About 56 million voters in Tamil Nadu will head to polls on April 23 to elect the state’s 16th Legislative Assembly in 2026. The contest is shaped by five major political formations. There is the ruling DMK-led Secure Progressive Alliance (SPA) under Chief Minister M.K. Stalin, the AIADMK-led NDA, and three significant third-front challengers that offer alternative ideologies to the governing Dravidian model.

This includes actor-politician Vijay’s Thamizhaga Vettri Kazhagam (TVK), the Naam Thamizhar Katchi (NTK) led by Seeman, and the PMK-Ramadoss alliance that includes VK Sasikala’s newly-formed party All India Puratchi Thalaivar Makkal Munnetra Kazhagam (AIPTMMK).

The election arrives at an inflection point. A decade of Dravidian welfare statecraft – free rice, higher education for girls, free bus for women, pensions for the vulnerable, insurance schemes, breakfast programmes in government schools – has defined the political baseline. Yet the state grapples with questions of fiscal sustainability; high unemployment especially among a young, educated workforce; and the unfinished agenda of caste and gender equity.

Perhaps the most defining factor is the increasing fragmentation of women voters who make up 51% of the electorate, who in the absence of strong female leadership, are a divided voting bloc this election season.

The elections serve as a critical test of the Dravidian movement that has governed the state for the better part of six decades, wrote journalist Narayan Lakshman in a paper. Without charismatic leaders like Jayalalithaa and Karunanidhi, going forward, the movement’s historic values as resistance to Brahmin dominance of state and public and private institutions may not be enough to secure power. Future state governments would be judged on their ability to deliver, or at least promise, “an acceptable standard of good governance”.

Aishwarya Govindarajan brings an issue-by-issue analysis of what these five formations are promising Tamil Nadu’s electorate.

Share

Issue 1: Welfare Policies

Tamil Nadu’s welfare architecture is among the most expansive in the country. Universal PDS, social security pensions for the most vulnerable, welfare schemes and targeted cash transfers have formed the bedrock of governance since the 1960s.

In 2026, the challenge is not whether parties can deliver welfare but how it can be expanded without deepening fiscal distress, and determining whose welfare philosophy resonates most with voters.

Issue 2: Women’s Welfare and Safety

Women’s financial empowerment remains a central electoral flashpoint in 2026. The DMK government’s Kalaignar Magalir Urimai Thittam (KMUT), a ₹1,000 monthly cash transfer to women, is a defining feature of the incumbent’s record. While the scheme has expanded women’s access to a predictable income stream, concerns persist that the amount is modest, often absorbed into household expenses.

While parties and state governments now promise cash transfers following this model, KMUT stands out for its near-universal reach among eligible women, and its framing as a ‘right’ rather than a conditional welfare benefit.

Party promises also range from increased SHG loans, coupons for household appliances and schemes to investments in safety infrastructure and labour rights.

Issue 3: Caste

Tamil Nadu is home to the original Justice Party, non-brahmin movements and the intellectual roots of Dravidian ideology, yet it faces renewed pressure due to caste-based violence. The state records over 1,000 cases annually under the Scheduled Castes and Scheduled Tribes (Prevention of Atrocities) Act, with persistent incidents of violence, discrimination, and social exclusion against SC/ST communities.

Against this backdrop, party manifestos offer assurances on reservation and scholarships, but diverge on how far the state should go in addressing entrenched caste inequalities and ensuring accountability.

Issue 4: Child Rights

NFHS-5 shows that while Tamil Nadu has reduced stunting to around 25%, acute malnutrition (14.6% wasting) and high child anemia (more than 57%) persist, with recent screenings indicating that 14–15% of children remain malnourished. Pandemic disruptions have also increased risks of child labour, while gaps in implementing the Protection of Children from Sexual Offences Act (POCSO) and rising mental health concerns around domestic and school violence have sharpened attention on child protection systems.

Significantly, for the first time, the ruling government’s manifesto allocates a dedicated section on child protection, marking a shift towards recognising children as rights-bearing citizens, rather than merely as a population without electoral voice.

Issue 5: Healthcare

Tamil Nadu has long been regarded as a model for public health in India, with relatively low maternal and infant mortality rates (19 per 1,000 live births and nearly 100% institutional deliveries). However, rising non-communicable diseases, healthcare inflation in the private sector, and the uneven distribution of specialist care remain acute challenges.

The manifestos reveal starkly different philosophies that range from universalising public coverage to leveraging private-sector tie-ups. No party has explicitly addressed the growing public health burden due to rising temperatures, water contamination, and increased vector-borne due to climate change.

Issue 6: Education

Youth literacy exceeds 96%, and gender gaps in schooling have largely narrowed, signalling the success of sustained welfare interventions in education. The state also leads in higher education participation, with a Gross Enrolment Ratio of over 50%, far above the national average.

Yet, this expansion has given way to new tensions about access. Exams like NEET are at the centre of political contestation, with the current government waging a legal fight against it. Opposition to NEET stems from concerns that a centralised exam disadvantages students studying in State Board, rural, Tamil-medium, and government schools who incur economic and psychological costs to earn a seat. Reports indicate a troubling pattern of NEET-related student suicides, with estimates citing over 20–25 deaths since 2017.

Education in this election goes beyond access – it is being fought on the grounds of equity, language, and the balance of power between state-led social justice models and centralised standardisation.

Issue 7: Elderly welfare

At least 60% of the state’s population is above 60 years of age and the old age dependency ratio has risen in the last 10 years, indicating rising health costs and pension burdens. a. A significant proportion of the elderly rely on state pensions as their primary or only source of income, with nearly 35–45 lakh beneficiaries covered under social security schemes.

However the current pension of ₹1,200 per month has failed to keep pace with inflation, especially in the face of rising healthcare and living costs. At the same time, the aging population carries a growing burden of chronic illnesses such as diabetes and hypertension (75% of elderly have at least one chronic illness), often without adequate access to geriatric care.

Parties offer competing approaches to elder care – focusing on pension increases, mobile clinics, pilgrimage support, and old-age homes.

Issue 8: Persons with Disabilities

The Rights of Persons with Disabilities (RPWD) Act 2016 mandated 4% reservation in government employment, but implementation has been uneven across Tamil Nadu. Parties speak of assistive devices, pension enhancements, and school support; NTK even plans to create a separate ministry. However, disability-inclusive urban planning and monitoring employment practices receive little attention.

Issue 9: Informal workers and labour rights

The state has a large and expanding informal and mobile workforce. Estimates suggest 7.7 million gig and platform workers across India, with Tamil Nadu accounting for a significant share in urban centres like Chennai and Coimbatore. The state also sees substantial overseas migration, with over 1.5–2 million Tamil workers in Gulf and Southeast Asian countries, alongside a sizable but undercounted internal migrant workforce in construction, textiles, and services.

This is especially visible in district like Tiruppur, the state’s knitwear export hub, where 3–5 lakh inter-state migrant workers sustain the industry but face precarious housing, limited access to healthcare, occupational safety risks, child labour concerns, and weak social security coverage.

Despite their central economic role, migrant workers, especially women, remain largely absent from manifesto commitments, with little clarity on portability of entitlements, housing standards, or labour protections.

While welfare boards exist and discussions on gig worker legislation and diaspora support are emerging, coverage remains fragmented and implementation uneven.

Issue 10: Fisher community welfare

Tamil Nadu’s 1,076-km coastline sustains one of the country’s largest fishing communities. The annual fishing ban, the longstanding Sri Lankan navy incursions, and competition from mechanised trawlers remain sore political wounds. 2026 sees unusually targeted promises that address longstanding concerns, such as accident insurance and sea ambulances, reserved assembly constituencies for fisher communities, and biometric cards for fisherwomen whose work goes unrecognised.

Issue 11: Housing access

Tamil Nadu’s housing deficit is sharply visible in Chennai’s resettlement colonies like Semmenchery, Perumbakkam, Ezhil nagar, Kannagi Nagar. More than 1-1.5 lakh people are displaced from inner-city slums, and face long commutes (20–40 km) and limited access to jobs, schools, and healthcare.

Schemes like PM Awas Yojana have expanded formal housing but have largely relied on peripheral relocation rather than in-situ upgrading. The manifestos diverge on this question – between scaling housing as infrastructure and addressing it as a rights-based issue tied to land, livelihoods, and access to the city.

Issue 12: Climate and environment

Tamil Nadu is one of India’s most climate-vulnerable states owing to the regular exposure to cyclones from the Bay of Bengal. Catastrophic crises like Cyclone Vardah and Cyclone Gaja have caused widespread damage to housing, agriculture, and infrastructure. Coastal districts face the threat of erosion.

At the same time, rapid urbanisation has intensified flooding risks: the 2015 Chennai floods and the 2023 floods exposed how wetland loss and unregulated construction amplify extreme rainfall impacts. In addition, the related water-stress, which includes a high ground water dependency of up to 80%, and the 2019 Chennai water crisis, have highlighted the fragility of urban water systems.

Despite urgent lived realities, ecological concerns remain an afterthought in electoral imagination There are plenty of contradictions: parties promise afforestation, river restoration, and wetland protection while also promising large-scale infrastructure, industrial corridors, and urban expansion projects. The result is a policy tension where climate resilience is acknowledged, but not fully integrated as a guiding framework for development.

📌 Have you joined Feminist Election Newsroom yet? FEN is BehanBox’s initiative to build a collective feminist practice of democratic engagement—tracking, interrogating, and holding power accountable, during elections and beyond.

Join the Feminist Election Newsroom

An earlier version of this story referred to the ‘DMK-led Secular Progressive Alliance (SPA)’ as ‘DMK-led UPA’. The error is regretted.

Elections in West Bengal are scheduled in two phases on April 23 and 29. The state had over 7.6 crore eligible voters before the Special Intensive Revision (SIR) exercise. About 27 lakh voters were removed from the voter list after judicial adjudication, and as of April 20, 6.82 crore people are eligible to elect 294 members of the 18th state legislative assembly. The votes will be counted on 4 May.

A triangular contest is currently underway between the ruling All India Trinamool Congress (AITC) party and opposition parties – Bharatiya Janata Party (BJP) and Left Front-led Communist Party of India (Marxist) (CPI(M)). Key issues being debated among parties include women’s welfare, SIR and citizenship, and employment issues.

The AITC, led by incumbent Chief Minister Mamata Banerjee, is seeking a fourth consecutive term. The party is fighting on the grounds of welfare, identity mobilisation and an aggressive anti-BJP stance.

The BJP, meanwhile, is banking on the anti-incumbent sentiment and has positioned itself as an alternative to AITC’s 15-year-long governance marked by corruption, misgovernance, and undermining youth development. Moreover, the party argues that a “double-engine government“ model will lead to faster development, administrative efficiency and job creation. Simultaneously, the party is consolidating Hindu votes by invoking contentious issues such as citizenship and threats posed by “infiltrators” and “outsiders” while promoting resentment against religious others.

The CPI(M), having won seven consecutive assembly elections between 1977 and 2011, is seeking to reclaim political relevance. Having been sidelined after AITC first assumed office, the party was instrumental in implementing land reforms in the 1980s, establishing a decentralised model of local self-government and defending the rights of minorities. Its issue-centric campaign focused on employment, welfare and ethical governance in opposition to what it perceives as communal and identity-based politics.

With West Bengal’s emergence as India’s most closely watched political battleground amid elections in Assam, Kerala, Tamil Nadu, and Puducherry, the contest reflects a state at a crossroads where competing visions of governance, development and identity are jostling for primacy.

Share

Issue 1: Special Intensive Revision (SIR) exercise

The Special Intensive Revision (SIR), a voter verification and electoral role updation process by the Election Commission of India (ECI), is the most contentious issue that has come to dominate election discourse in West Bengal.

The Union government says this exercise is necessary to create an “error free” voter list, ensure the removal of ineligible voters and to introduce “complete transparency“ in upgrading electoral lists.

Increasing urbanisation, periodic migration, young citizens becoming eligible to vote, non-reporting of deaths and the inclusion of the names of “foreign illegal immigrants” are some reasons it cites for introducing the process.

However, this has triggered massive condemnation and apprehension by opposition political parties, civil society groups and ordinary citizens. According to preliminary reports, of the 90 lakh voters excluded by SIR in the state, 57.47 lakh or 63% of the deleted voters are Hindus, while 31.1 lakh or 34% are Muslims. This is significant since Muslims, according to the 2011 Census, accounted for 27% of the state’s total population. Murshidabad district, having the highest number of Muslims in the state, recorded the highest number of deletions (over 4.55 lakhs), followed by Malda with 2.39 lakh omissions. Women bear the brunt of this exclusion – over 61.93 lakh women have been dropped from voter rolls, with women from Muslim, SC, and ST communities most impacted, according to an analysis by the Sabar Institute.

Meanwhile, the ECI has constituted 19 tribunals to hear appeals against voter deletions. Concerns have been raised particularly over delays in the disposal of appeals, uncertainties over jurisdiction, an absence of standardised procedures, and a lack of infrastructure and administrative support.

The AITC alleges that SIR is to enable the BJP to “grab power“ in the state. The CPI(M), in a letter to the ECI, has argued that SIR “represents a systematic exercise in mass disenfranchisement, marked by arbitrary criteria…and the increasing reliance on algorithm-driven exclusions rather than transparent, field-based verification”.

The Supreme Court, while stressing the need for a “robust appellate mechanism“ to address wrongful deletions, has nonetheless refused to intervene in individual cases. It has also declined to extend the 9 April deadline for freezing electoral rolls. Individuals cleared by appellate tribunals will be permitted to vote, provided their appeals are decided before April 21 and 27 for the first and second phases respectively.

Brown University Professor Ashutosh Varshney noted that the SIR contributes to a larger trend of “democratic backsliding”. “The erosion of democracy begins with an attack on the non-electoral dimensions of democracy—freedom of expression, vibrancy of non-governmental civil society and minority rights—and the next target is typically the electoral process.”

“[I]f the size of the Muslim electorate goes down sufficiently in Bengal and the BJP wins the state, it will not only be a matter of practical politics. Bengal’s political outcome will also enter the larger democratic discourse worldwide. Contrariwise, if the BJP loses in Bengal, it will also become clear that political parties can fight back by registering those communities that are often targeted by such exclusionary projects,” he added.

Issue 2: Women’s welfare

The Global Gender Gap Index 2025 places India at 131 out of 148 countries, with an overall gender parity score of 64.4%.

Against this backdrop, West Bengal’s gender dynamics demonstrate several structural shortcomings. The female literacy rate is 70.54% compared to the male literacy rate of 81.69%, reflecting a gap of 11.15 percentage points. Women remain disadvantaged economically, with the female labour force participation rate at 33.8%, well below the national average of 40%.

Women’s low economic participation and persistent social inequalities such as early marriage, domestic violence and gender discrimination, further exacerbates gender inequality in the state, according to NITI Aayog’s localised model of the UN’s Sustainable Development Goals 2023-24.

The ruling TMC has, in recent years, relied heavily on welfare programs and an institutional empowerment model to address gender inequalities with uneven outcomes. Key to this undertaking has been the Kanyashree Prakalpa program, which seeks to improve the well-being of girls through conditional money transfers. Studies demonstrate that the scheme has succeeded in reducing dropout rates and increasing female enrollment in higher education. Women’s self-help groups, on the other hand, have significantly enhanced decision-making power and community-level participation, with collective action helping women access resources and influence local governance.

Other programs such as the Lakshmir Bhandar Prakalpa for Scheduled Caste (SC), Scheduled Tribe (ST) and General Caste category women and Bidhoba Bhata for unmarried widows have also been instrumental in fighting poverty, building personal capacity and promoting self-confidence.

Issue 3: Informal workers and social welfare

The ruling AITC has implemented broad-based welfare coverage across vulnerable groups, including minorities, rural poor and elderly citizens, through cash transfers and subsidies. Such interventions have been effective in improving access to food, education and housing, especially in rural areas. However, concerns remain over distributional inefficiencies and access-related inequalities. Evidence suggests that access to welfare benefits can depend on social networks and local political connections, leading to uneven targeting across communities.

Moreover, the state continues to face underlying challenges of malnutrition, a high proportion of child marriages and a lack of regular employment opportunities. Parties this year have extended schemes to ASHA workers and gig workers, but no one has explicitly spoken about formalisation of these workforces.

Measurable gains in social protection and inclusion have also raised persistent questions about whether a populist-driven approach can sustain long-term improvements in human development and economic capacities.

Issue 4: Education

School education in West Bengal underwent major changes during the Left Front’s first term in office in 1977. Several measures, such as the removal of tuition fee in high schools, increasing teachers’ salaries, the construction of concrete school buildings and higher school enrollments due to the implementation of land reform policies, led to wide-scale transformation.

Per the 2011 Census, West Bengal has a 70.5% literacy rate, while India’s national literacy rate is 72.98%.

Despite past progress, the state’s education system is under strain. School closures, worsening student-to-teacher ratio, declining student enrollment since 2013 and corruption in the appointment of teachers have adversely impacted the system.

Furthermore, the closure of hostels, reduced budgetary allocation for education, and vacant teaching and non-teaching positions at madrassas have compounded the problem.

According to the state budget for the financial year 2025-26, total expenditure (excluding debt repayment) is estimated to be Rs 3.41 trillion, an increase of 14% from the revised estimates of 2024-25. The state has allocated 14.8% or Rs 505.59 billion of its expenditure on education.

Issue 5: Healthcare

Due to a severe shortage of public health personnel and management skills in primary healthcare for rural populations, along with a planning‐to‐implementation gap, the state’s healthcare system is underfunded. The primary healthcare system’s inability to meet the enormous demand in rural areas impacts poor people disproportionately. They suffer from impoverished health and a vicious cycle of poverty more than their urban counterparts.

The quality of rural healthcare is further hampered by inadequate services, high rates of absenteeism among available health professionals, and a lack of pharmaceuticals and medical equipment. Moreover, residents experience a wide disparity in the availability of critical healthcare services among various demographic groups, such as people experiencing homelessness, women, children and older people.

Per the 2025-26 state budget, 6.4% or Rs 219.39 billion out of Rs 3.41 trillion has been allocated for health and family welfare.

Issue 6: Agriculture

West Bengal is a major producer of paddy, wheat, oilseeds, pulses, potato and jute.

Per Census 2011, the state covered an area of 88,752 sq km or about 2.7% of India’s total geographic area of 3.28 million sq km. The total number of cultivators was 5.1 million people, constituting 14.72% of the total workforce of the state. Among them, 4.2 million people (82.16%) were main cultivators, while 0.91 million people (17.84%) were marginal cultivators.

West Bengal, per the inaugural Agricultural Statistics at a Glance 2021 report, had 5.1 million main and marginal cultivators, and 10.18 million main and marginal agricultural labourers. Of these, 0.61 million were main and marginal female cultivators, while 2.73 million were main and marginal female labourers.

However, increasing agricultural distress in recent years has been driven by land degradation, temperature extremes, water shortage, precipitation-related anomalies and insufficient institutional support. Coastal regions in the state are vulnerable, particularly due to cyclones, erratic rainfall, rising sea levels and saline water intrusion.

The 2025-26 state budget allocated 5% or Rs 232.83 billion out of Rs 3.41 trillion for the agricultural sector. This allocation places it in the middle tier among India’s major agricultural states, higher than Andhra Pradesh (Rs 14,138 crores), Gujarat (Rs 14,144 crores) and Punjab (Rs 14,407 crores) but below larger spenders such as Uttar Pradesh (Rs 24,215 crores) and Maharashtra (Rs 40,748 crores).

Issue 7: Economy, industry and infrastructure

The Bengal Presidency under British rule saw the rapid emergence of various industries such as tea, jute, engineering, coal, and iron and steel. However, this trajectory was disrupted following the partition of Bengal in 1905 and India’s independence in 1947.

The handicrafts industry also suffered with the ascendancy of the East India Company. British India lost a significant share of the world textile markets to its coloniser between the late 18th and early 19th centuries. Increased productivity due to the adoption of the factory system led to a global decline in the price of textiles.

Thereafter, there was a steady fall in the state’s share of national industrial output. This is partly due to increased trade union activity between 1967 and 1970, and partly due to industrial actions, political instability, lack of local capital and poor infrastructure. . Around the same time, agriculture witnessed some revival. The secondary and tertiary sectors also grew at 4.03 per cent and 4.08 per cent, respectively.

While the state entered a high growth phase post-2000, driven largely by the tertiary sector, industrial growth remains uneven and limited in employment generation. Micro, small and medium enterprises (MSMEs) have since emerged as the backbone of industrial activity. However, challenges such as outdated technology, infrastructural deficits and limited access to credit remain. Nonetheless, the industrial sector in West Bengal recorded a 7.3% growth in 2024-25, greater than the national average of 6.2%.

Issue 8: Environment

West Bengal is vulnerable to climate stressors, given its location between the Bay of Bengal and the eastern Himalayan range. This vulnerability is compounded by factors such as deforestation, rapid urbanisation, natural hazards and unplanned developmental activities. But environmental and ecological concerns find little space in the election manifestos despite acknowledgement by politicians to this effect.

Coastal flooding is prevalent in the state, given the annual occurrence of river floods, tidal floods, fluvial-tidal floods and storm floods. A rise in sea surface temperature over the last two decades has made the Bay of Bengal basin a breeding ground for cyclonic storms that have increased fivefold in number between 1970 and 2020. Such extreme events have been followed by embankment breaching and saltwater intrusions. Coastal and riverine households are acutely vulnerable, leading to the loss of land, livelihoods and other natural resources. Consequently, they are at risk of food insecurity and reduced ability to alleviate poverty.

Despite advancements in disaster preparedness and prevention targeting coastal populations and regions in West Bengal, there is a lack of critical infrastructure, government capacity and appropriate policy implementation.

Case in point is the Sundarbans, the world’s largest deltaic ecosystem bordered by West Bengal and Bangladesh. It is home to roughly 7.2 million people (4.5 million in India and 2.7 million in Bangladesh), including some of South Asia’s poorest and most vulnerable communities. It is also home to the Bengal tiger, estuarine crocodiles, olive ridley turtles, northern river terrapin, Gangetic and Irrawaddy dolphins, and various amphibians, invertebrates, birds and fish species.

However, pronounced drought, increased agricultural activities, high levels of precipitation, rising sea levels and temperatures, and the loss of intertidal mudflats threaten human existence and the habitats of many species. Meanwhile, adaptation strategies such as the cultivation of salt-tolerant crops and the construction of storm-surge walls and embankments have met with limited success.

Issue 9: Employment

West Bengal and Bihar have the highest prevalence of girl child marriages in the country, impacting 42.5% women married before the age of 18 years, according to the National Family Health Survey (NFHS) 2019-21.

West Bengal’s rate of unemployment was 3.6% for people aged 15 years and above, below the national average of 4.8%, according to the National Period Labour Survey (NPLS) for October-December 2025. However, persistent unemployment among educated youth reflects a mismatch between the expansion of higher education and the creation of stable and high-quality jobs.

Issue 10: Tea industry

West Bengal, India’s second-largest tea-producing state, manufactured over 166 million kilograms of the beverage in 2023-24.

However, despite its export significance and flavour profile, the tea industry faces several challenges such as low worker wages, rising production costs and stagnating tea prices. Moreover, the tea industry in Darjeeling faces a precarious future due to temperature and rainfall-related fluctuations, decreasing area under cultivation and high levels of carbon dioxide.

📌 Have you joined Feminist Election Newsroom yet? FEN is BehanBox’s initiative to build a collective feminist practice of democratic engagement—tracking, interrogating, and holding power accountable, during elections and beyond.

Join the Feminist Election Newsroom

In the thick of the election season, a legislative thriller writes itself. The Union Government on April 14 tabled three Bills purportedly to increase women’s participation and representation in the Lok Sabha and the assemblies of States and Union Territories. These are: the Constitution (One Hundred and Thirty-First Amendment) Bill, 2026, the Delimitation Bill, 2026, and the Union Territories Laws (Amendment) Bill, 2026. The women’s reservation bill, which was passed in 2023, came into effect from April 16 as discussions continue about its merger with a delimitation bill — a move the Opposition called “bizarre”.

On April 17, the Lok Sabha rejected the Constitutional Amendment Bill. The Union Government has subsequently withdrawn the Delimitation Bill, 2026, and the Union Territories Laws (Amendment) Bill.

The Bills proposed a 50% increase in the Lok Sabha strength, and to amend the Constitution in a way that would give the Union Government powers to decide when and using what data are future delimitation exercises carried out.

Without any proposed safeguards, the Bill gave arbitrary discretionary powers to the government, which could be misused to manipulate future electoral outcomes, says Maansi Verma, a lawyer and founder of Maadhyam, a civic engagement platform that tracks the parliament and policies.

Under scrutiny was the timing, the premise underlying these legislations, and the hasty process with which they were introduced. Why link delimitation, a constitutionally-mandated exercise, with women’s reservation bill? Why do it before the Census 2027 data is published? Why 850 seats?

Share

According to Maansi, the three bills together were a smoke screen, which violated the spirit of the women’s reservation bill. “What they actually wanted to do is not upset the existing arrangement; they did not want any male MP to make space for women,” she says. Opposition MPs have further noted the Bill will negatively impact women from SC and ST communities.

The bills gave discretionary power to the government, which without judicial oversight, changes the nature of our electoral democracy. “This is not how bills should be made, how the constitution should be amended, not how parliament should function,” says Maansi. Edited excerpts below.

What do you make of the timing, intent, and ongoing debates behind the delimitation push?

There was absolutely no inkling that the government was planning something like this till about a month ago. The Transger Persons (Protection of Rights) Amendment Act, 2026, was brought on March 13, then it completely took over the news cycle and at that point of time, when the budget session was going on, the government did not indicate that it was planning something like this.

That is strange, because on April 16, while speaking in the Parliament, the Prime Minister said that they have been doing deliberations and consultations, not just with different political parties, but also with constitutional experts and women activists, that there have been structured and informal conversations. But who are these experts? Who was the government speaking with? Where did these consultations happen?

Apparently, the government called two all-party meetings which happened during the budget session, but even then, it wasn’t very clear whether the government gave any justification for why this needs to happen now.

On the last day of the budget session — which was originally scheduled to end on April 2 — it was announced that it would reconvene during April 16-18. Elections were going to take place in four states and one union territory… why choose these three days? The Opposition also asked: if you want to implement the Women’s Reservation Bill quickly, why could you not have waited till the elections got over? The government has so far not come forward with any justification. All it is saying is: women have waited too long, this needs to happen. But then why did they wait from 2023 till 2026?

It remains a mystery why it had to happen now. Some people link it to elections – that the government wants to get political mileage in the middle of election season. And they are building a narrative that they are doing this for women. So if the opposition opposes delimitation, the government will spin that to say the opposition is against women’s reservation. And that is what we see. The Prime Minister very categorically said in his speech on April 16 that everyone can take the credit if they support the Bill, but if they oppose it, he is definitely going to get political benefit out of it.

Why is delimitation linked to the women’s reservation bill, what is the government’s argument? Is there a precedent for tethering a constitutional process to specific legislation?

In 2023, when the government brought the Women’s Reservation Bill, it said that whenever this act comes into force — and this act came into force on April 16, 2026 — the census which happens after it, delimitation will happen on that basis. And then whatever is the number of seats in the Lok Sabha finally, one-third of that will have to be reserved for women.

A lot of questions were raised even then. Instead of implementing the one-third reservation on the existing 550 seats of Lok Sabha, everyone asked: why postpone it, deferring it to a date when a new census happens, then a delimitation happens, and then it is implemented? The concern was the next census would have been the 2031 census, and once the census happens, then delimitation happens, so before 2034 at least, the reservation could not have been implemented if you go by the formulation mentioned in the Act. A lot of people also criticised the government for this formulation, noting that they don’t need to link women’s reservation to census and delimitation — just apply this one-third formula on the existing 550 seats and implement it from 2024 onwards.

The government offered no real justification at that time also, and even then the government did not do any public consultation on the bill. A special session of Parliament was suddenly called and the government kept saying that some historical decisions would be taken. The bills were introduced without any knowledge of the contents or indeed any consultations.

The most plausible theory of why the government insisted on census and delimitation first, is it would lead to a reallocation of seats, and eventually, seats in Lok Sabha would have to be increased. Then giving one-third of that to women would be more acceptable to the men who are currently holding on to these seats — men who somehow consider it their birthright to rule, so they would not be upset when their seats suddenly become reserved for women. So, to not upset the existing arrangement — where we had only 14% women in Parliament and the rest were men who were not willing to make space for women — this formula was proposed.

Tejasvi Surya on April 16, when speaking on the new bill, more or less confirmed this to be the government’s thinking. He categorically said that if they would have reserved seats on the basis of the existing Lok Sabha strength, a lot of male MPs would end up losing their seats. What the government wanted to do was to prevent the existing seats from “encroachment” — that was the word he used for women’s reservation.

The idea is that the existing makeup of Lok Sabha should be disturbed as little as possible, and additional seats be created which can then be reserved for women.

If the government had implemented the Women’s Reservation Bill in 2023 or if it were to do it right now, then conduct the delimitation post-Census, how would that play out?

If this would have been implemented in 2023 itself — without linking to it to census and delimitation, and simply said one-third of existing seats have to be reserved — then in 2024 itself, out of 550 existing seats, one-third would have been reserved for women.

What complicates things now is that there was another provision in the Constitution [the proviso to article 81(3) and the third proviso of Article 170] which had frozen the reallocation of seats till 2026. The Constitution currently says that the Lok Sabha seats have to be reallocated and readjusted after every census. But this was frozen earlier — in 1976 and then again in 2001 — to say that this arrangement should not be disturbed till a census after 2026 is conducted.

Share BehanBox’s Substack

This was done because some states would have ended up losing out on their seats in Lok Sabha because they had implemented population control well. What is written in the Constitution is that as far as practicable, the population-to-seat ratio should be the same. That would mean that if, for instance, in southern states there are 15 lakh people in every constituency, and if you use that ratio to apply to Uttar Pradesh, the number of seats would have to increase. The population disparity was such that if this principle was followed, the seats for certain northern states would disproportionately increase in comparison to seats for southern states. Therefore, a political arrangement was reached between the union government and state governments to freeze it till the time a more just division of seats could be achieved.

In 2026, the government again has an opportunity to freeze it for a certain number of years. But some people — and this argument also has weight — have been saying that since 1971, population has drastically increased, so there is a need to reconsider the seating arrangement in Lok Sabha. Right now, some constituencies are unduly large and no MP is able to do justice to representing lakhs and lakhs of people.

The assurances the Prime Minister and the Home Minister gave in Parliament — that whatever is the existing proportion of seats each state has will be maintained and on top of that every state will gain 50% — are not written anywhere in the bills.

In the parliament, the government has said the proportion of seats will remain the same for states like Karnataka and assured southern state’s seats would not be decreased. Is the arithmetic correct? How can they claim that without any data or recent census figures?

The government is doing a delimitation right now, before the 2026-27 census figures are available, and is proposing to use the 2011 census for this purpose. According to their calculation — and I don’t know how they have done these calculations, because psephologists like Yogendra Yadav have done their own calculations and have given different numbers — the government is saying that no state will lose, because they are also increasing the total number of seats, and every state will eventually gain seats.

But this is not laid down in the law — it is just what the government is saying, assurances they have given. Yesterday, the Home Minister in his speech gave some numbers but this is their calculation on the basis of the 2011 census.

What the bills themselves actually say is very different: they say that the government is empowered to conduct delimitation whenever it wants to, and can use whichever census data it wants for that purpose.

What is also very strange is that the government is already coming up with these formulations and deciding the reallocation of these seats, which is actually the job of the Delimitation Commission to determine on the basis of the principles laid down. The Delimitation Commission is supposed to function independently, place its draft report in the public domain, invite comments, interact with all political parties, and go on the ground. A very technical but also very consultative process is supposed to be followed, after which you arrive at these numbers.

The government can tomorrow take back these promises. They can say that we gave these suggestions to the Delimitation Commission, but it is an independent body and ultimately has to come up with its own numbers, and we can’t do anything about it. The government would be legally correct in saying that.

Even if the government brings a schedule to the delimitation bill laying down the maximum number of seats for each state, to what extent that is going to be binding on the Delimitation Commission is an open question, because the body is supposed to function independently.

It is very difficult to understand why the government is insisting that all these assurances should be believed, and that just on that basis we should take all of this — the 850 number, the delinking of delimitation from census, the use of the 2011 census — on face value. Clearly, the government has not made any attempt to justify its decisions or to do open public consultations on this.

Do the three bills address reservations for SC, ST, and OBC women within the broader framework? What would that arithmetic for SC/ST reservation look like?

The Constitution Amendment Act of 2023 did say that within the one-third of seats reserved for women, seats have to be reserved for women from Scheduled Caste and Scheduled Tribe communities. But the Scheduled Caste and Scheduled Tribe reservation happens on the basis of population — the idea is that those constituencies where people from Scheduled Caste and Scheduled Tribe communities are in greater proportion should be reserved for them.

So if you use an outdated census of 2011, it is quite possible that it doesn’t reflect the true reality on the ground of where people from Scheduled Caste and Scheduled Tribe communities are more concentrated and should be represented. Constitutionally, the representation of Scheduled Caste and Scheduled Tribe women within the larger quota of women is protected. But how it will actually play out on the basis of the 2011 census remains an open question.

The government is also planning to do a caste census in the upcoming 2026-27 census. It is quite possible that the political demand for OBC reservation would become stronger after that data came out — the opposition is already raising that demand. To avoid that situation, what the government is saying is: we will do a delimitation right away, using the 2011 census, before the caste census data comes out.

[MP Brinda Karat wrote in The Hindu: “In the name of urgency for women’s reservation, implementing a delimitation exercise which ignores the current population of SCs and STs will be a manuvadi injustice to oppressed communities, denying them a fair increase in the proportion of reserved seats.”]

Why are we proposing a 50% increase in seats? What’s the arithmetic? Does it align with past increases?

No, it has never been a 50% increase in one go and the government has not given any justification for this. This is a very big jump from 550 to 850. If you look at past increases, they have always been very modest and gradual. For instance, 1971 is the last census we used to increase seats. The Constitution was amended in 1973, and if you look at the Statement of Objects and Reasons of that constitution amendment, you see that first the 1971 census happened, and the delimitation process was triggered immediately after. It was realised that if the existing 500 seats remained, some states would lose out and some would gain. So to make sure no state loses out and everybody gains, an increase in the number of seats was proposed.

Also note that this increase in seats was proposed after they already had the census data from 1971 and the delimitation process was underway — so the government had some idea of what the reallocation might look like. And even then, it was a very modest increase from 500 to 525. Then in 1987, some reorganisation of states happened and this number went up from 525 to about 530 for states. Every time, these increases have been quite modest — you want to make sure nobody loses out, that even if it is a gain of one to two seats every state is gaining — but it’s not like the gain is 50% in one go.

Even if we lived in an ideal world and every census since 1971 had seen a modest increase — 25 to 30 seats every time, in 1981, 1991, 2001, 2011, 2021 — even then, it would have amounted to a total increase of maybe 100 to 125 seats at most, which means from 550 you would have gone to 650 or 675. Just because that didn’t happen and there is this pent-up need to increase seats, you can’t suddenly jump from 550 to 850. There have been MPs and experts who said that a 10% increase would have been sufficient to ensure no state loses.

Also consider how women’s reservation fits with this. If you do one-third of 850 it is roughly 267. And if you subtract that from 850, you arrive at the current number of seats in Lok Sabha. The number of seats being added is exactly the number of seats that need to be reserved for women. What they wanted to do is not upset the existing arrangement, they did not want any male MP to make space for women.

The Constitution requires the delimitation exercise to happen after every census, but the amendments want to remove this clause and give the government arbitrary power to decide when to do this exercise. What are the potential risks with this change?

One of the reasons why this is laid down in the Constitution is to give a certain kind of certainty and predictability. You know for sure that every time a census happens, it will be followed by delimitation. Every political party can prepare itself accordingly. It is important to have a set frequency laid down in the Constitution as it enables a process which can be followed in a rhythmic pattern, and is not left to the arbitrary whims and discretion of whichever party is in power at a particular point of time.

Now, this is what the government is proposing to change. The government is saying that now they can decide when delimitation has to happen. If the 2026 freeze had been left alone per the Constitution, whenever the 2026-27 census data gets published — say, by next year — delimitation would have had to happen automatically. The government would have been forced to use the 2026-27 data only for the purpose of delimitation.

Since the government is also planning to do a caste census in this data collection, it is quite possible that the political demand for OBC reservation would have become stronger. To avoid that situation, what the government is saying is: we will do a delimitation right away. And the government could not have done that without giving itself the discretionary power to decide when delimitation should happen. So the government is proposing to delink delimitation from the census so that it has the power to decide when delimitation should happen and which census it should use for that purpose.

And this can also mean — and legally there would be nothing stopping the government from doing this if this amendment is passed — that every five years, just before elections to Lok Sabha or before elections to a particular state, the government can order delimitation and also decide which census to use for that purpose. This can become a tool in the hands of the government to manipulate electoral outcomes. It may not happen in the Lok Sabha because of the kind of attention this may garner, but it could happen in the states.

Subscribe now

We have already seen something like this in the case of municipal elections. The 74th Constitution Amendment gave municipalities a constitutional status and provided for elections to municipal councils every five years. But just like a frequency for delimitation was laid down for Lok Sabha and Assembly elections, that was missing from the 74th Constitution Amendment. State governments were given absolute discretion to decide when they want to do delimitation, which census to use. And we have seen instances of extreme misuse of this discretion, that every state government — irrespective of party — has done it. If they are not expecting favourable results in the municipal election, they would suddenly announce delimitation or a change in reservation status. This has happened every five years, and sometimes more than once within that five year period, they would arbitrarily announce delimitation or they would delay delimitation, they would announce just before the elections are due to delay the elections. We have seen this in Bengaluru, in Mumbai, in other cities as well.

There will be no judicial oversight of the delimitation committee. Should this be a point of concern?

That has always been there in the Constitution. There is a Supreme Court judgment from the 1970s which has upheld it — that delimitation is not open to judicial review, and so delimitation exercises from the very beginning have been shielded from judicial review.

But now, with so much discretion being given to the government — a government that is not bound to give any justification for why it is ordering delimitation in a particular state at a particular point of time, or why it is using a particular census data for that purpose — to also not have judicial review on top of that becomes problematic. Maybe this might also mean that this question itself — of whether delimitation decisions should be open to judicial review or not — may be up for reconsideration, because the basis has changed.

What have we learned from the delimitation exercises in Assam and Kashmir?

What happened in Assam and Kashmir is a lesson in why this kind of discretionary power should not be given. In Jammu and Kashmir, the commission set up for delimitation purposes ended up delimiting constituencies in such a way that the Jammu division gained more seats than the Kashmir division. BJP was more confident of its performance in Jammu as compared to Kashmir, so it would have definitely gained advantage out of it.

In Assam, geographical consistency — which is one of the principles that has to be followed — was violated: constituencies have been drawn across rivers and across mountains, in ways that geographically would not make any sense on a map. Essentially, wherever Muslim populations were concentrated, those areas were divided into smaller constituencies so that their concentration came down and the number of seats that Muslims could influence was reduced. The BJP, wary of not getting support from Muslims, essentially reduced the influence that Muslim votes can have on the eventual outcome of elections.

Experts have also noted that even all of these new 850 seats can be redrawn in such a way that the population makeup of particular constituencies is engineered to benefit a certain party.

The ruling government is arguing that doubling the MP count and having longer Parliament sessions will lead to better representation. What do you make of that argument? How does an altered parliament composition affect legislative processes in the future?

With the way Parliament is functioning currently, we are seeing very, very few sittings in a year. We are seeing MPs’ time — which is supposed to be for Question Hour, Zero Hour, private member business — often lost because of disruptions, protests, and repeated adjournments by the government. And when that doesn’t happen, the government simply prioritises its own business over these other things. A case in point is the session happening right now: an ordinary session should have had Question Hour, Zero Hour, private member business. But the government suspended these three things to say that only these bills will be discussed and nothing else will happen in this session. And this happens very frequently. MPs hardly get any time to represent people, raise their concerns, ask questions.

Even when debates happen — and debates don’t happen that often in Parliament — very few MPs get to speak. Smaller parties, regional parties, get very little time. Usually it is BJP which gets a lion’s share of the time, and even there, you repeatedly see the same MPs making speeches. A lot of MPs may have had an opportunity to speak maybe once on a bill in a year. So if this continues, and you have more MPs in Parliament, essentially those MPs will just come to sit in Parliament, not get any real opportunity to speak much, vote occasionally, and then leave.

And most of these new people who are going to come to Parliament are going to be women MPs. You want more women to come to Parliament — but to do what? Just come and sit. Vote occasionally, then leave. That is all that you are looking at: women as just numbers.

Legislatively, we don’t see a lot of debates happening in Parliament. Even committees — as of now, committees have about 30 MPs, 20 from Lok Sabha and 10 from Rajya Sabha. Now, Rajya Sabha numbers are not going up, so Rajya Sabha MPs will continue to be 10, but Lok Sabha MPs will increase, so you will have 40-MP committees. How much time would MPs get to actually make a contribution even within committee meetings? And the government anyway is not sending bills to committee as much as it should. So we are seeing not just the functioning of Parliament coming down, but also the functioning of committees — and to these processes, which are becoming less and less alive, we are just trying to add numbers without creating the space where these additional MPs can make their presence felt.

A lot of people have also raised questions about the role of Rajya Sabha being further diminished. You’ll have an 850-MP Lok Sabha and just about one-fourth of that in Rajya Sabha. If a joint session of both houses needs to be called to pass a bill — and it hasn’t happened in the last two decades at least — but if it does, Rajya Sabha MPs would hardly have any influence. Even in committees, if you have 30 MPs from Lok Sabha and 10 from Rajya Sabha, even there their influence will go down. What role would Rajya Sabha then even play in the kind of deliberations that need to happen in Parliament?

Another valid point that a lot of people have raised is that essentially all these numbers are being generated just so that the government can ensure it has the numbers it needs to pass bills — even, in the future, constitutional amendment bills. If it has a larger number of MPs from the northern states, it will become easier to get bills through even without any debate. There are also concerns about cost — how much it is going to add to the cost of running Parliament, the salaries and all other perks and housing that has to be given to parliamentarians.

All of this is what one should remember from this entire episode: this is really not how laws should be made. There is something called a pre-legislative consultation policy of the government from 2014, which essentially says that every draft bill, before it is introduced in Parliament, should be placed in the public domain. The government should invite comments for a minimum of 30 days, do open consultations, give justification for why it is bringing a certain bill, and only after having done that process should a bill be introduced in Parliament. After introduction, ideally it should be sent to a committee for a further round of scrutiny, and only then should it be taken up for debate and passage.

All of this has been sidelined and fast-tracked. No consultation, no discussion, no deliberation, no committee scrutiny.

This is not how an ordinary law should be made, and definitely not how the Constitution should be amended — in such a rushed manner without any consultation. There are very real implications of what happens on the ground after you make these changes in the Constitution.

This story was first published minutes before the Lok Sabha rejected the Constitution (131^st Amendment) Bill, 2026. The story has been updated to reflect the change.

Leave a comment

Political alliances contesting the election include: the ruling Left Democratic Front (LDF) led by the Communist Party of India (Marxist) (CPI(M)), the United Democratic Front (UDF) led by Indian National Congress (INC) party, and the National Democratic Alliance (NDA) led by Bharatiya Janata Party (BJP).

The LDF, under incumbent Chief Minister Pinarayi Vijayan, is seeking a third consecutive term. The NDA, having won 0 seats in the 2021 assembly elections, is looking to disrupt the bipolar LDF-UDF contest by promising a corruption-free governance, farmers’ welfare, women’s empowerment, and employment opportunities for the youth.

Other issues dominating campaign season include healthcare, education, housing, connectivity and modernisation, and climate and environmental governance.

Here is what manifestos of the main coalitions are promising the electorate.

Issue 1: Welfare support

Policy development in Keralam has been largely defined by welfare measures initiated by successive state governments since the 1960s. This resulted in food security, improvement of healthcare and education, and the expansion of pension schemes.

In 2026, the government spent 154,675.53 million rupees on social security and welfare, up from 148,073.70 million rupees in 2025. However, concerns regarding increased public spending remain amid rising debt, inadequate employment opportunities and lack of incentives for attracting new talent. Despite a roughly eight-fold increase in revenue expenditure from 20,000 crore rupees in 2000-01 to 1,60,000 crore rupees in 2023-24, concerns remain over fiscal and revenue deficits.

Issue 2: Healthcare

Keralam’s healthcare sector has made sustained gains for over two decades with improved neonatal (3.4 per 1,000 live births), infant (4.4 per 1,000 live births), under-five mortality (5.2 per 1,000 live births) rates, and maternal mortality (19 per 100,000 live births) rates. High literacy rates, public health awareness and decentralized governance have been credited for this development.

However, a rising proportion of non-communicable diseases, rapidly aging population, and gaps in emergency and trauma care have overwhelmed the system. Moreover, aggressive investments by global private equity funds in the state’s privately-owned hospitals have sparked concerns over rising out-of-pocket expenditure and the closure of small-scale hospitals in rural areas.

Issue 3: Education

With universal literacy, near-total primary school enrollment, and strong gender parity, Keralam’s model of education continues to draw global scrutiny. However, a lack of inclusivity and equity, weak infrastructure, and gaps in academic governance are straining the system.

Issue 4: Agriculture

The agricultural sector plays a crucial role in the state’s economy as it offers livelihood to approximately two-thirds of the total population. The average land holding size is 0.12 hectares while the proportion of small and marginal farmers is 92%. The most cultivated crops in the state include paddy, coconut, rubber and arecanut.

Current problems confronting the sector include limited land availability for cultivation, rising production costs, erratic monsoons, and challenges in selling fresh agricultural produce.

Share

Issue 5: Women’s empowerment and security

The state government launched the Women’s Security Scheme in February 2026, aimed at supporting women and transgender women not covered under any existing social welfare programs. A monthly pension of 1,000 rupees has been earmarked for beneficiaries between 35 and 60 years of age. An initial allocation of 105.87 crore rupees has been made for the same.

Issue 6: SC, ST and OBC welfare

Numerous welfare schemes focusing on education, housing and financial independence are offered to SC, ST, and OBC communities by the state government.

However, despite such measures, Keralam was found to have a larger income gap between general and socially marginalised categories in rural and some urban areas, according to a Moneycontrol analysis of Household Consumption Expenditure.

For Scheduled Caste communities in rural Keralam, The consumption gap between SC communities and members of the general category was 1.58 times in rural Keralam, compared with the national average of 1.38 times. A general category person spent 1.42 times more than an individual from a Scheduled Tribe residing in Keralam, compared to the national average of 1.2 times.

Issue 7: Connectivity and modernization

Keralam is currently undergoing a transformation in connectivity and modernization, with the expansion of ports, roads, railways, and digitization. It is the first state in the country that declared access to the internet as a basic fundamental right in 2019. Efforts have also been made to promote sustainable development through wind and solar power projects.

However, a growing population, high rates of unemployment among educated, young people, and a heavy reliance on remittances have led to low industrial growth.

“Keralam’s experience illustrates the complex trade-off between social progress and economic sustainability,” according to academics Maya K. and Neeraj Kumar. “The state has demonstrated that prioritising human development can lead to remarkable social outcomes, even without rapid industrialisation. Yet, sustaining this model requires a reimagined approach, one that strengthens fiscal discipline, expands revenue bases, promotes sustainable industries, and protects the environment,” they add.

Issue 8: Housing

The Livelihood Inclusion and Financial Empowerment (LIFE) Mission was launched by the state government in 2016 to provide safe, permanent housing for landless and homeless residents. According to the Keralam State Planning Board (KSPB), about 30.78 lakh homes were completed in 2025.

The government also launched a landmark housing scheme for transgender persons in August 2025, with priority for non-home owners living in temporary or unsafe conditions, the elderly, and persons with disabilities.

Moreover, The Kerala Single Dwelling Place Protection Bill 2025, passed in October 2025, offers protection to families from losing their mortgaged homes due to non-repayment of loans.

Issue 9: Climate and environmental governance

The loss of approximately 500 hectares of Kerala’s coastal land due to erosion affected at least 79,000 people between 2002 and 2012. Moreover, projections indicate that four districts could be submerged completely by 2050.

The Kerala State Action Plan on Climate Change 2023-30 projects an increase in instances of extreme rainfall, adversely impacting natural resources and socio-economic systems. Wayanad, Kozhikode, Kasaragod, Palakkad, Alappuzha, Idukki, Kannur, Malappuram, and Kollam districts are highly vulnerable to climate change.

The tone and emphasis of how this issue will be addressed differs in the election manifestos. “Infrastructure-led growth remains the central political promise, while ecology is treated as a constraint to be managed, not a boundary that defines policy,” journalist K.A. Shaji writes for Down to Earth. While the LDF proposed Keralam as an industrially-diverse economy, the UDF manifesto looks at individual projects rather than focus on systemic policy. The NDA, on the other hand, prioritizes high infrastructure-driven growth with little consideration of the diverse ecological vulnerabilities the state faces.

“Roads, railways, ports, and corridors dominate the imagination of progress. Ecology is secondary, often framed as an obstacle or an afterthought. Climate impacts are normalised rather than politicised… [W]hat emerges is a continuation of incremental adjustments within an existing model of growth,” Shaji adds.

Issue 10: Gig workers, expatriates and migrant laborers

The manifestos address these groups unevenly and with little specificity.

The Kerala State Platform-Based Gig Workers (Registration and Welfare) Bill 2024 was expected to be tabled in October 2025, but no further developments are available.

Meanwhile, over 3 million Keralites live abroad and send remittances worth approximately 2.16 lakh crore.

No official data on the number of migrant workers exist in the state. However, the government says that over 5.16 lakh migrant workers have registered under the Aawaaz health insurance scheme. Moreover, about 4.2 lakh migrant workers have registered with the dedicated migrant workers’ Athidhi portal/app and 1.65 lakh under the Inter-State Migrant Workers Welfare Scheme.

📌 Have you joined Feminist Election Newsroom yet? FEN is BehanBox’s initiative to build a collective feminist practice of democratic engagement—tracking, interrogating, and holding power accountable, during elections and beyond. Sign up here.

In the fray are the NDA (BJP + Asom Gana Parishad), and a Congress‑led alliance (Congress + CPI(M) + Assam Jatiya Parishad), among others.

The discourse this season is dominated by identity and religion. The incumbent NDA government’s priority, it has time and again stated, is to expedite detection and pushback of ‘‘illegal immigrants’’ to “protect the land and dignity” of indigenous people and provide land rights to “genuine citizens”. Common to all parties is the implementation of the Assam Accord, signed between AASU, AAGSP and the Central Government on 15 August 1985.

Other issues that parties are focusing on include youth unemployment, flood mitigation, recognition of tribal communities, welfare for tea tribes, women-centred cash schemes, and Zubeen Garg’s death.

Summary of election manifestos of main parties and coalitions

Issue 1: Identity politics

Illegal immigration has been a chronic flashpoint since the 1960s in the state. Hemanta Biswa Sarma’s government has been stoking fears about a ‘conspiracy’ to change Assam’s demography and reduce Hindus to a minority. This explains the uniform demand to implement the Assam Accord and a violent crackdown on Bengali-speaking Muslims.

The systematic shrinking of citizenship has played out through three bureaucratic exercises. In 1997, the Election Commission introduced the ‘D-voter’ system to remove suspected ‘illegal immigrants’ from voter rolls by putting a ‘D’ (doubtful) before their names. As of 2024, the state has designated 96,987 voters as D-voters. [Sarma said lakhs of D-voters were deleted from Assam’s electoral rolls this year.]

The same D-voters are stuck in a democratic limbo, required to appear before foreign tribunals which have sent hundreds to detention centres [decisions that have also been struck down later by high courts]. The National Register of Citizens (NRC) in 2019 left 19 lakh people in a state of suspended franchise.

In the lead up to the election, the citizenship of thousands of Bengali-speaking muslims (deprecatorily called ‘miyas’) hangs in balance. The Sarma government has “pushed back” hundreds of Bengali-speaking Muslims to Bangladesh, or evicted people from homes and bulldozed their properties.

“For his agenda, [Sarma] must create [a spectre] of illegal migrants in Assam. The detention model serves him as a tool to this end,” an advocate with Gauhati High Court told Scroll.

Issue 2: Land Rights and Indigenous Identity

The pitch is loudest to indigenous tribes. Parties promise to ensure the economic and social security, and “protect the land and dignity” of indigenous people. This translates into offering the ST status demand from six communities—Tai Ahom, Moran, Matak, Chutia, Tea Tribes, Koch-Rajbongshi—which has been an electoral issue since 1996 (a primary issue in more than 45 seats in Assam). BJP has promised it since 2016 but no Cabinet notification was issued.

Issue 3: Informal Workers and Social Security

The central government raised its ASHA honorarium contribution from Rs 2,000 to Rs 3,500 from March 2025 — but Assam had not implemented even this revision as of November 2025, triggering statewide strikes. No party has talked about formalising ASHA and Anganwadi workers.

The Sarma government promised regularisation and 7th Pay Commission salaries in his very first cabinet meeting in 2021—neither was delivered.

Issue 4: Women

The state implemented in 2020 the Orunodoi program – a cash transfer scheme where Rs 1,250 per month is paid to 40 lakh women. The incumbent government announced a transfer of Rs 9,000 on March 8 (Women’s Day) to the beneficiaries – the largest such disbursal in the state. Recently, Congress MP Priyanka Gandhi Vadra at a rally alleged women beneficiaries were being forced to attend political rallies.

Notably, the Orunodoi excludes contractual workers, ASHA workers, Anganwadi workers and government employees.

No party talks about structural challenges: domestic violence, poor enforcement of laws, economic dependency, prevailing child marriage concerns (researchers have questioned the Sarma government’s policy on criminalising child marriages), or bodily autonomy. Additionally, the state government passed an amendment to repeal the Muslim Marriage and Divorce Act, 1935 in February 2024, which as we reported, would do little to empower women.

The Mukhyamantri Mahila Udyamita Abhiyaan (MMUA) aims to provide almost 4 million women who are members of self help groups (SHG) a minimum annual income of Rs 1 lakh each, excludes women who have more than two children, disproportionately affecting women from the most marginalised groups.

Share

Issue 5: Education and Employment

Assam faces high youth unemployment and no big local industries to provide people jobs – there are almost 21.16 lakh educated job seekers as of 2024 (a number that has risen over the years). The unemployment rate is at 3.9%, a sharp increase from 1.7% in 2022-23 and above the national average of 3.2%, according to the Economic Survey of Assam 2025-26.

Unemployment has led to rising out-migration to urban metropolitan areas like Bengaluru, mostly by tribes like Mising who are not included in the Sixth Schedule.

Demand both for secure government jobs (more so in rural areas) and for startup support among the urban youth.

Issue 6: Tea Workers

Assam has more than seven lakh tea workers (of Adivasi ancestry), spread over 800 tea gardens and several other unorganised small gardens, and they constitute about 18-20% of the electorate. Daily wages are a pain point – workers earn around Rs 250 in private gardens in Brahmaputra, but this can come down to Rs 150-180 in the unorganised sector.

BJP promised a wage increase in 2021 but never legislated it, so implementation is patchy. The incumbent government notified an interim increase of Rs 30 in wages.

Workers are for better access to healthcare, education, sanitation facilities, and a stricter implementation of the Plantations Labour Act 1951. There is also a demand for land ownership rights for families who have lived on these plantations for decades.

Issue 7: Floods, Erosion & Climate Change

6 of India’s 8 flood-prone districts are in Assam. Floods, embankment breaches and erosion displace communities, destroy crops, and cost millions annually. The incumbent NDA government has aggressively pushed for expanding embankment infrastructure (it has built 1,000 km of new embankments since 2021). But scholars like Arupjyoti Saikia have argued against boosting these projects, citing the Garg Committee reporting of 1951 that warned these measures disrupt natural processes, fail to provide long-term relief, and worsens flood control efforts.

Urban flooding is a stress point for Guwahati, the new Gen-Z Kunki Chowdhury AJP candidate is one of the few talking about drainage failures.

Issue 8: Health & Elderly Care

Parties have focused on higher insurance coverage ceilings and medical college infrastructure—but structural gaps in terms of primary care, availability of rural doctors, and high outpatient costs go unaddressed. The state faces an acute shortage of doctors, specialists and technical staff, some of whom were suspended for demanding higher pay.

The state’s geography poses challenges with annual flooding and erosion that damages infrastructure. Remote riverine areas face unique barriers, often making boat clinics the only access point. And due to poor road connectivity, about 39% of rural people report transportation as a major barrier, with nearly 31% spending over Rs 60 to reach a health facility. Transportation issues create barriers for women, informal workers, Scheduled Tribes, and elderly individuals.

Issue 9: Governance

Everything from corruption and development to culturally sensitive issues like Zubeen Garg’s death have found space in Assam’s political vocabulary. Congress has repeatedly called Assam’s the “most corrupt government”. This year, Congress representative Supriya Shinate said Sarma’s personal assets doubled over the last decade and their family wealth increased six times (he declared no immovable assets in affidavits). This was an ‘asset accumulation model’ and not ‘Assam model’, she said.

Almost all parties have made singer Zubeen Garg’s death a poll issue, alleging the incumbent government was involved in a cover-up, as emotional resonance across communities.

📌 Have you joined Feminist Election Newsroom yet? FEN is BehanBox’s initiative to build a collective feminist practice of democratic engagement—tracking, interrogating, and holding power accountable, during elections and beyond. Sign up here.

Joan Kinyua once loved technology, but she is now betrayed by its promise. Enticed by the pay and glimmering future of the rising data annotation industry in Africa, she found disillusionment setting in swiftly and suddenly: the pay didn’t justify the work, there was no social security, and the job and environment often led to burnout and trauma. Worse, the manual labour powering AI revolution went unrecognised, and these digital futures replicated imperial power structures of the past.

She, along with her colleague Michael Geoffrey Asia and Ephantus Kanyugi, started the Data Labellers Association that enables workers to have a shared space, vocabulary, and bargaining power against global tech companies. “It is African Intelligence powering European intelligence, which they are now calling Artificial Intelligence,” Michael says.

In the last two years, the DLA has mobilised workers across the country, developed standards for dignified work conditions, and sparked a conversation on mental health, an occupational hazard in this field.

In a conversation with Saumya Kalia, Joan and Michael talk about the painstaking labour of mobilising data workers and the fight against racial discrimination. There is a need for cross border organising with the Philippines and India, which house the majority of data workers, they say. And the world needs to see beyond the manufactured hype sustained by Big Tech, governments, and media. “Big Tech needs to understand this is someone’s job, someone’s economy. They need to understand that these are human beings who are making them billionaires…It’s not a crime to have a win-win situation. It’s practical, it’s been proven, and it can still work,” Michael says.

Edited excerpts below.

How did you hear about data labelling jobs, how did you get into this space?

Joan: I started in an organisation called Sama (back then it was called Sama Source). I am professionally trained as a business administrator and was working in a school earlier, but I did not like that work. When my sister got a job at Sama, she would tell me about what this work is and how they have an open office and are working eight hours and how everyone is really friendly and how they’re working on this technology of self-driving cars. I felt this is what I wanted to do.

I soon resigned from my job and applied there; within a week or two, they called me for a data labelling job. The training was supposed to happen in a slum called Kibera, where I live. The companies wanted candidates to present themselves as having only a school education, regardless of whether they had degrees, professional training, or other expertise. If you mentioned additional qualifications, you risked being labeled “overqualified”. I followed the expectation and only stated that I had finished school in order to secure the job. I love technology and I love computers, and I was extremely good at the training which lasted for two weeks.

It was only three days before I was informed of an opportunity. The conversations in the start were very welcoming. They told us about working on cutting-edge technologies like self-driving cars, but we were also told not to communicate with anybody in case we saw something bad or if something wasn’t working. As a newbie, you’re excited and grateful, so I obeyed the rules.

Fast forward to around eight months later, I started noticing things. Firstly, the salary did not make sense. It was only facilitating your conditions of work, but not doing anything to make you independent in the long run. It’s like you’re just fueling the system but you’re not benefiting from the system.

Things became very bad when the company started directly targeting me when I became vocal about issues.

What was the consequence of speaking up?

Once our team leader warned they were going to box [isolate] us. I asked: if we are grown-ups, why would you box anybody for bad quality of work? I was called out for “speaking too much”. It got bad. I was targeted for my work; I used to get a quality [rating] of 100% on my tasks but they started removing chunks and finding mistakes to the point I was getting 56%. Then I was given a PIP, a Performance Improvement Plan. I moved from having good quality work to a PIP stage four, which meant that I was seen as a liability to the organisation.

[PIP is an example of manufacturing performance decline to justify dismissing workers who raise voices. Employers also routinely employ surveillance systems that alert management when people are talking in a bathroom or if they say the word ‘union’, to track “troublemakers” and manipulate task allocation to hinder organising, per this research.]

I did not want that bad reputation to affect my career prospects so I started looking for another job in the same sector. I was lucky enough to get a job at another organisation called CloudFactory; I transitioned from a night shift at Sama and started at CloudFactory the very same morning. I thought I had found a better opportunity, CloudFactory was marketed as a top-ranked BPO with higher pay and better projects. However, the reality was quite different. At Sama, the organisation was bad but you had medical cover and a contract. At the new organisation, you sign monthly contracts and are expected to earn 24,000 Kenyan Shillings or below because anything above that is taxable. There were no medical covers, social protections, or other benefits – you were essentially on your own.

Share BehanBox’s Substack

Here, I also learned about Remotasks, a separate data annotation platform owned by Scale AI, where individual tasks fetched $30–40, significantly more than what I was earning.

Back then, I didn’t have a laptop or Wi-Fi at home, so I would wake up extremely early, around 4 am, so that by 5.30 am, I’d have reported to the office. During the day, I focused on CloudFactory work, and later in the evening, I would pretend to leave, hide in the washroom, and access work for Remotasks once the scene was clear.

Eventually, I was found out and fired. The official reason given was that I had breached my contract by working for another organisation performing the same type of tasks. But I continued with the formation of the association and talking to people.

Is this when the Data Labellers Association came up?

Somebody reached out to me for a research project in this field. My role as a research assistant was to interview 50 to 100 people of the same background as me and speak about the landscape. People spoke to us about many different challenges. For instance, the skilling issue, there is no way we can take the skills that we’ve got here to other spaces; once you’re fired from this organisation, you look for another bad organisation and then just continue the cycle because you do not have the [other] skills for the market.

When it comes to mental health, we are constantly living on the edge. People are asked to watch damaging, devastating content. These jobs are also not predictable… When it comes to the salary, back then when we started doing it, we were being paid very well. As more people started working on this platform, now you’re being paid some 0.something dollars for a task you’re working for five hours. Things are starting to get bad.

When I was working, I just thought these work conditions are normal and this is how people are supposed to be treated. But everyday, as we discussed the interviews with the head of the research, we realised more and more the exploitation built within the system. These things were not supposed to be happening.

I sat my friends down, Michael was one of them. We wanted to come up with an institution that can address the skilling risk. But the researcher noted this is a bigger issue than just skilling, and encouraged us to form a body that speaks about worker challenges.

We researched for a year and spent the whole of 2024 just speaking to people. Back then, it was difficult to get people to understand what data labelling was. There was conversation about gig workers but people were focused on ride-hailing sectors. We had to establish a space for ourselves. We formalised the association at the end of the year. It was recognising these systemic challenges that pushed us to form the association.

How did you connect with other workers? In India, unions have started using digital methods to mobilise gig workers but their reach is still low among data workers, who work remotely and include a lot of women with carework responsibilities.

Joan: Workers sign NDAs that explicitly state they’re not supposed to be part of unions so we had to be careful. People do not want to lose jobs so we have to come up with mechanisms to protect the workers. That meant no online interaction, because you do not know who is going to be part of the conversation and they might face retaliation from the organisation. [Platforms like Urban Company and Blinkit in India have employed similar union-busting tactics.]

Because we are very protective of the workers, there is no direct retaliation from the employer because our numbers are not made public. We do not give anybody information about who is or who isn’t part of the association.

In this field, we both know and don’t know each other. I knew Michael personally from earlier but I first saw his face in 2024. We have worked with these people for more than five years, but we do not really know them. These groups – on Telegram, WhatsApp, Signal – give us a new opportunity to reach workers.

Share

Also, besides the three of us, the rest of the members are still actively working their data labelling jobs. That makes it easier for us to have conversations because we haven’t lost touch with people on the ground. We all started in the same place so there is that trust.

Michael: It also depends on how you’re related to them. Personally, I was very vocal at work and so someone meets you somewhere and is like, “I think I’ve seen you. You’ve been working on this and this space.” That trust and familiarity is how it started.

What is the perception about this work among data workers themselves? What did you want to demystify for the public?

Michael: People understand that these conditions are exploitative but they do not have an alternative, so they remain in the jobs out of desperation. We try to tell them a job is supposed to be a source of satisfaction, not just something to survive on. These jobs strategically have very poor working conditions, to ensure that we are always in survival mode. They target people between the ages of 18 and 23 – young people who do not have responsibilities. If someone gives you $200 per month and you are still living with your parents, you do not feel like it is a burden. But for someone like me who has a family, who has to lay rent and school fees, I will realise that $200 is not enough. It opens eyes for me.

It is hard, trying to bring these people to the table; most people do not know they are in survival mode. But they are beginning to understand what we mean by lack of adequate income.

Joan: We are also seeing people self-harming because of these jobs – that also makes people join associations or unions.

There is a knowledge lacuna about this work among policy makers and the general public too. Most people did not know what data labeling is – we had to establish its definition and how important it is in the Artificial Intelligence field. We took it upon ourselves to define these roles. We took it upon ourselves to be part of conversations where we were not invited. And we took it upon ourselves to speak about how people are being treated, to name and shame these organisations.

Big Tech organisations are very clever in squashing our voices. An organisation like Sama, for instance, is present in Kenya and other countries for the exact same job. Once they figure out that Kenyans are mobilising or are making noises or want more say in deciding the final product, they shift the work to another country. People are afraid these jobs will move so they end up being defensive of tech companies and these jobs. But we try to explain to them the companies are only concerned about the job, not the workers.

It is important for us to speak about the reality of this work not only on a regional but a national level. These Big Tech companies deliberately do not want workers to speak up or to be known – we are creating a space to challenge this.

Can you tell us more about the mental health aspects of this work? What makes it an occupational hazard for data workers?

Michael: You’re right, mental health is not recognised as a professional issue. We need to address it from a sector point of view, not a general point of view. Right now people believe content moderators have it easy – they are working online, facing a computer and earning a dollar. That is not the case.

For example, people are being killed in the ongoing war between Iran and Israel, but there’s someone who’s filtering this information so that it doesn’t reach consumers on the platform. What a content moderator is seeing online is not what a doctor sees in an emergency unit of the hospital, but both can cause mental health issues.

I have personally worked on moderating pornographic content. You cannot expect someone to work on pornography for eight hours everyday, for months, and still be the same. It was not easy.

Joan: Mental health is a very touchy subject but you can see the effects of this work. You can see how desensitised people have become due to the nature of the content they are exposed to, much of it harmful or disturbing. Many workers are affected, yet they either do not recognise the impact on their own mental health or are reluctant to acknowledge it. For example, in Kenya in 2024, when the platform Remotasks shut down, many workers were suddenly left without income, and the stress and pressure they had experienced over time were clearly visible. There is a hidden emotional toll of this work, even though it is rarely discussed openly.

[Workers at Sama described their work as “torture”. They found the sessions with ‘wellness counsellors’ unhelpful and rare because the focus was always on worker productivity.]

It’s not only about content but the environment itself. The 23-year-olds being targeted will potentially have families in the future and they will realise $200 is not enough. What happens to your family adds to your stress; you’re constantly living on the edge.

What triggered my panic attack and anxiety was this moment, around my daughter’s birthday. My account was deactivated because I violated community guidelines I was not aware of, and I did not get any money. I tried to appeal but nobody responded to my emails. That caused me to panic, I had already planned what I needed to do with that money, and then I didn’t even have money for diapers. It makes you feel so vulnerable – you’re vulnerable to the platform, to the system.

Mental health is manifesting in very different ways, but some people do not want to address it because of the stigma. Especially as Africans, we do not want to associate with mental health in any way, shape, or form – people are afraid to be labelled. When I suffered from anxiety attacks, I could not speak about it because I was afraid of so many things. How are my friends going to view me, is anything I’m going to say taken into consideration, or are they going to think I’m ‘psycho’ or something like that?

How do you weave mental health into your advocacy if the stigma contributes to silence?

Joan: This has been the hardest issue to surface. We didn’t want to start with mental health at first. Even the mental health department [Taskforce on Mental Health, Ministry of Health] has not been part of the conversation. But now when we speak to policymakers, we do not sugarcoat – we tell this is what these jobs are doing, this is how desensitised people are.

Platforms say they have mental health practitioners but usually it’s wellness coaches, and as content moderators you’re not supposed to speak about the content you’re working on. It’s quite unhelpful.

We are trying to look for mental health practitioners but we do not have any funding, so we want the mental health department to be part of this conversation. To be able to pay for mental health workshops or individual counseling sessions.

We have also found that community engagement — just having a community of people – can also ease workers’ mental tension. It’s like walking on eggshells in this sector. You do not want to harm the workers, but you also do not want to portray them as victims.

How important is cross-border solidarity or international organising for you? What would you tell data workers in India?

Joan: We want everybody on board because if it’s not Kenya providing the manpower, it’s India or Pakistan or the Philippines. Data workers should focus on building strong, cross-border solidarities. Collective organising, shared knowledge, and mutual support across regions are key to ensuring that their voices are heard and that they are not left invisible in a system that depends so heavily on them.

People are currently settling for less than what they deserve. For Kenya, platforms impose quality standards requiring work accuracy of 98% and above [workers whose output falls below a certain quality score get paid less, deprioritised for tasks, or removed]. Most people are doing work that is almost perfect. The quality standards are so high, which only shows how skilled and valuable workers are. But the irony is we are still treated as disposable.

Workers should not feel that they need to despair or overextend themselves, breaking their backs and sacrificing their health, just to deliver for a technology that was never designed with their well-being or interests in mind. Instead, they should recognise the value of the work they do and use that as a foundation to demand better pay, stronger protections, and proper recognition for their contributions.

Researchers have found that colonial power dynamics also show up in India – platforms employed ‘time zone hierarchies’ (reserving higher paying jobs for workers in Global North) and racial wage code; and having an Indian accent meant lower pay. Can you talk more about the racial dynamics at play in Kenya?

Joan: There is also a lot of gatekeeping in payment. People in Kenya are paid $2 an hour but those in the US or EU regions are paid close to $30 to $60 an hour. We were discussing this recently at a session when somebody from the audience said, because you’re Black, you should accept being paid $2. That’s the mentality we’re working with — I’m Black, they’re white. But aren’t we all human, aren’t we putting the same effort to do this job? We need to get rid of that self-discrimination. Arguably yes, these are different economies but the pay rate difference should not be this much.

Michael: I would call this a form of discrimination slavery. In their own country, training tasks earn people as much as $30. In Kenya, someone is working for two weeks without being paid even a single penny on training tasks. The act of someone not paying you or paying you poorly, that is slavery, and it has never had a good track record.

We need to put these conversations into perspective. Today, most Kenyans are struggling to apply for jobs because of what this work has created. When you submit a job application, it gets flagged as AI-generated because of the Kenyan English used. We fed a lion, but today the same same lion that was supposed to protect us is coming back to eat us. It’s pathetic.

[Kenyan English has its own syntax, rhythm, and phrasing, and those same patterns were fed into AI training datasets by workers. So the detection software reads what a Kenyan writes naturally as machine output, as ‘AI-generated content’. AI detection tools have been shown to produce higher false positive rates for Black writers than for white ones.]

After the Second World War, Europe had to survive and the European Marshall Plan was put in place but this post-war recovery assumed continued extraction and access to resources [land, labour] in colonies like Africa. It has happened earlier and it’s still happening. I see AI in this manner: we are still using Africa, for example, to sustain Europe and the West.

I came across an article someone wrote. Why do we have white cars with black tires? Why is it that when you block someone from accessing you, they are put on a ‘black list’? In a war field, a ‘white flag’ denotes peace but a ‘black flag’ shows war. We have regions where people die in most cases through drugs and it’s called a ‘black spot’. Why black? We call it ‘talent’ in Europe and the US, but when you come to Africa, they call it witchcraft.

Unless we stop manufacturing cars with white bodies and black tires, discrimination will never leave our society. Until we get to a point where this philosophy of one race being superior to the other is permanently abolished, where the colour of a man’s eyes is of no significance, then this space is going to continue being messy.

People like the tech, the glitz of AI tools, but we often fail to see what labour and effort goes into making it. What would you want people to know?

Joan: It is very important for people to feel secure and appreciated for the work they do. We only know the importance of our work or feel like we did a good job when we see self-driving cars work, knowing they are not causing accidents. We can see the robot vacuum is doing a good job. But in making that possible, we also had to see people sitting on toilets and doing very weird things which this robot took photos of.

[A recent investigation found Meta’s smart glasses recorded people’s intimate and private lives, collecting sweeping footage which was reviewed by data annotators.]

My point is: where am I in this conversation? Most people rely on basic research out there but they’re not willing to interact with the people who do this work. I don’t know what they’re afraid of.

When we see cutting edge technology, let’s ask the questions that nobody wants to ask. How did this come to be? How was it made? Let’s ask questions; let’s not just embrace AI.

Michael: I work on self-driving cars, but I am not sure if I will ever see self-driving cars here in Kenya before I die. I’ve never even seen streets of Kenya being pictured for the same. This is purely African intelligence powering European intelligence, which they are calling artificial intelligence.

Rights need to be centered in these conversations; people need to stop being blind to what is so obvious and just address the elephant in the room. The elephant in the room is Big Tech and human rights violations. People need to open their eyes.

Leave a comment

AI Summits have a template: government-Big Tech deals at the centre, civil society in consultative roles, a non-binding declaration at the end. The people most affected – gig workers, data annotators, grassroots data rights groups, marginalised communities – are rarely in the room at all.

A feminist AI Summit asks: what would it take to build the room differently.

That means being precise about what’s actually being diagnosed. Phrases like “inclusivity by design” and “responsibility by design” – used generously in the recently concluded AI Summit – assume exclusion is an implementation problem to be engineered away. Getting that diagnosis right would change who counts as an expert, how gender enters the conversation beyond harm-mitigation, and what participation should actually mean while the agenda itself is still being written.

See what Saumya Kalia heard on the sidelines of the India AI ImpactSummit – from researchers and practitioners, including independent technologist and researcher Rohini Lakshané, and Shohini Banerjee, Knowledge Specialist at Point of View.

All illustrations by Urvi Sawant.

Leave a comment

At India’s AI Impact Summit, the buzzwords came thick and fast: inclusive, sovereign, ethical, resilient, democratic. The language of rights movements and civil society, deployed, with a straight face, by the same companies whose supply chains run on underpaid data workers, mined minerals, and depleted water tables.

Shivangi Narayan, researcher on gender and technology, sat through it and offered a tongue-in-cheek decoder of what each term actually signals – for the communities that will live with these systems long after the summit wraps.

Share

The AI Impact Summit has ended, and everyone – from tech companies and industry lobbies to policymakers and researchers – spoke about building equitable AI systems and a future where AI is a public good. What was missing, however, was evidence, specifics, and communities themselves.

AI regulation falls into the larger arc of India’s approach towards IT and digital industries. We asked Anita Gurumurthy and Nandini Chami, researchers at IT For Change, a non-profit working on technology and society, to put into context the emerging debates around AI. How is India trying to govern AI technologies? What economic and political thinking influenced this? Where are gender and caste in these conversations?

Over email, before and after the summit, the researchers spoke about how India’s ‘market-first’ and ‘light touch’ approach has allowed tech companies to continue building their surveillance empires without safeguards.

While the focus seems to be on privacy, protection and sovereignty, we fail to ask fundamental questions, the researchers told Saumya Kalia. “We need to press the delete button on the halo around AI and the myth of ungovernability – AI is firmly and squarely an issue for public reason,” they say. Edited excerpts below.

Let’s start with the summit. Why does it matter that it is happening in India, and why now? What economic and political trends are shaping the agenda?

In the official narratives, the USP of this summit is that it shifts the conversation from mere declarations to tangible action. The India AI Expo, Knowledge Compendiums and Case Books on Applied AI, CEO Roundtable on Responsible AI–showcasing Indian industry and DPI [digital public infrastructure] for AI models–is clearly the centrepiece of this agenda.

At the same time, what remains uncertain is the question of the extent to which the Summit will produce concrete gains for India. 88 countries may have signed onto India’s ‘AI for all’ vision in the Summit’s final declaration, but it remains to be seen whether India will be able to lead a South-first AI paradigm, overcoming dependence on US big tech in practice, and neo-colonial equations in digital trade.

Geopolitically, India is on the backfoot in protecting its nascent data and AI industry from the onslaught of UK, US and EU markets. In the BRICS [a coalition of 11 countries], India has supported declarations on data sovereignty and development-oriented visions of AI governance. This rhetoric comes back at the Summit through themes like Indic language models, open cloud infrastructure etc, but it is unclear how technocratic choices and innovation start-ups will tackle social inequalities or deepen public purpose.

A lot of AI startups are in fact valued at several times over their actual valuation. This makes them moonshot projects, with only a few players making money right now. India had recognised 1,61,150 startups by January 2025, yet only a subset builds core AI products.

Tech companies and AI startups had a dominant presence at the Summit. Will the presence of industry reshape the ‘global IT’ agenda? Who or what could be left out?

Big Tech has always shaped the global IT and digital governance agendas. This is neither new nor surprising. The ‘Key Visionaries’ page of the Summit, the processes of the thematic working groups – all give primacy to an expert-led narrative of AI governance.

The agenda of public accountability and social purpose innovation–integral to the core sutras of People, Planet and Progress–were often a postscript to the market-first narratives that dominated the discussion.

With a beast that brings unknowns, even with its scintillating promise, the governance question is all-important. It is necessary because AI pervades everything. It is a technology that is re-shaping market, civic, and governance spheres, opening up new questions about power and agency and troubling pre-existing benchmarks.

There is also value in seeing the AI governance question in terms of constitutional values. This will logically open up the ‘who’ and ‘how’ aspects of agenda-setting–where principles of federalism, societal well-being, freedom from harm, the rights of future generations, and many others–will guide and determine which publics emerge as relevant to what debates.

In 2019, you wrote about the ‘wicked problem’ of AI governance. Wicked because of the struggle to find the right combination of legal, ethical, and technological regulations. Which of these has dominated in India and what does the choice reveal about India’s priorities?

‘Wicked’ problems elide simplistic, linear fixes. They need a complete grasp of the complex socio-political nature of societal structures.

Share BehanBox’s Substack

The technocratic imaginaries that have defined India’s digital governance approach over the past two decades have persisted in the AI context. Building locally relevant AI in the reality of global infrastructural divides is not easy. We have to think about orchestrating a governance environment that aligns our concerns around inclusion, transformation, and more with resource allocation, public interest, ethics, citizen participation, public-private boundary management etc. It is about building a pathway to industrialisation all over again in the AI moment. Aspirations to become the AI garage of the world may need to wait till the home environment can throw up sufficient learnings.

Since 2019, India has tended to valorise the speed of AI adoption in the domestic market. Diffusion is not an intrinsic public good; the terrifying stories of social media influence on young people testify to the crises of the AI society. Public policy is mainly directed towards getting on top of the global AI race – underwriting compute investments of domestic firms, opening up public datasets to market actors, and providing preferential tax incentives, IP protections, and access to funding for deep-tech firms.

This market-first approach to AI may not do justice to India’s socio-economic challenges. Our current ‘light-touch’ regulation which wants to establish market supremacy of India’s AI sidesteps intentional, whole-of-society governance.

We reflected a few years ago about the North-South AI divide and argued that data and AI are two sides of the innovation coin, and that resource (re)distribution at the global level is vital for an equitable AI economy. Our concern about equity of data access remains.

Today, we also see that the country AI trajectory badly needs a rule-of-law-based approach rooted in constitutional values. As things stand, the rights of citizens, the duty of the State, and the accountabilities of the market are but weak and tenuous in the national policy vision for AI. Also, how the data domain will be organised and governed for social good remains an open question at the national level.

What would a constitutional and rights-centred approach to AI governance look like? Would a community-led approach also be automatically anti-caste?

This is an important starting point to define the relationship between tech and society. What public interest tech communities are telling us is that AI ecosystems are built on and deeply reinforce the axes of caste and gender. Discrimination, exclusion, and exploitation on the lines of caste and gender are encoded in large language models (LLMs) and AI applications (consider how people generated non-consensual intimate images on Grok).

The Constitution of India recognises the right to equality as a positive right – moving beyond freedom from discrimination to enshrining affirmative action for furthering an equal society. It is important to extend this constitutional, anti-caste framework to address new challenges in the AI moment. What is needed is a horizontal omnibus legislation for extending fundamental rights in the AI moment (including affirmative action for the protection of vulnerable groups), sector-specific legislation, and new institutional bodies/mechanisms for the enforcement of these rights.

Some steps have been taken by the government, like developing sector-specific strategies. However, this thinking currently privileges unhindered AI innovation. A patchy and fragmented rights approach will not provide the same protection that a re-reading of constitutional morality can offer for governing AI.

Going forward, we need to move away from focusing only on product ‘bias’ and instead, examine the mode of production itself in order to better address caste, gender, and other social biases. This means foregrounding and making visible human and socio-ecological costs in data and AI value chains, and preventing structural harms, including those based on caste and gender lines, from being entrenched.

Share

Without effective rights to information about and participation in AI policy, social progress is bound to be thwarted, directly undermining the ability of people – historically marginalised communities, environmental defenders, small economic actors, indigenous people – to shape AI futures.

Summit conversations on data sovereignty–Indic language models, open cloud infrastructure—happened without the participation of grassroots organisations or unions. What do you make of the representation gaps?

The India Summit could certainly have encompassed a more diverse range of voices. The representation gaps at the Summit reflect the wider terrain of the AI debate: Those who matter, and those whose lives are already entangled in disempowering ways in the dominant AI paradigm, are seen to be peripheral. Having more seats at a Summit table does count. But what is more important is that demands are consolidated and channelled across multiple political spaces to expose and address the public accountability gaps in AI governance.

The short history of global AI Summits reveals a largely celebratory and homogenising narrative about AI. In any case, as with large spectacles, these moments are rarely the right avenues for serious public debate.

The India Summit did open up some space for a South-led AI vision, which civil society can use tactically for asserting pluralistic visions of the South – those that can emerge from the ground, those that challenge state-led rhetoric. But civil society and global justice organisations need to work beyond the surface-level questions about infrastructures in order to reject both technological determinism (this dogmatic faith in tech revolution as a driver of social progress) and long-termism (the inevitability of AI’s existential risks to humanity becoming a reality).

Our concerns about structural justice must address the material foundations of the AI economy and how they play out in practice. Yes, openness is important, local language models are critical. But we need to ask what kind of openness, and what kind of public goods, can support a resilient society that truly has the capability and freedom to pursue data science towards goals that it sets for itself.

AI is not a brand new technology and we’ve seen cycles of AI hype and ‘AI winters’ earlier. Can we cast a wider look: how has India tried to regulate or steer conversations around AI? Do today’s debates continue or disrupt earlier thinking?

There is a persisting blindspot, when we attempt to place the present discourse on AI governance in the long arc of IT and digital policy. There is an absence of a robust industrial policy that leverages the general purpose nature of network, data, and AI technologies.

Scholars have highlighted how the growth in software and IT-enabled sectors since the 1980s has not been leveraged to add value to the domestic economy and transform core sectors. These sectors have also employed only a minuscule percentage of the educated elite. New Gen AI developments, such as AI productivity assistants and ‘co-pilots’ that automate thinking, threaten these gains.

After the liberalisation of the Indian economy in the 1990s, the country has largely been a consumption market for digital products and high-end digital services, with some limited contributions to the low-value segments of the global digital value chain [such as call centres, data processing, and back-office services in the IT and ITeS industry].

Since 2014, the Union Government has responded to this challenge in various ways, including focusing on strengthening the domestic digital start-up and innovation sector. But as commerce and industry minister Piyush Goyal reflected in April last year, start-ups have mainly come up in ‘rent-seeking sectors’ [those that manipulate existing economic conditions to generate profit without creating new wealth] such as quick commerce. And India lacks deep-tech innovation in sectors like EV, semi-conductors, industrial robotics.

The limitations of our innovation culture have haunted us over a long time; the past decade of attempts to bring data and AI industrial policy as a corrective has been insufficient.

India first wanted to keep data within the country to protect its economic interests. Over time and with the LLM boom, it became more willing to share data with certain countries and companies. But it still sees data as crucial for national development. Similar lines of thinking pervade the AI governance discourse as well – AI for Atmanirbhar Bharat is a running thread in the narrative, which is also one of the pillars of the India AI Mission.

Where does India stand in terms of AI and data governance in the world?

The shifts in India’s stances reflect a moving carrot – an ongoing tension between domestic digital industrial priorities and the opportunities offered by the global digital economy. We have come a long way. Earlier, India had taken unique positions, such as recognising community rights over non-personal data and placing limits on cross-border data flows. But those were still the days before the ChatGPT moment.

Our position, today, as the digital trade negotiations of the India-UK agreement show, is pretty close to the mainstream agenda pushed by the Big Tech lobby. Our industry wants access to global data markets, open government data, foreign capital and so on. This does not sit neatly with India’s rhetorical assertions about development sovereignty in the global digital/AI order; after all, development cannot mean greater wealth for some while deepening social inequalities for others.

What is needed is intentional governance that places equitable development at the centre of our data and AI governance approach. One thing we have partly got right is the connection between DPI and AI innovation. The next step is to ensure smaller actors also benefit and that social-purpose innovation is not left to market interests.

Signal’s founder Meredith Whitaker described today’s tech economy as a “surveillance empire” where you monetise data to a point the whole system becomes unstable, like a centipede with too many legs that eventually “collapses”. Has any Indian framework taken note of this business model, do you see any real safeguards?

No, and that’s what worries us. We currently have a ‘light-touch’ regulatory approach to the AI market and lack sufficient purpose limitation and other ethical boundaries in our data stacks for key social sectors such as health and agriculture. This may be a ticking time-bomb or a formula for slow societal atrophy. This may sound hyper-sceptical – but we believe that policy frames must adopt a worldview that treats even the magical as an ingredient to be examined for the old question of power and control. The glitz of AI may be undeniable; but it may leave behind, or conveniently erase, the ordinary.

Research has shown women and gender-diverse people remain less represented in AI jobs, especially leadership positions. How does that impact governance frameworks? Have Indian frameworks planned to engage with these concerns, particularly risks of surveillance and violence?

The politics of representation in the industry does impact bias in products. But it is also true that India’s laws don’t deal with AI systems that profile or harm whole groups like caste communities and women. Discrimination is rooted in deep social structures; to challenge it, we need continuous post-market assessments of products. One-time certification of risk is partial at best and dangerous at worst; you can’t approve an AI system once and forget about it, harms often show up only after people start using it. These systems need to be watched continuously, and their limits should be set by constitutional values like equality and dignity.

There is some policy attention to the risks of synthetic media and image generator AI technology for gendered violence, but progress still remains to be made to build public trust. We may need to learn from other jurisdictions – the debate on societal data rights in Brazil, ongoing discussions in Australia about treating online digital footprints as personal data, the experience of the Network Enforcement Act in Germany (the NetzDG law), EU’s AI omnibus and so on.

At the same time, there are very interesting civil society developments – the Samiksha LLM pipeline evaluation and innovations like Shhor.ai and other research efforts to map gender-based harms in AI.

Our current focus on weaponisation of AI technology has been on DigiYatra and facial recognition technologies (FRTs). What are we missing?

The weaponisation of AI technology is witnessed in so many instances – commercial spyware, clandestine state surveillance, cyberespionage. The manipulative impulses of surveillance capitalism destroy the social fabric, they annihilate democratic culture. We are mute spectators to bots and troll armies running amok, and AI functionalities that are inimical to the dignity of women and marginalised groups. This weaponisation of social media has acquired new dimensions with the advent of synthetic media.

Another emergent threat is the potential for agentic AI to be weaponised by malicious actors.

Subscribe now

As AI agents become ubiquitous in public service sectors – banking, welfare delivery, etc – this will become a major security risk, as entire systems, public systems, can now be held under siege. We are missing a nuanced conversation on how to set redlines for the deployment of agentic AI.

So far we have been focusing on parity, privacy, and bias. What rights are missing from this conversation?

We need to be talking more about economic, social and cultural rights in AI value chains. This is beyond decent work for data labour. It is about the concerns of farmers and producers in agritech value chains, cultural and indigenous communities whose knowledge commons and creations have been taken into generative AI data pools, communities displaced by data centres, and so much more.

We also need to be centering questions of epistemic rights – what is AI doing to our societal knowledge commons? Our sense-making and meaning-making as societies? Our knowledge and lifeworld pluralisms? In essence, the right of all to live a life of vitality and real autonomy, and our collective right to infrastructures of care and creativity. These are the faultlines that life under AI is visibilising.

You wrote about the silence around thinking of AI as a ‘public common’. Why does that silence exist?

When we wrote our paper in 2019, we were concerned that social media algorithms would commodify all of sociality; the commons of our everyday interactions. The concern was that not many were taking the threat seriously. But today, after the Gen AI boom, it is evident that fundamental questions about market-led AI innovation are deliberately buried. Questions about the data that is used for training models – originating in the plunder of the knowledge commons – have been reduced to semantics about copyright law and its fair use exemptions.

We are of the view that an internationally binding framework that centres a commons-based approach to the governance of training data – covering the entire lifecycle of AI, including innovation capability – is essential to address the current challenge of knowledge colonisation and cannibalisation of the public domain.

As an innovation built on the collective resource of social data, AI cannot be allowed to be monopolised by a few. It must be protected and preserved as an epistemic infrastructure that underpins and enriches people’s knowledge commons. But how this framework will be enacted and arrived at in the global political economy – that’s the million-dollar question.

Leave a comment

It is unmissable, this exposition of India’s AI vision. The India AI Impact Summit 2026 brings together governments, tech companies, venture capitalists, and civil society to show solutions and shape what ‘inclusive, resilient’ AI should look like. The agenda–dominated by government ministries, industry associations, Big Tech, with little to no representation from human rights defenders or grassroots unions–is focused on building ‘AI for All’ and ‘AI for Public Good’.

It’s the first global AI summit hosted in the Global South, and that positioning matters: It creates space, at least in principle, for questions about equity that rarely surface in rooms – mostly in the Global North – dominated by those who build and sell AI technologies. Who profits, who bears the cost of this technology, whose problems it is designed to solve, whose lives it reorganises, whose knowledge and innovation are valued, and who defines ‘impact’.

The value of this event depends on whether the rooms make space for these questions.

Share BehanBox’s Substack

The Summit’s framework is organised around three ‘Sutras’ (People, Planet, and Progress) and seven ‘chakras’ that articulate India’s focus. Human capital, safety, social empowerment, innovation. The agenda is more ambitious: taking decades’ worth of AI discourse and cramming it within five days. But some questions don’t fit into these neat categories, and may even expose what or who the chakras leave out.

What follows is an attempt to surface and reframe those questions.

Chakra 1: “Human Capital”

The phrase tells us something. People are not workers with rights; they are capital to be optimised, inputs to be measured for productivity and efficiency.

The upskilling conversation assumes AI work is a future to prepare for. But an Indian AI workforce already exists; it just is not recognised as one. Educated women and young people in small towns label images, transcribe audio and moderate horrific content for pennies, training models that will be sold back to us as innovation. They are building AI, usually without any means to know whose systems their labour feeds or where the value goes. Theirs is invisible, gendered labour with no employment protections, no compensation for emotional and linguistic work, and no recourse when platforms change terms overnight. This is not “the future of work” because it is happening now. And it is colonial extraction of their cognitive labour: Global Majority countries like India provide the labour, the Global North captures the returns.

The upskilling conversation around AI also rarely asks whether the jobs on offer are worth skilling for. Will they pay a living wage? Are you told who you are working for, where your data sets will be used, is there a say in decision making? Will they offer contracts, leave, health coverage? Can you file a grievance against an algorithmic verdict? Can you collectively bargain with a platform that classifies you as an independent contractor?

Is it possible to create dignified, safe work conditions when algorithms are your managers and grievance officers, and you are not a person – but ‘capital’?

Chakra 2: “Safe and Trusted AI”

Yes, these systems must be transparent, accountable, and aligned with the public interest. But accountable to whom and through what mechanisms does the public actually shape those interests? The communities most affected by AI systems are rarely in the rooms—including those in and around the Bharat Mandap— where “responsible AI” frameworks are drafted.

Trust is not a technical problem. How do we build trust in AI when ‘digital trust’—trust and comfort with phones, payment apps, attendance apps, and technology at large—lags, especially among women whose access and usage of the internet is patchy and surveilled? Trust also requires understanding and dialogue, but algorithmic systems are designed to be opaque.

The harms are not hypothetical. Deepfakes as gendered violence, generative AI tools weaponised to spread anti-Muslim hate, surveillance and suppression of dissent – these demand oversight. But we also know how ‘safety’ and regulation become tools for censorship. Consider the new deepfake regulations coming into effect February 20. They claim to protect women from non-consensual synthetic imagery – a genuine harm that demands response. But the three-hour takedown window now applies to all content deemed unlawful, not just deepfakes. Platforms facing liability won’t distinguish between a synthetic video and a political post flagged under the same rules. When you build a system to remove content fast, the system removes content fast, regardless of what is being removed or why.

And when AI systems fail—deny wages, lock someone out of welfare, flag the wrong face, declare them dead—accountability becomes diffuse. Responsibility is split across vendors, agencies, the State and algorithmic black boxes, making meaningful appeal difficult.

So when we ask for AI accountability, we must ask: accountability within what system? Can you hold algorithms accountable when the states and corporations deploying them evade accountability themselves? Can you configure systems for transparency?

Chakra 3: “Inclusion for Social Empowerment”

Inclusion into what, exactly? The narrative assumes marginalised communities are simply left out of AI’s benefits, so the solution is diverse datasets, indigenous language models, representation in tech. The frame accepts the existing system as legitimate and asks only how to distribute access more widely. For every ‘can this AI solution scale’ or ‘is it ethical’, we forget to ask: why do we need this?

The discourse on ‘mitigating bias’ and exclusion assumes discrimination to be a technical problem, and debiasing is a patch to be applied. This week, the patch was literal: unsheltered people were cleared from Mathura Road before the summit began. Out of sight, out of dataset.

Caste, gender and religion do not appear in training data as variables to be corrected. They are embedded in what counts as authoritative text, whose surnames get associated with “doctor” and “professor”, which pin codes credit algorithms flag as high-risk, whose faces trigger police databases. Debiasing assumes the goal is a neutral AI. But an AI trained on existing datasets reproduces existing hierarchies.

The question, then, is not how to make these systems neutral. It is whether neutrality is even possible when the training material is historical exclusion.

Consider indigenous LLMs, often cited as inclusion success stories. There is genuine value here that should not be dismissed: documentation of endangered languages, accessibility tools for the majority of Indians who don’t speak English, educational resources and government services in regional languages. Language is identity, but also cognition and worldview; its preservation is not trivial.

But the questions remain: who funds these models, who owns them, who controls how they are used? Are they built to be scaled or for public good? If it’s venture capital seeking scale and returns, communities provide linguistic labour and cultural knowledge while investors capture value. If it’s the State, communities still don’t govern how their linguistic heritage is used, adapted or modernised. Then, the question isn’t whether indigenous LLMs have value – they do – but whether the current structures of AI development can deliver that value to communities, without extraction.

The deeper question is whether marginalised communities can say no. Can they refuse to be included on terms set by others? Can Adivasi communities refuse forest “monitoring” AI that serves conservation NGOs but criminalises their livelihoods? Inclusion without the power to refuse denies people economic self-determination and forces incorporation on terms set elsewhere.

Chakra 4: “AI for Economic Development & Social Good”

India is targeting up to $200 billion in data centers and chip manufacturing (as compared to the current $70 billion in investment), while villages lose water tables to cool servers. This is “development” – GDP growth, foreign direct investment, compute power, and global competitiveness. But whose development? Measured how? By whom? The discourse of AI for economic development treats technology as inherently progressive, benefits as naturally trickling down, and costs as acceptable externalities. BehanBox asks: Who bears those costs and who captures the gains? Where is the duty of care from the government and providers of social benefit?

NITI Aayog’s AI strategy frames India’s population as an asset, a “test bed” – for scale for training data, volume for testing deployment. Citizens as raw material for AI development, not participants in deciding what gets built for their good.

The “social good” framing similarly obscures who benefits. AI for healthcare often means diagnostic tools trained on data from patients who never consented. AI for agriculture means solutions designed for landowners – but most women working farms don’t own land, so extensions services never reach them. AI for financial inclusion means credit scoring that, at best, ignores (and, at worst, penalises) informal work, irregular income, lack of collateral – the economic pattern of most women’s lives, made more stark by caste oppression. What remains unclear is if we are fostering ‘AI for social good’ or fostering an AI industry.

Chakra 5: “Resilience, Innovation & Efficiency”

This is a false equivalence. Innovation and efficiency don’t always make systems or people resilient. Efficient systems often have no space to hold community knowledge. Innovation frequently strips away established rights and privacy. The framing collapses three different things into one aspirational package, but they pull in different directions.

Look closely at the three terms. To be caught in the language of “innovation” and “entrepreneurship” disguises what is lost, risked, or deliberately hidden from view. Who trains the algorithms that improve disaster alert systems? Who pays for the minerals and resources extracted to build them? Can these systems document women’s indigenous, oral knowledge without becoming spaces of extraction?

Efficiency for whom? Welfare “streamlined” means fewer humans to appeal to when an algorithm fails. MGNREGA made “efficient” through biometric authentication means workers whose fingerprints don’t scan don’t get paid. Efficiency measured by reduced “leakage” – but leakage only counts downward, never the massive expenditure on tech vendors framed as “investment”.

And efficiency is not a neutral value. AI systems like Lavender and Gospel, used to generate bombing targets in Gaza, processed intelligence at unprecedented speed, enabling thousands of targets in weeks where human analysts would produce dozens in a year. The same logic that makes welfare delivery “efficient” makes warfare efficient. The technology does not distinguish between benevolent and sinister use cases; the question is always efficiency toward what end, and at whose cost.

The “resilience” framing also shifts responsibility from systems preventing harm to communities absorbing shocks. Climate resilience through AI means better flood, drought, and early warning systems. But a warning is not a boat or a road out. The woman farmer who gets an SMS that floods are coming still has nowhere to go, no compensation when her harvest drowns. She is expected to be resilient, to adapt, to bounce back, to absorb. The system that created her vulnerability through deforestation upstream, absent drainage infrastructure and systemic landlessness stays unchanged.

Chakra 6: “Democratising AI Resources”

Policy discourse is stuck in a loop, equating democratising with localising compute power and technologies. Necessary, but is that all you need to democratically govern AI?

Sovereignty and “sovereign AI” have an undeniable appeal – national startups devoted to owning and governing our own compute, storage, and data flows. But does this redistribute power from Big Tech, or just shift it to domestic corporations and the State? No real effort has been made to build alternate, people-led AI governance models that give communities a say.

Indian states compete against each other—and against other Global South nations—offering cheaper land, power subsidies, tax holidays, looser regulations to attract data center investment. Where are the conversations on shared frameworks for worker protections, community consent, ecological limits? Is sovereignty a service, a design feature, or for social good?

None of this is democratisation if the conversation is confined to state and market options. Democratising AI would mean including people affected communities to shape what gets built, for whom, under what constraints—effectively rewriting the hierarchy of global AI.

Chakra 7: “Science”

Science is presented as neutral, with AI simply accelerating discovery in health, climate, agriculture. But the feminist question always is: whose research is getting funded, who bears the costs versus who captures benefits? And whose knowledge is valued?

Take AI for health. AI diagnostic tools developed through partnerships between big tech and private hospital chains are trained on patient data, often with consent waived or patients unaware. And systems trained on populations elsewhere may not work for Indian patients: IBM Watson for Oncology, deployed in Indian hospitals, was found to be biased toward US patients and standards of care.

AI for climate carries a contradiction at its core: data centers drain water tables and consume massive energy, so building compute infrastructure to model climate change has real environmental costs borne by communities who will not see the benefits. And forest monitoring AI—satellite imagery that could track deforestation, illegal mining, corporate violations—is predominantly deployed to surveil communities instead, flagging Adivasi movement as “encroachment” while corporate violations go untracked.

Agricultural AI is designed for landowners, trained on their data, optimised for their crops. Most women working farms do not own land, so extension services, credit, advisories built on this data do not reach them. And the climate adaptation knowledge women farmers hold is often oral, contextual, embodied. If AI cannot capture women farmers’ knowledge entirely – the problem becomes what counts as data, as science.

The pattern across all three: AI extracts data and knowledge from populations and communities, while the tools built from that extraction serve institutions, corporations, and states.

The larger question BehanBox would like to raise is: What would feminist governance of AI actually look like?

• Affected communities in decision-making rooms: Not as consultations after the fact, but in the design stage, with power to shape outcomes and say no.

• Accountability with recourse: When an algorithm denies you rations, wages, or your place on the voter roll, someone must be liable, there must be somewhere to appeal, a human who can override the machine.

• Decentralised governance of data: Communities as custodians of their own data and knowledge systems, not sources of extraction.

• Labour made visible: The women training algorithms, moderating content, collecting data recognised, compensated and protected as workers.

• Consent as ongoing, not one-time: Including the ability to revoke tools and systems that are not democratically agreed upon.

• And the willingness to ask “should we build this at all” before “how do we build this fairly” because existing structures are already unequal. A feminist governance of AI rejects techno-solutionism: some problems cannot be solved by automation, some knowledge should not be datafied, and some systems may need to be undone, not optimised.

  Thanks for reading BehanBox’s Substack! This post is public so feel free to share it.

  Share

This is part two our series on AI labour. Read part one here.

Muskan* sits in front of a mic in a make-shift studio carved out of her room in Govandi, an eastern suburb of Mumbai marked by dense settlements and waste-processing sites. It is midday: there is enough light in the room, and the children are away at school. A phone balanced on a stack of books, earphones plugged in, she records a script on medical information and transcribes interviews, which then become training data for a women’s health chatbot powered by AI.

Just a little over 40, Muskan taught at a school before she quit and took up this freelance gig with a local NGO that is building a women’s sexual and reproductive health chatbot in partnership with the Gates Foundation. First, she learnt to iterate small sentences – like “delivery didn’t happen” – and then she started playing around with emotions to reflect concern or urgency. Then the scripts became longer, about 45 minutes long on issues of pregnancy, fertility, family planning, which she would narrate with enunciations, and ideally, without background noises. She mumbled and fumbled at times, asked her children to practice with her, but kept going.

“Slowly the confidence came. Now it feels easy, really easy,” she says.

Without this emotional and linguistic labour–her voice, accent, pacing and emotional delivery–the final AI chatbot cannot reliably interact with users, who are the women from her slum. What will look like an autonomous function is actually sustained by her creative effort.

In part one of this series on the hidden architecture of intelligent machines and chatbots and platforms, we showed how the intelligence of AI and machine learning rests on an invisible workforce of people who collect, clean, and create data. Still, these continue to be seen as ‘technical’, ‘entry-level IT’ tasks, routine in their make-up, requiring little skill and thus low pay. In this, the second part, we turn to the lives of workers, whose tasks demand attention, context, judgment, and emotional endurance but whose skilled labour will be finally rendered replaceable and risky by opaque AI value chains. This manual labour is further devalued in India, where a history of tech outsourcing feeds the aspirations and economic mobility dreams of many, as we detail later.

The tech economy is reproducing colonial patterns of extraction and dependency, and the historic structures of caste and gendered hierarchies determine whose labour is seen with dignity, says Kai-Hsin Hung, a scholar of labour and value chains at the Interuniversity Research Centre on Globalisation and Work.

Scale of Harm

Rising Worker Vulnerability

Think of AI as a tiered global data economy, with action unfolding across geographies and channels. At the top are firms competing over chips and compute power; AI labs and product companies like OpenAI and Meta build and deploy models. At the bottom of this stack is the layer of human labour of workers like Muskan who keep the data pipeline going. We wrote about the racial logic governing this pyramid earlier: capital and decision-making power are concentrated in the hands of select countries while data and labour are sourced from a dispersed periphery in the Global South.

We found through researchers that contracts given to third party vendors and start-ups for this data work are squeezed on price and vendors compete aggressively on labour costs. This has the effect of lowering wages and creating work arrangements without accountability. The digital factory has no clear boundaries and no obvious site of accountability. Someone in Chittoor could be working at an outsourcing centre, employed by a Hyderabad start-up, to label data for a company in London.

In India, with declining formal employment opportunities, data work draws in young people and educated women into feminised, low-paid jobs for its promise of flexibility and remote work. At the same time, the rise of generative AI since 2022 has scaled the demand for invisible labour, while also often intensifying workers’ exposure to extreme, violent content.

The Union Government has partnered with global tech firms such as Microsoft and OpenAI under IndiaAI Mission to skill hundreds of thousands of developers and establish AI labs across training institutes, without any recognition of the risks and rights in this work. “The gap between the scale of harm and institutional response is widening,” says Kai-Hsin Hung. It leads to poorer work conditions, health risks, and jobs that erode workers’ skills over time.

Colonial Data Factories

Before the pandemic, transnational firms carried out AI work remotely but it has now expanded to include BPO and KPOs that work with varied structures, as we reported earlier. One regime is contractual work with no job security or grievance mechanism. In another, workers are formally employed by BPOs and ITeS firms but are more surveilled, and have stricter quotas tied to speed and accuracy. Work is also increasingly being governed by dashboards and faceless algorithms.

Share BehanBox’s Substack

As we said earlier, workers like Muskan take these work opportunities for want of options and the promised flexibility. “I gave many interviews–to be a receptionist, real estate agent, or tours and travels–but there was the age limit, demand for fluent English or unpaid training,” she says. She took up teaching but the commute and care work wore her down. “I used to wake up at four in the morning, make food for everyone, pack my tiffin, then take the train, walk five-ten minutes to share an auto, then take another auto to the Vikhroli station, then walk 10 minutes to school,” she recalls.

Her current AI gig may go on for three more months, or six, if she’s lucky. “I didn’t know this world before, but the work gives me peace and I hope it continues,” she says. Working from home now has changed the shape of Muskan’s day. She starts working after her children leave for school, then sometimes for three-four hours in the evening, weaving her paid data work around household chores. “This work feels much better…sitting at home and getting the same payment as other work.” Her husband is an auto driver, and with rising costs and their children’s education fees, the two still struggle but try to make ends meet.

When Gowri, 28, joined a vendor-based company near Chennai she saw it as a path to transition to a more ambitious software engineering role. Like her, many hope that it leads to a “real” tech job. The prestige associated with IT roles makes it easier for women in particular to negotiate with families who may otherwise frown upon women working–a reason why platforms like Amazon Mechanical Turk and Telus saw a surge of women joining during the pandemic.

Another reason is also the rapid spread of outsourcing centres in tier 2-3 towns since the pandemic. This displacement is happening against the backdrop of a larger structural shift where hiring in tier-2 and tier-3 cities has surged. Job openings in cities like Nagpur and Indore grew nearly 42%, far ahead of 19% growth in tier-1 metros, according to a 2025 Randstad report.

The pitch to local workers is simple and persuasive – global IT work is coming home, young people need not migrate for steady pay and a foothold in the tech economy. But women like Ashima*, in her early 20s, realise that these promises are illusory. She was planning to move from Chittoor to Chennai for a Master’s course, when a data outcentre company opened its first office in her town. There is work here now, why do you need to move, her family had argued. But it is a decision she came to regret, as we detail later.

“In many cases, it came in the way of women seeking that independence and moving away from home,” says Srravya Chandhiramowuli, a researcher of data annotation work for AI at the University of Edinburgh.

(De)Valued

Broken Education-Employment Link

The Union Government is encouraging young Indians to embrace machine learning skills. Data annotation and model training roles are touted to absorb a large portion of India’s workforce impacted by automation, and by 2030, the market is expected to exceed $7 billion. The government think-tank NITI Aayog believes tasks that workers like Gowri and Muskaan undertake require “low levels of expertise and present an opportunity to exploit labour cost arbitrage to serve companies globally”.

But this assessment is not necessarily correct, interviews show. Hyderabad resident Deepa* spent years annotating geospatial data – tracing roads, identifying signs, correcting maps to the point she can navigate parts of London by memory and at speed. AI trainers too have spoken about how crucial human oversight is.

Srravya’s research too showed that start-up companies hide the labour sustaining them. “You don’t want the data annotation team to also grow as you grow as a company. You have to show that you’re scaling efficiently, so you displace that work and move it out of sight,” one manager told her.

What’s damning is that women’s participation in this sector tracks closely with gender norms around cleaning and care, creating parallels to traditional caregiving, says Kai-hsin. Across iMerit, Accenture, Telus, Teleperformance, and other work arrangements, women are often employed in low-level annotation teams, quality assurance, or content moderation. These in practice become “digital cleaning” jobs with workers scrubbing datasets or tidying labels.

Subscribe now

This also has consequences for the future of women’s labour participation in India. “For the last 50 years we’ve been pushing to educate women through initiatives like Beti Bachao, Beti Padhao… but the link between education to dignified employment is broken,” says Aditi Surie, a scholar at the Indian Institute of Human Settlements.

It’s not that this work is meant to be essentially numbing but it has been made so because it requires little engagement and is done without transparency, we found. This leaves no room for meaning or creativity in the labour.

‘No Future In This Role’

Tasks involved in data work are repetitive and decomposed, designed to strip away human judgment, says Kai-Hsin.

That was Gowri’s experience too. She was annotating real-time business insurance data for a company in Chennai, but left after 18 months because there was “no future in this role”. Most days she felt like “just a label marker”, doing “more work than was humanly possible”, and often without the training to handle complex data. She applied to 20 roles over eight months and did not bag any of them.

In Hyderabad, Beena* made a quieter calculation about exiting the system. After two years of working with a data annotation company, she asked for a work experience certificate for only one year. She worried that without any promotion or role change, her stint would look like stagnation. “I don’t think I’m really helping myself by saying this is what I did for two years,” she told researcher Srravya.

In India, studies of Adivasi women working with iMerit found they were channeled into the lowest-paid and monotonous segments. Their knowledge disappeared into the model without recognition or authorship, researchers found.

One worker spent time trying to decipher the difference between the red cap of a Dr Pepper bottle versus that of a Coke. “It should not be anybody’s job to know the difference between these kinds of details, and yet, that’s what they’re trying to learn,” Srravya adds.

High productivity targets and volume of work also strip judgment and learning out of content moderation tasks. Content moderators in India with companies like Accenture and L&T–working without proper training and mental health support–prioritise quick content work (like removing posts or live streams) over those that require complex thinking.

“Would never recommend de-ranking content as it would take time,” a 28-year-old told a researcher. This additional time would interfere with performance evaluation. This is textbook deskilling, Kai-Hsin says, where complex judgment work is broken into point-and-click operations governed by speed-based metrics, quotas, and dashboards. And without regulation or collective bargaining, these work opportunities lock many into a cul-de-sac.

Research by scholar Nithhya Sambasivan and others showed that qualities such as care, diligence, and contextual reasoning–central to good data work–are systematically undervalued, even though they directly shape the quality and safety of AI outputs.

Share

There are exceptions, such as in Muskan’s case, where there is an attempt to engage with ‘ethical AI’. While recording and transcribing notes, she learns about women’s sexual and reproductive health. She has also begun picking new words and pronunciations and learned how to zoom in and steady the screen. She shared this with other women. “Talking to each other helps, you get solutions,” she says.

Data work needs development skills, as is the case with Muskan, which are undervalued and misidentified—similar to paid care work like that of Accredited Social Health Activists (ASHA) and domestic workers. Workers are the experts on the data sets they produce and train, so their expertise needs to be valued better and recognised, a process that researcher Srravya thinks can begin with expanding their work beyond the scope of a task.

What if in Humans In the Loop, the company recognised protagonist Nehma’s creativity, treated her as an equal, and encouraged her to upload and tag datasets that reflect the culture and contexts of the Adivasi community? Or what if content moderation for social media giants was organised the way it is for Wikipedia, with guidelines and transparency protocols in place, that incentive accuracy and judgment? “Working with people as equals can make this work far more integrated and meaningful,” says Srravya.

Caring For Machines

The costs of making AI machines possible often shift to workers’ bodies and household budgets, we found, but neither the government nor employers recognise this. Muskan, for instance, complains of eye strain and back pain due to long hours of sitting and staring at a phone screen. She paid for a doctor’s consultation out of her pocket. She also pays for the invisible infrastructure for her work. “We don’t have Wi-Fi [because it adds to monthly expenses]; everything runs on mobile data,” she says. So the couple decided to wait and see if her gig lasts longer than six months.

In outsourcing centres with strict targets, where days unfold as doing 120 tasks per hour engaged in numbing work with strict bathroom breaks, workers speak of fatigue and exhaustion. “Workers may not articulate it as mental well being concerns but these issues are distressing and have impacts on workers’ mental and physical health,” says Srravya.

Similar patterns have been documented in Kenya, Philippines, and Colombia. Workers pay for internet connectivity, ergonomic chairs, therapy–especially content moderators who are exposed to violent content. A Kenyan court in 2023 ordered Meta to provide “proper medical, psychiatric, and psychological care” to content moderators who spoke of depression, anxiety, post-traumatic stress disorder, and suicidal ideations as outcomes of filtering pictures and videos of assault, self-harm, child abuse, necrophilia, rape, and more.

Subscribe now

Under the recently enforced labour codes, “gig workers” and “platform workers” are defined only in the Code on Social Security, 2020. But they do not automatically fall within the scope of core protections in the Occupational Safety, Health and Working Conditions Code because they are outside the traditional employer-employee framework. And because they are classified as not “employees” but “independent contractors”, provisions ensuring minimum wages, maternity benefits, and occupational safety are absent unless the government implements specific schemes.

Asmita Bhutani, an assistant professor of labour at York University, Toronto, says despite this work being intensive to the body and mind, “no employer or government pays attention to women’s embodied labour or issues related to health and safety”.

Kenya’s AI workers formed a Data Labelers Association last year to ask for mental health support, fair treatment, and upskilling opportunities. Similar demands echo among Indian workers too but more than everything, they want to be treated as employees, says Asmita: “They are not surfing on their phones doing low-skill work; they are spending time, energy, and cognitive skills…they want to be treated not as contractors but as employees.”

Past Continuous

India’s Tech Mythification

Unlike India, in Kenya and the Philippines, activism and collective action since the pandemic have led to sustained scrutiny and legislative action. One reason is the visibility of the work itself: Samasource (a contractor for Meta) ran large physical content moderation centres in Nairobi where people met each other and shared work conditions, as compared to Ind
ia where most content moderation work is dispersed, routed through multiple actors, hybrid offices, and remote platforms, as we explained in an earlier report.

Secondly, tech work in India has been framed as meritocratic and aspirational, rooted in prestige and independence. This, like scholar Hemangini Gupta wrote in her book Experimental Times, contributes to a “nationalistic mythification that valourised tech products over the process and labour that goes into”.

When India liberalised its economy, the IT sector and BPOs were seen as these silver bullets to solve employment problems in India and in turn got regulatory leniency. They enjoyed labour protections, such as exceptions from the Industrial Employment (Standing Orders) Act, which happened most recently in Karnataka which extended the exemption running up to 2029. These exemptions weaken standards on termination, working time, and grievance processes–what experts call “regulatory arbitrage”. Rising automation and AI layoffs have led to burnouts, insecurity, and a mental health crisis in India’s IT sector.

Data work is being absorbed into the same architecture. Recent protests tend to focus on forced resignations or retrenchments but sustained advocacy remains difficult. “IT workers are a classical example of people who believe their lies,” says Rikta Krishnaswamy who has organised tech workers earlier. There is another work myth involved here: how can workers with phones and laptops, sheltered in offices or homes, be exploited the same as those doing manual work on the streets, in sweat shops and factories?

Identifying where large-scale data labour is located is itself a challenge, says Alagunambi, an organiser with UNITE, a union for IT and ITeS workers in Chennai whose members also include those data tagging work. Firms like Teleperformance could be key sites, but the work is constantly being subcontracted, reclassified, and relocated.

India’s masculinised idea of tech also distances women. Asmita says: “That’s one contradiction — women are a part of the tech economy, they’re instrumental in building up AI systems and contributing to profit for the tech ecosystem, but they’re not really seeing themselves as part of that system, nor are the companies pitching them as such.”

Workers do find ways to outmaneuver the system: some finish, according to Srravya, hourly targets just a little ahead of deadline and use the spare few minutes to take “micro breaks”. Others might develop a personal equation with their manager and seek leeway. These acts of resistance become less possible the further you are from the centre where these systems are built, Srravya adds.

Regulatory Challenges

The recent legislative wins focus on location-based gig workers and make no mention of data workers or content moderators. Governments continue to see data workers as an extension of the ITeS and BPO industry, or as ‘future ready’ labour, experts say.

India’s national AI strategy and the IndiaAI Mission focus on innovation, compute capacity, datasets platforms, startup financing. They frame ‘responsible AI’ in terms of limiting “technical harm caused by AI applications”, while leaving workplace conditions for people building or training those systems outside the frame.

The forthcoming AI Impact Summit has sessions on “skilling and productivity” of the future labour force, but not an explicit stream dedicated to worker rights or labour conditions for data workers, shows the official agendas released so far. There are talks on compute capacity, datasets platforms, and startup financing, but conversations on labour protections remain in the shadows.

The new Digital Personal Data Protection (DPDP) Act also only governs data governments and user rights, and not the workers around it or their work conditions.

It is unclear if representatives for these groups were included in the consultations for these bills, says Nighat, a policy analyst with Aapti Institute. “These issues are passed on between ministries of labour, technology and trade, so it’s hard to pin accountability,” she adds.

Kai-Hsin sums up the paradox. Despite frequent rhetoric about ‘responsible AI’ or ‘AI for job’, no regulatory energy is going into protecting upstream data workers. “There is no coherent regime that treats AI data work as hazardous, skilled labour deserving strong rights,” he says.

Further reading:

• Aapti Institute has a voluntary code of conduct for companies employing data workers.

• Modern tech economy and AI cannot work without millions of workers training algorithms and labelling data. Kenyan activist Joan Kinyua talks about transparency, basic rights, and fair pay for data workers, and how workers are already pushing back. Read this AI Now Institute interview.

• A documentary about the about human workers behind AI in Nairobi, and the working conditions that push their physical, economic, and emotional well being to the limits.

  ---

Leave a comment

On December 18, 2025, the Sustainable Harnessing and Advancement of Nuclear Energy for Transforming India (SHANTI) Act was passed in the Parliament without any public consultation, outside the ambit of RTIs, and amid a walkout by the Opposition. The Act repeals and subsumes two laws – the Atomic Energy Act, 1962, and the Civil Liability for Nuclear Damage Act (CLNDA), 2010 – opening up nuclear energy to the private sector, diluting accountability, and raising concerns for public safety. Companies like Adani, Jindal, Reliance, Tata, and Vedanta have already shown interest in nuclear energy and the Adani Group has begun exploring a 1,600 MW project in Uttar Pradesh.

The Central Government claims that the Act was overdue and plans to install 100GW of nuclear energy by 2047 to contribute to its climate and energy transition goals. Yet, nuclear remains a ‘false climate solution’ (with high carbon emissions, long operational timelines, and poor cost-to-energy generation ratios) and poses significant environmental, health and weapons proliferation risks.

Often built on coastlines and in high-seismic zones with a poor track record of management, radioactive contamination poses a greater risk to those already marginalised by caste, religion and gender during extreme weather events, says Misria Shaik Ali, PhD from Rensselaer Polytechnic Institute and currently a postdoctoral fellow. In this interview with Anjali D, Misria draws on her decade of research to trace opaque bureaucratic systems, debunk claims of nuclear techno-utopias, and discusses the way forward. Excerpts below.

What is the new SHANTI Act and why has it alarmed civil society and scientists?

“SHANTI” is a strategically chosen acronym meant to evoke peace and development. But I want us to return to this word at the end of the interview and consider whether the Act truly brings shanti (peace), and for whom.

The Act comes at a moment when nuclear energy is being promoted as environmentally safe and climate-responsive. However, my work over the past decade shows that nuclear energy is far from safe. Jadugoda in Jharkhand, also called the nuclear graveyard of India, is a stark example where uranium mining and radioactive contamination have resulted in widespread disabilities and diseases, worsened by lack of government support.

Nuclear reactors involve radioactive materials with severe biological risks, whether through contamination or warfare. Placing such an unstable energy source in the hands of profit-driven companies raises serious concerns about accountability and public welfare. Beyond privatisation, the Act also fails to address longstanding regulatory gaps and dilutes provisions for accountability and liability. It continues to treat nuclear energy as the domain of experts alone, ignoring the fact that it directly affects the public and requires broader democratic oversight.

Under the SHANTI Act, nuclear power and human-made radioactive contamination will become ubiquitous, without sufficient regulatory safeguards or jurisdictional clarity.

Why did you choose to study nuclear contamination in India?

I was first introduced to nuclear issues in 2011 during the Idinthakari protests against the Kudankulam nuclear plant. Growing up in Tamil Nadu, I would sometimes hear about the protests on TV while walking past my grandparents’ room. Back then, I would say I opposed the protests, because figures like APJ Abdul Kalam taught me that India needed nuclear energy.

It was during my Master’s research on the Neutrino Observatory in Theni district that I realised this view came from not knowing enough to form my own opinion. I was just repeating popular arguments: that nuclear energy is safe, necessary for rising energy demand, and allows continued economic growth without compromise.

During my studies, I noticed that the neutrino facility was being established at the cost of sacred mountains, and why local communities resisted this abstract idea of “scientific advancement”. Even if they didn’t know what neutrinos were, they knew they didn’t want their sacred lands destroyed. Over time, I realised how narratives of nuclear progress often weaponise expertise, suggesting that only experts understand these projects and their national importance while dismissing ordinary people as unknowledgeable. This pushed me to study nuclear contamination more seriously and form my own opinion.

You worked extensively in contamination affected areas like Tummalapalle in Andhra Pradesh, and Kudankulam. What is radioactive contamination, and what can it look like on the ground?

Unlike other chemicals, radioactive contamination has no colour or odour. Simply put, it is released when unstable elements like uranium and thorium are disturbed. This happens throughout the nuclear fuel cycle, from mining to waste disposal (both of which remain unclear and poorly defined under the SHANTI Act). Once disturbed, these invisible particles seep into water, air, and food, and enter our bodies, where they remain and continue emitting radiation, causing cellular and genetic damage.

This invisibility makes radioactive contamination especially dangerous. People only realise they have been exposed long after it has entered their bodies and through small signs like skin discolouration or irregular menstrual cycles. Over time and across generations, this can manifest as spontaneous abortions, children born with disabilities, and chronic diseases.

This pattern was seen around the uranium mine in Jadugoda. Although the mine began operating in 1967, it was only two decades later that people began noticing that children were born with more/less than five fingers or two limbs and people experienced severe health issues, including heart disorders and reproductive failures.

Similarly, in 2016, near the Tummalapalle uranium mine’s tailings (waste dumping) pond, farmers noticed white-coloured water in their borewells with diminished yields. Months later, testing revealed uranium levels as high as 4,000 ppb. While no amount of uranium is truly safe, the WHO considers 30 ppb acceptable, whereas India’s standard is 60 ppb.

Often, movies and images depict radiation as neon green, but in reality, it looks banal – like wind-blown dust settling as pink residue on your skin. And that is what makes it frightening, because people cannot see it, cannot access monitoring data, or make contamination legible in their everyday lives.

Just as the invisible LPG was made detectable by adding the odourant Ethyl Mercaptan, making radioactive contamination ‘visible’ should be a priority. Instead, the nuclear sector continues to prioritise profit over people’s wellbeing.

Does contamination have specific gendered or caste-based everyday impacts?

Radioactive contamination and disability are deeply linked and complicated by caste and access to healthcare. In Tummalapalle, I found fewer children with visible disabilities than in Jadugoda because families here have greater geographic and financial access to medical technologies like CT scans. As a result, doctors often advise parents to abort disabled foetuses, eugenically erasing an entire group of people from the world. Jadugoda, a predominantly tribal area with limited access to medical facilities, has a much higher visibility of children living with disabilities.

Contamination also has a distinctly gendered impact. Women in Tummalapalle are disproportionately affected because the water they use every day for cooking, washing and cleaning is contaminated by a nearby tailings pond that receives over 73,000 tonnes of waste each month. Although Pollution Control Board (PCB) stipulations required the tailing ponds to be lined with multiple non-porous materials, the Uranium Corporation of India (UCIL) followed simpler Atomic Energy Regulatory Board (AERB) guidelines, lining it only with clay.

As a result, many women have developed skin disorders, endometriosis, kidney issues, irregular menstrual cycles, and experience high rates of spontaneous abortions. My 2021 surveys found that 44% of pregnancies in Tummalapalle ended in miscarriage, compared to a national average of 30%.

During one of my focus group discussions, women kept recalling all the abortions they had experienced – so many, that until prompted they only counted abortions as those conducted by doctors. Many women have also undergone hysterectomies, and several families have sent their children to live outside the village to avoid contamination.

It is impossible to describe the enormity of the grief and pain they carry in their bodies.

Furthermore, while permanent employees are protected by externally monitored radiation badges, contract workers are routinely sent into radiation zones in their place without safety gear or the monitoring badges so that companies can avoid accountability. I myself have seen photographs of workers standing in radioactive dust with exposed legs and no protection. Yet, most don’t have access to health insurance or adequate medical facilities.

Ultimately, nuclear contamination deepens existing social hierarchies. Women and oppressed caste workers bear the greatest risks, facing new and intensified forms of exploitation – risks that are likely to worsen with privatisation and weaker liability clauses under the SHANTI Act. Any discussion of nuclear energy must address social costs in a society already shaped by caste, gender, and religious inequalities.

You’ve spoken about the impacts of contamination on agriculture, water, labour, climate, health and more. Given this, where has the management and monitoring of nuclear energy and contamination fallen short?

Regulation of nuclear energy in India is deeply fragmented, and Tummalapalle illustrates this failure clearly. Let’s revisit the uranium mine’s tailings pond, where the Pollution Control Board (PCB) and the Atomic Energy Regulatory Board (AERB) had conflicting standards.

When the Andhra Pradesh PCB found groundwater contamination through its assessment, it issued a show-cause notice to UCIL. A public sector undertaking of the Department of Atomic Energy, UCIL responded that it had followed AERB guidelines and was therefore not required to comply with PCB stipulations. UCIL even shifted the blame for the contamination onto farmers, claiming they dug borewells “too deep.”

The matter resurfaced in 2021 before the National Green Tribunal (NGT). However, local communities were never informed. The NGT eventually ruled that since UCIL had followed AERB guidelines, it could not adjudicate the issue. In 2023, the PCB went a step further, stating that because radioactive risks fell outside its purview, the site should be removed from its contaminated sites inventory and transferred to the Department of Atomic Energy.

What this case reveals is extremely alarming: an issue of pollution was handed back to the same authority responsible for running the facility, and once again, non-nuclear regulators were sidelined while the idea of “expert” was narrowly defined to exclude farmers as experts of their own environments.

The SHANTI Act worsens this trend. While it proposes a Nuclear Damage Claims Commission, it excludes representation from affected communities. It also grants statutory independence to the AERB, without clarifying how it will coordinate with other regulators like PCBs. Furthermore, several regulatory aspects continue to be allocated to the AERB or the central government, making true political independence impossible.

There are also serious unresolved gaps in radiation safety. In 2010, a radioactive research device from Delhi University ended up in Delhi’s Mayapuri scrap metal market, causing a Level-4 radiation incident. One worker died and several others suffered severe burns. Such “orphaned”, or mismanaged, radioactive materials pose enormous risks, including accidental exposure, environmental contamination, and even weaponisation. Yet the AERB still does n

Share

ot have a comprehensive list of radioactive sources in use across India.

Despite these failures, the SHANTI Act proposes privatising nuclear energy and introducing Small Modular Reactors in industrial settings with minimal exclusion zones. These reactors are to be placed inside the factory campuses and data centres employing thousands of workers, while day-to-day controls will practically shift to private players. If radioactive materials are lost, mishandled or leaked, who will be responsible?

These are the rhetorics parroted by the establishment to make nuclear sound like a techno-utopia. Unless we have the humility to acknowledge our own errors, we will continue living in a contaminated world.

Given nuclear energy is classified as an issue of National Security, what do systems of consent, accountability and redressal look like for contamination affected communities?

To be honest, there is no real consent, accountability, or redressal in nuclear project areas. There is no space for agency. After being stripped of their lands, livelihoods, and safety nets, people are involuntarily pushed into a radioactive, contaminated world.

Again, caste plays a big role. When landless agricultural labourers lose their jobs after land is acquired for nuclear projects, they are left with little choice but to work in the nuclear mines or plants as contract labourers. Even those who do have some choice are rarely informed about the risks by the nuclear establishment.

Audio : Consent, choice and the illusion of a future in radioactive zones

When I was in Kottala village, Tummalapalle, people told me that they didn’t believe they would ever be healthy again. Everyone, cattle, dogs, and people, had diseases, and the land was completely contaminated. There were no functioning health facilities and no stable income. Abandoned by the state, they now demand that UCIL acquire their entire village, believing that such an acquisition will give them access to compensation and health insurance.

People have been structurally excluded from the very beginning, when the Atomic Energy Act of 1962 explicitly kept the public out of nuclear decision-making. The SHANTI Act, intended to harness nuclear energy through private players at a time when uranium and thorium mining are exempted from public hearings altogether, deepens the exclusion of the public.

This creates a climate for protest and consequent criminalisation. In Kudankulam, around 8,800 people were charged with sedition for opposing the nuclear plant, many of whom still carry those cases. They were accused of receiving foreign funding, when they were just demanding protection for their health, livelihoods, and oceans.

Share BehanBox’s Substack

On liability, there has been a clear rollback. Under the “right to recourse” clause of the Civil Liability for Nuclear Damage Act, 2010, suppliers could be held responsible in cases of accidents. Under the SHANTI Act, this liability is shifted away from suppliers towards operators to attract foreign investment. Yet, even operator liability is being evaded – by labelling potential operators as “users”, operators are released from the responsibility of compensation altogether. In practice, this means that private companies will be let-off scott-free while compensation is pushed back onto the government and delayed indefinitely.

How has India historically justified the proliferation of nuclear energy, and how has the narrative around nuclear energy shifted today?

Historically, India’s nuclear lobby has always attached itself to whatever policy language is most fashionable at the time. In the 1950s, nuclear energy was framed as a symbol of post-colonial strength and self-reliance. By 2006, nuclear energy was rebranded as a safe, environmental solution that could be reused.

Today, nuclear energy is being sold as a carbon-neutral climate solution and a pathway to net zero. By framing nuclear as a climate solution, the state is trying to project India as a modern, progressive nation, while distracting from deeper social and political violence like the daily lynching of minorities.

The benefits of nuclear power, whether employment or scientific research, have always been highly concentrated in the hands of the upper caste. Nuclear energy has always served only a Brahmanical idea of ‘the nation’.

What should a good nuclear policy look like?

I’ve been studying global nuclear regulatory systems, and honestly, I haven’t come across a single ‘good’ model till date. Many people hoped Fukushima would offer lessons for a path forward, but in the end, they too just flushed out millions of gallons of treated radioactive water with tritium into the sea.

After every disaster, we get a new round of safety updates – Chernobyl, Three Mile Island, Fukushima. Despite prediction models, the complexity of nuclear disasters makes planning difficult. Just because a plumber designs a drainage system for safety, doesn’t mean there will be no leaks. And imagine that when there is a leak, the plumber rejects it despite water damage on the wall, just because he has designed it for safety.

That is exactly how nuclear safety works. When people say their safety has been compromised, ‘experts’ deny it, insisting that all the design has been tested for safety and all safety procedures have been followed. The possibility of failure is dismissed outright. And although many experts genuinely try to prevent accidents, this isn’t about intentions. Systems built only around technical safety can never account for human error or social realities.

Safety is a very limited framework, especially in societies shaped by caste and patriarchy. The real question should be whether humans and non-humans are protected from exploitation and contamination, and technical safety is only one small part of that.

It’s just impossible to be responsible with nuclear energy. It is undemocratic, irresponsible and keeps the public out. For me, the only genuinely responsible nuclear state would be one that does not pursue nuclear energy.

How do we reckon with energy security for the country amidst climate change?

There are different ways to think about energy security, but right now we’re following what I’d call a substitution approach. We’re simply replacing fossil fuels with other sources like hydro, solar, and nuclear without questioning the system itself.

For me, neither coal nor nuclear are real options and large-scale electricity itself needs to be questioned. We should be moving toward decentralised energy systems that build on existing infrastructure, rather than creating massive solar parks or nuclear plants in the name of “clean” energy.

We need an urgent shift toward degrowth: producing and consuming less. This applies both at an individual level and to large-scale consumption by crony capitalists. We also need to redirect funding away from outdated debates and toward research on radioactive contamination, its impacts on agriculture, women’s bodies, and structural inequality.

I’m against all systems that benefit a small elite while the majority bears the cost.

As a Muslim woman in the current political climate, what do you risk while documenting nuclear contamination?

Audio : Listen to unedited parts of this section by Misria

After I submitted my PhD dissertation in 2023, I found it being discussed on a defence forum under the heading “New Anti-National Activity.” My work was linked there, and one comment said something along the lines of, “Did you see the author’s name? Of course she’s anti-national.” At that moment, my Muslim identity was attacked to discredit ten years of research.

Being a woman in this field makes it harder as we are not considered smart enough to work on nuclear issues. And since I’m not a nuclear engineer, my work is often dismissed altogether.

I’ve also been warned by senior scholars not to study nuclear energy, especially because I write about the marginalisation and genocide of Indian Muslims. After I leave the field, people I work with are often questioned by authorities. So the risk isn’t just mine – it extends to all those whose lives are completely tarnished by nuclear energy and still choose to share their experiences with me. This is a reality that many activists face, simply because they cannot let go of the horrors they have witnessed.

Subscribe now

What keeps me going is the blind trust people place in me. I am holding so many people’s vulnerabilities and stories – women whose bodies have suffered immensely just to give birth, and entire villages living in constant anxiety about displacement and contamination. When they have shared their pain and paranoia, how can I even think about protecting myself as a scholar?

Walking away isn’t an option. Even if this work makes me more vulnerable and less employable, the risks I face are small compared to what these communities endure every day. I will continue this work in whatever form I can, and in the future I want to be able to take those at the frontlines of radioactive contamination with me to conferences because their voices are the ones that matter most.

How should we keep ourselves better informed about nuclear energy?

Start by thinking about your own everyday exposure to radiation through medical systems, food systems, and technologies we rarely question. Even those of us with class privilege don’t know about the impacts of X-rays or radioactive dyes used in scans.

You can also read MV Ramana’s The Power of Promise, Shannon Cram’s Unmaking the Bomb, and Thomas Wellock’s Safe Enough?. Additionally, read the SHANTI Act carefully. If nuclear reactors are going to become ubiquitous, we need to know what rights and avenues of recourse we have.

And finally, I want you to reflect on my initial question – Do you now feel that this act brings shanti? For who?

This story is produced as part of the Asian College of Journalism’s Climate Change Media Hub Mentorship Program.

Leave a comment

“How will I survive without NREGA?” asked Kamla Devi, a widow from Beawar, Rajasthan, at a press conference in Delhi on December 17. Widowed some years ago and with no land or resources in her name, she has relied on work as part of the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) to provide her income.

There are 12.16 crore workers like Kamla Devi whose livelihood stands threatened with a new Bill that stands to repeal the historic MNNREGA. On December 15, the Union Government tabled the Viksit Bharat Guarantee for Rozgar and Ajeevika Mission (Gramin), or VB–GRAMG, Bill, 2025, which was passed by the Lok Sabha on December 18 through a voice vote and by the Rajya Sabha post midnight the same day, amid Opposition protests. The Bill will come into law once signed by the President of India.

Among other provisions, the Bill increases the number of promised days of employment from 100 to 125. The Centre would bear 60% of the costs and give states a “notional allocation” based on undefined criterion; any expenditure beyond this allocation will be 100% the state’s responsibility. With a pre-determined budget and discretionary powers to the Centre, the Bill violates fundamental principles of a rights-based legislation and dismantles the right to work, unions say.

The Bill also centralises power with the Centre while shifting the financial and implementation burden on to the States, says Dipa Sinha, an economist and an assistant professor at the School of Liberal Studies, Ambedkar University, Delhi. “There are many issues on ground but many would have been resolved if the initial MGNREGS itself was implemented in the spirit of the Act,” she points out.

The legislation came on the heels of India’s long history of public spending on rural development and poverty reduction programmes, such as the Food For Work initiative. In 2005, the ruling United Progressive Alliance government passed MGNREGA amid high unemployment, inflation, and resistance by worker unions. In promising 100 days of guaranteed unskilled employment in a year to the rural poor, it legalised the right to work. It remains the world’s largest employment guarantee programme, and at the time of its inception was believed to be a “game changer” and the “single most innovative programme from India and a lesson to the whole world”.

However over the last few years, MGNREGS has been hit by budgetary constraints, delayed and inadequate wages, irregular work flow, and issues of digital compliance in the last few years but it still offered social security benefits, encouraged local land development, and bolstered the bargaining power of workers.

In an interview with Saumya Kalia, Dipa talks about the proposed reforms and how they dilute the right to work. Edited excerpts below:

Does any aspect of how the recent Bill was introduced – the name, timing, core provisions – strike you as significant or surprising?

In terms of the content, some parts of the Bill codify what the Union Government has been doing stealthily for some time. It fits in the larger narrative of how MGNREGA has been undermined over the years, so that is not entirely surprising. What is surprising is the manner in which it was introduced. Even with the labour codes, they put up drafts and asked for some responses — although they didn’t take anything into consideration — but at least there was a pretense of some process. This Bill did not even see the amount of public consultation the labour codes had; it was just abruptly listed on Monday.

The way it’s being done is a bit puzzling. They want to repeal a rights-based law without due consultation with workers, unions, or activists.

Does the new Bill address any of the limitations of MNREGS in terms of budget constraints, wage delays, or work demand-supply gap?

It’s been 20 years since the Act was passed and times have changed, so there is a need to rethink it for sure. The Amarjeet Sinha committee was constituted in 2021 to review the implementation. But I don’t see the need to repeal the whole Act and replace it with new legislation.

The many issues on the ground could have been resolved if the MGNREGS was implemented in the spirit of the Act. Many difficulties were created in the name of reforms. A major set of these relate to the digitisation of processes in the last few years – insisting on digital attendance records, and real time monitoring through the National Mobile Monitoring System app which have caused many issues on the ground. Now these digital solutions touted as reforms are being made into laws. (Research shows that digital fixes have not brought 100% success. Reetika Khera argues that technological solutions have reduced transparency instead of enhancing it, resulting in corrupt practices in states that could earlier control malpractices.)

Also, the tendency towards overcentralisation began to happen in different ways earlier. Under MNNREGS, the Centre makes an initial allocation at the beginning of the year, states implement the programme, and revised budgets are sought based on actual demand. Earlier this year, the government brought in a notification placing caps on how much states could ask for. The government circumvented questions on this, maintaining that MGNREGS is a demand-based scheme and they will “make funds available as per demand on ground”. Our argument even then was if the Centre signals a cap in advance, states will inevitably underestimate demand during implementation because they are unsure whether funds will be released.

Share BehanBox’s Substack

The third concern had to do with wage revision. NREGA has a separate wage rate not linked to notified minimum wages, and its updation each year has been so meager that the current NREGA wages are quite low compared to even the prevailing market wages. The demand for NREGA exists on the ground, and the work being provided currently is inadequate. Also its broader effects – on the labour market for instance – are getting diluted. The Bill is not meeting or addressing any of these concerns.

Activists argue that the new Bill centralises power by giving the Centre discretion over state-wise allocations, while also shifting financial responsibility to the states. How does this change the nature of the job guarantee?

In this new legislation, the Central Government will use some parameters – we don’t know what – and in the beginning of the year make a ‘normative allocation’ for each State. The Bill still uses the word ‘demand’, but what happens if the demand is more than the normative allocation in a state? It also very clearly states that for any expenditure above this normative allocation, states will have to bear the entire additional expenditure themselves, and even this has to be approved by the Centre. So you’re shifting the entire liability to the state government which has to operate within centrally determined norms.

And all these benefits of floor wages, restricting migration, and more are realised if work is available when people need it. So it’s also about shifting power within the labour market, giving the worker a say over where they’ll work and for how much, because currently the power is tilted towards the employer.

Estimates show that states might be expected to shell out almost Rs 55,000 crore. Do they have the fiscal capacity for this and what will this burden do to the delivery of job guarantee?

States are already struggling; we’ve been seeing state chief ministers coming up to Delhi, holding dharna and lodging complaints. It’s not only NREGA, so many things have been shifted to the states. There are delays in GST transfers which are being contested. Over time, GST has also hit the revenue raising capacity of states. At the same time, the Centre has shifted its revenue composition between cess and tax revenue in a manner where cess is becoming higher – this matters because cess is not shareable with states.

So you’re squeezing state revenues but asking them to spend more on development schemes while curtailing their freedom to spend. Take the National Rural Health Mission for example: if the signage over a Health and Wellness Center (Aarogya Mandira) is not per the Centre’s design, they wouldn’t give you the money.

Why is decentralisation important to a job guarantee legislation?

The whole point of NREGA was also to empower the community. NREGA said that you would have a shelf of works which would be decided in the Gram Sabha which would decide if there is a need to construct a road, clean the talaab, dig up the school playground, and so on. And then as and when people demand work, they would look at that shelf of works and pick. The new Bill still empowers the Gram Panchayat to make a Gram Panchayat Development Plan but the main objective and core priorities will be defined by the Centre. And once these plans are made, they will all become part of the Viksit Bharat National Rural Infrastructure Stack [a unified framework under the new Bill for listing public works], from which the centre will pick its priorities. If you have a decentralised mechanism, why insist on a National Rural Infrastructure Stack?

The country is very diverse, and these things should be left to be planned locally. Only when there is local accountability and people have ownership in the scheme can these problems of leakages and so on be addressed. In that sense this Bill is similar to the earlier Food for Work programme, and other such schemes, when the government opened up work opportunities when it felt like it.

The Bill turns a legal right into a centrally sponsored scheme. What would be the impact on the broader landscape of employment security in rural India, especially the landless labourers and women who depend on it? Could there be further casualisation of labour?

The Bill says it will provide 125 days of work but people are not even getting the 100 under MGNREGS. MGNREGS is a fallback option, it does certain things. For instance, it offered basic social protection, ensuring there are no days when people go without food and other necessities. People don’t have savings, so the availability of work through NREGA mattered when there was no other work.

Having this security as a demand-based right had a positive effect on the overall labour market as we saw in the first few years of NREGA, before it started getting diluted.

One effect of this work was that it paved the way for a fallback wage. If there is work available at minimum wages — particularly for women, because there is no gender wage gap here, unlike in agricultural labour markets — it matters greatly. At least 55% of MGNREGS workers are women. If a woman knows that she can demand work under NREGA and earn Rs 250, which is the current average wage, her bargaining power in the labour market increases. She can refuse to work for less.

Studies from the initial period showed that MGNREGS did contribute to raising market wage rates. More recently, however, market wages have risen to around Rs 300 while the MGNREGS wage has remained at Rs 250, and so it no longer plays that role. That is why one of the key demands has been that MGNREGS wages should also be revised, in line with minimum wages and inflation.

NREGA also had the potential to, and in the initial phases did, curtail distress migration. In lean work seasons in villages, the option of MGNREGS ensured that workers wouldn’t have to come to a city to work as rickshaw pullers. With the new programme, that protection weakens.

Subscribe now

What are the contradictions built into the Bill when it claims to preserve the job guarantee? What does this mean for the right to work – are we diluting or dismantling it?

One contradiction is the restriction on employment during the agricultural season. [The Bill restricts work for 60 days during peak agricultural season. This “60-day blackout” will hurt the bargaining power of women, landless and other marginalised communities, said the NREGA Sangharsh Morcha in a press release.]

If work is meant to be on demand, then why say that for 60 days you cannot demand work? One section says the schedule of rates should be estimated on the basis of a normal working day of eight hours with a one-hour break. But the very next clause allows for work extending up to 12 hours. Again, like with the labour codes, they are diluting the right to work. If you’re working 12 hours a day and not getting minimum wages, then that’s bonded labour, not decent employment.

The new Bill proposes digital compliance through geospatial planning, AI for fraud detection and more. There is enough evidence that things like biometric authentication have created hurdles. What risk would tech interventions pose for workers?

The Bill uses a lot of big, vague language around things like geospatial planning and AI-based fraud detection, without explaining what these mean. The Bill frames these digital interventions as great successes and therefore making it part of the law.

These digital interventions have already undermined the right to work, as we saw in the last few years, where wages were not being paid on time and many were being excluded. Wage delays push people to migrate, especially to jobs like brick kiln work where they are paid in advance. These are not people with savings so the timing of payment became very important. And many of those delays were the outcomes of centralisation.

A lot of activism and policy debates had led to the recognition of rural employment as a right. Why do these labour gains need to be remembered today?

NREGA came at a time of great rural distress and across the country there was a demand for a solution to unemployment in the run up to the 2004 elections. Towards the passing of the Act, there was huge mobilisation by civil society, by Left parties; on the street, in the Parliament, many of these issues were discussed. Should it be universal or not? Should it be 100 days or 300 days? Should it be an individual or household entitlement? What about accountability measures? Each of these was a struggle, it didn’t come easily. And it is in locations where the struggle was intensive and people’s participation was overwhelming that the programme functioned the best.

Now it has also become difficult to protest. After the passing of the Act, NREGA unions were formed in different parts of the country and they came together in various forums like the NREGA Sangharsh Morcha. They have all been fighting for issues like higher wages, and against digital exclusions among others. They’ve been trying to come out on the streets as much as they can, given the prevailing constraints of, say, obtaining permission to protest at places such as Jantar Mantar.

Leave a comment

Further reading:

• Nikhil Dey and Aruna Roy explain why the new Bill completely does away with the right to work as an entitlement.

• Jean Dreze’s analysis of MNREGA through the lens of employment generation, women’s participation, and more.

• Read the whole Bill here.

Once, the internet was a place of discovery. You logged in to find others like you: messy fan pages, personal blogs, queer communities, labours of love. Now, 40 to 60% of those links no longer exist. Five profit-driven companies own most of what we see and AI-generated content drowns spaces.

“We are made to forget that the internet wasn’t always like this,” says Smita, who works on gender, sexuality, and internet rights at the Association for Progressive Communications. For instance, the internet was always queer and feminist but has yielded to a grey homogeneity. AI slop floods digital spaces, and government policies have tethered us to online spaces that code dependence, exclusion, and repression. “We have not meaningfully consented to this internet,” Smita adds.

What does it mean to reclaim the internet as a feminist space centering care, joy, and the right to be offline? In a conversation with Saumya Kalia, Smita talks about the history of the Indian internet, the possibility of feminist AI regulations, and why pleasure should be centred in digital rights conversations. Edited excerpts below:

What can be characterised as a feminist internet?

At the core of it, a feminist internet is a space that offers meaningful access – not just accessibility for persons with disability but also access in terms of language, for people with different degrees of digital literacy and familiarity with technology. It is access to information that is not programmed but has multiple viewpoints and dimensions to it.

An internet will only be feminist if it’s not just equal but also equitable. We need a space where there are provisions made to ensure the internet works for people who are at the absolute margins, who are not included in conversations. Unless the internet is equitable, it will not become the level playing field that everyone hopes that it will be. And then ultimately, a feminist internet is an internet for fun, to meet people and have community.

Tell us about the early conversations around a ‘feminist internet’.

My Master’s thesis looked at how lesbian and bisexual women used dating websites as a space to form communities, so it seemed like a direct segue into talking about sexuality, gender and technology. Later, as part of Point of View [a non-profit working on gender and sexual expression], we built curriculums on teaching digital rights and digital security with a feminist, intersectional lens. Usually when cyber security people parachute in, they don’t really see what the community needs – they impose whatever technology they think is right and leave. But it’s not helpful, especially for communities where English is not the first language; if something is inaccessible, people leave it.

The Association for Progressive Communication later in 2014 and 2015 did consultations with 50 activists and academicians from different movements – feminism, LGBTQI rights, digital rights, child rights – along with feminist techies most of whom were from Global South countries, and came up with what was called the feminist principles of the internet. It almost sounds fantastical if you look at these principles and then look at the internet today. But for a lot of people, it was the first time they had a guide on thinking about a feminist internet. It’s idealistic but we encourage different communities to take different things out of it. Say someone looking at freedom of assembly or association could talk about its implications online.

It’s almost like designing a virtual city.

Very much so. Though the internet feels virtual, it’s a reflection of the life we lead outside. So it would be a safe city, but it would be a ‘non smart’ safe city – in that it should not be reliant on surveillance for security but rather the design should be secured by virtue of including people from diverse backgrounds from the very beginning. It’s a subtle but important difference.

What feminist principles has India looked at or should be focusing on?

All of it. India is falling behind in the digital rights space. There’s no index which looks at digital rights particularly but other metrics are telling, like India’s ranking on the Press Freedom Index or the number of government-ordered internet shutdowns. India has one of the highest government requests to Google to take down content. Everything that can be done badly is being done. One of the most horrendous examples is compulsorily pushing people online, which violates one of the most important feminist principles, that you need the right to be offline as well as online. During the pandemic, everyone was forced to get an Aadhaar card for Covid vaccine registration. Recently they have implemented a facial recognition system and an OTP for procuring ration. This is problematic – facial ID is a racist technology, it doesn’t read people (especially people of colour and non-men) properly. What also happens is that the OTP goes to the husband’s mobile, whereas the woman is the one who has come to collect the ration, and she’s unable to take it home. Before this, they had mandated fingerprints when Aadhaar was big but that also didn’t work.

Your basic survival necessity should not be connected to the internet, especially in a country like India where internet penetration is highly gendered and much more complicated to measure than a simple percentage.

What do we mean by a ‘right to be offline’, and what does its denial entail?

I would articulate it like this: if you need a smartphone and an internet connection to get basic facilities like water, access to healthcare, access to vaccines, safety – or basic necessities like roti, kapda, makaan – then you do not have the right to be offline. If a shopkeeper can only give you ration through an online system that scans your fingerprint or your face, you’re being forced to go online.

Ultimately, it’s a matter of consent. During the pandemic, people were forced to buy a mobile phone with money they didn’t have to get regular updates from the government – they were not consenting to being online. For trans women based in small towns in Tamil Nadu, for example, the only way they could get information was when a tempo with a loudspeaker would make announcements about the lockdown, timing of shops, availability of doctors, medical precautions. They couldn’t recharge their phones because the mobile shops were shut. If the government forces the internet to a degree that it’s the only source of information, can you really say the internet is a free space and the government cares for its people?

And if the government insists on pushing people online, they have to build infrastructure to that degree, which is where India differs from other countries. Take IRCTC or the postal tracking website, before you can find the little box for tracking your parcel, you have to click 10 times and scroll past the Prime Minister’s face for five pages. Even with PM Cares and the Covid apps, there was no transparency about where the data is being stored, how long it is going to be stored for, who has access to it, whether the database is secure. When your infrastructure is not up to the mark, what are you forcing people into?

[The Union Government recently notified the Digital Personal Data Protection Act. Activists believe the privacy laws are opaque, undermine informed consent, and perpetuate “information asymmetry between individuals and large platforms”.]

What were the key moments in how Indians experienced the early days of the internet?

In India, people flocked to the internet when internet cafes became a thing. Broadband providers like Sify had cyber cafes everywhere but were immediately taken over by men who wanted to watch porn there. There is nothing wrong with that except watching it in a public space creates discomfort and contributes to a lack of safety, so women and young people couldn’t really access the internet beyond a certain time.

The biggest surge happened when smartphones came up – it was a turning point in terms of access; between 2015 and 2019, cheaper phones, affordable data plans and more operators led to a mobile internet usage boom and more people went online. It was much more affordable than a computer, and even now, most people don’t have laptops or a PC. Demonetisation was another subtle but big jump. After that many other schemes went online because [the government] thought enough people had the internet, which data shows they don’t. And then the next big surge would be Covid.

Subscribe now

These small spikes have been there but the instances of online gender-based violence unfortunately are present as early as the 2000s. The first big ‘scandal’ was the Mysore Mallige – when a college-going couple filmed themselves having sex, and the video was leaked by a friend of the man in the video. And after that, there was the DPS MMS scandal in 2004 which became very big and had very gendered consequences. The boy missed a cricket match but the girl had to leave the school and the country. But I think the reason why people keep defying and going back to the internet, is because there is pleasure in there, there is connection, there is community.

The internet was also a tool for building movements such as the SaveRTI and Pink Chaddi campaigns. Has that changed with the prominence of Big Tech companies?

One of the reasons why governments and private sector are trying so hard to control what people are seeing and doing, or why internet shutdowns have become so prevalent, is because it played such a big role in so many social movements, not just in India but across the world — from the Arab spring and the MeToo movement to the recent Nepal uprising.

The internet has helped in informing and educating people, a better-informed population will also protest when the government is not doing what it’s supposed to do. It has also played a role in bringing together a critical mass of people. Before the internet I could go to Azad Maidan and see what’s happening there, but that’s not the world we live in today. The CAA-NRC protests were another pivotal moment also because the internet was used to build knowledge.

In a lot of countries, the moment a protest erupted was when the governments touched mobile phones. For example, in Lebanon, it was when they decided to charge for video calls on WhatsApp; in Malawi, it was because they decided to increase mobile data charges; in Nepal, they decided to ban social media, which, combined with massive growing dissatisfaction and economic disparity. India has not outrightly banned anything – banning is what causes people to protest – but it has insidiously changed laws to increase control and surveillance, and that’s why people notice slowly. The Income Tax Department this year said that very soon it will have access to our messages and social media. What is the logic? It’s a way to increase control online. And it also shows ways the government misuses the internet. In the Bhima Koregaon case, evidence was planted on activist phones using a WhatsApp link. Pegasus spyware was used to hack phones of politicians, journalists, activists, judges; it should have been a huge issue but there was barely a whimper.

Could a Pink Chaddi-like campaign happen today, in this media ecosystem?

I worry about the corporate control and the government control, and there’s a collusion between them, but more than that, it would be the widespread biases online and offline.

There’s a way to create visibility, even with corporate control, like Nepal showed us recently. The social media and the internet intermediaries were never on the people’s side; it so happened that people figured out how to hack their spaces for desired needs. Facebook never wanted lesbians from Chennai to form a group secretly on Facebook and then meet in person, but it happened because people hacked the system. People can figure it out, but the reality is that prejudice, populism and Islamophobia in the offline world completely permeate online as well.

For marginalised communities, the internet is both a space for self assertion and of witnessing violence. At least 13% of hate posts on Facebook are related to caste-based hate speech, one 2024 study showed. How have these two dynamics of self-assertion and safety played out?

Whatever prejudices exist offline, they translate online and it just changes form. Recently, a Dalit man was beaten up because he posted a Facebook profile picture in which he had a polished, turned up moustache. The other side of the coin is when Kavin Selvaganesh [a young Dalit engineer] was murdered for being in a relationship with a dominant caste woman, his murder and murderer were celebrated online. Like our lives, prejudices flow both ways online and offline.

In many ways, this particular balance that has to be struck is very similar to what happens in the physical world. If you’re visible about certain issues or as certain kinds of people, there will be violence and the internet mirrors the offline world in this aspect. Whenever there is a documentation of joy of people who are Dalit, there is a repercussion. The good thing is that the Internet also creates space and reach that was previously not possible. The risk of online trolling is ever present but there are ways to protect yourself and get support that one could not gather in a physical world.

Share BehanBox’s Substack

When it came out, the internet was seen as this space that will be a leveling ground for marginalised people – for Dalit people, women, queer and trans people, Black people, all of us. That’s also why it got regulated very quickly. The people who created the internet, who are probably white, cis, straight men in global North countries, moved to contain it the moment it challenged their power. They can’t shut it down because it makes them money, so they take to regulations and create more divisions online.

The internet challenges the status quo as it is, and the feminist internet all the more so. I always feel like the Internet has always been a little queer and feminist but somewhere along the line we were made to forget that and it has been taken over. The inventor of the Web very intentionally did not patent it, there’s a reason for that.

What is your first memory of experiencing the internet, and finding pleasure?

There are many. I love the internet. I love computers. I grew up in the UAE, I got a computer, one of those big ones, in the second standard. And the internet was very expensive at that point. I would go play on Neopets, this website where you can have virtual pets. It was a really silly thing, but so much fun.

Because I was queer, the internet was also kind of my life, it’s what kept me sane for a long time. When you grow up in a country like UAE, where even the mention of the word is banned, you start to think you’re going crazy, maybe this is not how you should feel about your women friends, but the internet in many ways showed me I’m not alone.

After I came to India for my undergraduation, I joined dating websites. I didn’t even want to date, I just wanted to talk to people, and that was wonderful, because I met some really nice people there. I have always been nerdy, so finding websites which spoke to specific interests was always exciting – like okay, there are other people doing this.

What did the internet look like then? Messy website, chaotic?

Terrible fonts.

What happened to the design then?

I think the world is homogenised in terms of design today, and it’s inevitable that it also goes online. If you look at Instagram, everything is plain and white and beige, so boring. There’s a theory that when design goes minimalist, it’s also an indicator of fascism. The white walls design in a museum was appropriated by and popularised during the Nazi regime — they started using it because white symbolised purity. They wanted to sanitise how art is supposed to be seen, and also wanted to control whose art is shown.

Everything about cyber security and the internet is about safety and violence. Why don’t we think of the internet as a space of joy?

It’s for multiple reasons. At its core, it’s patriarchy. Before Covid when I was at PoV, we worked with young women and LGBTQ+ persons from different communities, and found young women are often the last ones to get a phone in their households, and even when they get a phone, it’s highly surveilled. The worry is always: who is she talking to, is he from our religion, our caste, is he a Muslim? Families don’t even think about women talking to other women, that’s a very far off concept. This surveillance shows that the immediate worry is that whatever freedom is not afforded to women in the physical world, the phone will become a way to access that. So if you push for digital rights, which is joyful and pleasurable, you will have to accept that your daughters can use a phone for things other than classes or college applications or work. And that’s a very difficult thing for Indians to accept.

Globally, everyone knows there’s actually a lot of work to be done around gender, but if you go anywhere, from the Internet Governance Forum [a United Nations-convened group including governments, civil society, and companies to discuss all things internet and ethical AI] to a UNDP consultation, it starts and stops at online gender-based violence. It’s about how women are attacked, how they are abused – all of which is true – but that’s not why we are online. That’s not why we are fighting for space.

The internet is a place of violence but it’s not the only place where violence happens. What makes online violence feel exceptional and pervasive is there is no accountability and no consequences; it is not really seen as violence. There are also no regulatory mechanisms on social media platforms to handle this in a meaningful way.

Before AI, we poured money into smart cities. This approach to think of tech as a silver bullet has always been there. After the Delhi 2012 rape case, government agencies launched some 70-80 apps on women’s safety, but all of them worked on either geo fencing – if you go beyond a certain point it will send SOS messages – or you could click a button and send information. Both had to do with surveillance. But the app is not going to keep you safe until you change actual policing on ground, put street lights in an area, and challenge social norms.

What is the cost of neglecting these affective elements of our online lives, say the right to pleasure, within the digital rights movements?

Digital rights are already sidelined and when you create hierarchies, some people will never get their turn. The TikTok ban in India is an example of this. TikTok was used primarily by those from poorer communities for pleasure. Daily wage workers – say, brick layers, construction workers, people making utensils – were posting videos of dancing, of teaching English and more. Instagram was too sanitised; Tiktok was free and easy. That was a really rich time for people to document their lives and create their own videos and have fun. I’m not saying that short form content is good or there are no data concerns, but it gave people freedom.

TikTok is also a tool of political expression – in many countries, including the US recently, it gets banned when there is a lot of political activity online. When it got banned in India, there was barely any protest. Imagine if Twitter were banned today – people and parliamentarians would call it an attack on freedom of expression. Why is the right to be on TikTok not seen as the same? There are caste and gender implications there.

The government recently released the AI Governance Guidelines ahead of the India AI Summit. Can AI regulation happen in a feminist way, and what should India be looking at?

Firstly, we shouldn’t be holding it in India. The current government and its policies have made enemies out of our neighbours. People from Pakistan can practically not get a visa, same for those from Afghanistan and Bangladesh. India stopped visa-free entry for Myanmar passport holders, and the Rohingya people in particular are unlawfully detained and deported.

If half of our neighbours cannot come in easily, what is feminist about [this summit]?

Also, I don’t think it’s possible to imagine a feminist AI – at the core of it, it’s an exploitative technology. The question of regulation is complicated, it has to be done on multiple levels. First, we need to regulate disinformation. How do we hold anyone accountable when platforms aren’t accountable? Regulating AI has to include regulating hate speech, online gender-based violence, and deepfakes. So policymakers may have to look at a broad framing, that AI should not be used to harm or infringe on rights. But it can’t be so broad that it is abused and becomes a tool for censorship. If we leave it to the Indian government, they will try to do what they tried with the obscenity law and anything involving digital laws – they would go with a Victorian mentality, and now a Hindutva approach as well.

The harder question is what are we regulating. Its use? Its creation? Its impact? Are we talking about server location, consent in data collection, or whether an AI model can be course-corrected while it’s being trained? And how do we ensure the training data is actually diverse, when lack of diversity is the starting problem? Regulation is still far away because you need to ask these questions to understand what you need to regulate. As with other regulations, it could further sideline the conversation on the internet as a space of joy and pleasure.

(One recommendation in the AI governance guidelines pushes for the “use of locally relevant datasets to support the creation of culturally representative models and applications”.)

How can we improve Instagram or Facebook as a feminist space for pleasure?

Firstly, they need to un-black box their reporting mechanisms. They need to tell us how when you report violence that’s happening, how it is actually being addressed. They need to increase moderators on these platforms, which they have cut further. And they need to stop AI moderation, that’s a recipe for disaster.

They need to work on the design to make it more people-centric rather than profit-centric. Like stop the infinity scroll. And Instagram knows how to do this — it was like this before, so go back to that. Go back to a chronological timeline, showing posts from the people who you actually follow, and not the retweeted random posts. Make it more accessible in terms of how you post, and reduce AI (including AI-generated video content).

One thing we could push for immediately is to ask platforms for proper AI labelling. If a content is AI-generated or it has used AI at any part of the video, it needs to be labeled. Instagram has a ‘made with AI’ tag but there’s nothing regulating it. Clear labelling will help check information pollution – you will know if something is misinformation or disinformation immediately. And it would help social media companies too: the industry is training its own models on AI slop, which is rubbish, so if you do this labelling properly companies will be able to create better AI products. Even for purely capitalist reasons, companies should do this; people are very unhappy with having to ask, every minute they are online, if something is real or AI.

What is an example of a feminist internet in practice in India?

It can be on Instagram or anywhere, but there are pockets where people try to create spaces which are queer and feminist. During the pandemic, a bunch of non-binary trans people who were stuck at home – which are violence spaces – created a group on Instagram where they would send photos of themselves as they present and appreciate each other from there. This queer organisation in Manipur used WhatsApp and those small phones to get deliveries in person to people who needed HIV medication and hormone therapy. It was beautiful, and so necessary.

I don’t believe that you have to leave everything, go to the perfect platform or the perfect technology – it’s impossible. We don’t have the money, we don’t have the skills. Fediverse [a decentralised, more democratic way of social networking] is an aspiration, but it’s currently very inaccessible to a lot of people here.

It’s almost like it’s not about the digital space, but what people make of it and how they use it which makes it a feminist internet.

Precisely. I don’t think the internet is inherently bad. It’s just become highly capitalist, and people are not intentional in how they use it. Earlier with a modem, you had to be intentional because it costs money, but now you don’t have the intentionality partly due to how platforms are designed.

Like with the physical world, the online world also needs intentionality.

When we are out in public, a lot of us don’t wander aimlessly, it’s with some sort of purpose. I want to go for a walk in my neighborhood at this time because the weather is nice. I want to go to a park. Or I want a chai. Online, you sit and suddenly you’re scrolling on Instagram. It’s the equivalent of ‘oh I’m standing here might as well go walk down the road’. If we don’t do that in the offline world, why do we do that online?

That is why the right to be offline matters. Platforms and governments make it harder and harder to be offline. If you feel like you need the phone to travel, eat food, get a train ticket, how will you not live on it? And when this has been enforced so much, it’s very hard to be intentional on the more frivolous, joyful corners of the internet.

But pleasure comes back when you broaden your language and idea of technology. Remember Walkmans? It’s technology. Your paintbrush, a good pen. Silly games that you play, Tamagotchi, it’s all technology and it’s all pleasure. That’s why we started going into phones – remember the snakes game? We forget these things sometimes and lose joy in the process also.

Behans, what are your early memories of the internet? What brought you joy, and where do you find pleasure now? Write to us at contact@behanbox.com.

Leave a comment

Startups are everywhere, so are tech bros. They speak the language of innovation, flexibility, empowerment, selling the vision of a tech future that is an individual’s to mould. But beneath the hackathons and the VC-fuelled unicorn craze lie histories of exploitation and indignities.

India’s current startup ecology is deeply tied to its neoliberal roots and global tech structures, says scholar Hemangini Gupta, author of Experimental Times: Startup Capitalism and Experimental Futures in India. She traces the caste and gender privileges that go into the making of a tech bro, women’s innovation and belonging in masculine spaces, and the obscure promises of ethical AI solutions. Companies have to start including workers in decision making and profit sharing to build equitable tech futures, she tells Saumya Kalia in an interview. Excerpts below.

What sparked your interest in elite tech entrepreneurs?

For my PhD I wanted to look at questions of information technology and the women in it, how they were reconfiguring some of the expectations around gendered respectability, caste and class. But IT companies were very reluctant to have an ethnographer amidst them. There were concerns around patented information and work processes, and I wasn’t able to get access to an IT company despite my best efforts.

This was also a time [the late 2000s] when a lot of my friends were moving back to Bangalore to be entrepreneurs. There was a discourse around entrepreneurship and building Bangalore as a startup city that was attractive for entrepreneurs. The city offers a combination of government infrastructure – in terms of funding actual labs and large fairs and meetings and conventions – and investment by middle class entrepreneurs, many of whom returned from the US over the last 15 years and have been trying to build an ecosystem conducive to entrepreneurs.

I eventually became interested in several ways through which Bangalore was being repositioned. One was this idea that Bangalore is not just the site for back end work but also for entrepreneurship from which creative work can be exported to the rest of the world. The second was that it was the city where many middle class professional migrants were moving back intending to do meaningful and creative work. I wanted to see what it meant for people to move to the city and to ask: how is work being reshaped in this current moment?

You speak of two types of experimentation in the book, one practised by male entrepreneurs for profit (and eventually for nation building), and one by women entrepreneurs for maintaining infrastructures of care. How do you distinguish between these?

The first half of the book is focused on understanding the male entrepreneur, a very masculine figure, and trying to unpack the expectations embedded in this idea. I attended startup meetings, networking sessions, and startup festivals where people were taught how to train themselves to be this entrepreneurial figure. This question came to me: what happens after that moment of inception, after you birth an entrepreneurial idea, how does it sustain itself? What forms of labour are necessary for that idea to grow?

Some of the inspiration came from cultural theorist Miya Tokumitsu who wrote a book about this ethos, Do What You Love. She looked at Apple as an example, how one never really thinks about who makes the Apple iPhone or under what conditions they make it. So the second half of the book focuses on a particular entrepreneurial company and tracks the everyday life and work of people employed there. I found that the expectation was that all employees become some sort of entrepreneurs themselves.

The values and habits of the male founder were normalised and expected to be adopted by everyone – they were expected to be experimental with the food that they ate, how they held themselves, the kind of leisure and creative pursuits that they’d have. There was this expectation that our bodies would be prepared for an entrepreneurial life, which is not just about work.

The book contrasts these two kinds of experimentation, one by the male entrepreneur who heads the company, who enjoys the profits of the company, and the other by women who experimented for survival. Most of the women were young, middle-aged employees; none of their parents, certainly not their mothers, had professional office jobs. For them, the experimentation was not just in being with each other and trying to figure out how to move in these spaces of leisure – it also became about experimenting with their personal lives. They took a lot of risks in terms of who they made friends with, how they fell in love or how they got divorced. The workers in the company came together to form these infrastructures of care and socialising, and it’s through that that they survive this everyday expectation of experimentation in the office.

Share

Do women entrepreneurs enable or upend the model of startup capitalism?

I still ask myself that question. It is sometimes phrased as, is there an alternative to capitalism? At the end of the day, people want to know if there is a possibility for change and at this moment, I don’t think there is. We are encountering women who are entrepreneurial but that doesn’t necessarily mean the nature of entrepreneurship is changing. Women are entering a world that is already dominated by masculine expectations around how they are available to work, where they can work, and how many hours they put into it.

But then there’s also the question, are these women doing something to undo from within the expectation of capitalism? The answer is probably they were working within it. For instance, on days when you could order food at IT companies, women employees would often order biryani but the CEO would come in and say: ‘Come on guys, you can’t just order the same food every time. We’re a global company. Why are you doing the same thing every day?” To him, it meant you’re in the startup neighborhood, you should be eating experimentally. But I think among women was the understanding that there were others with you who were also equally a bit tentative and ambiguous about entering these new worlds.

So they took that ambiguity with them, and sometimes it manifested as a tentative approach to work, which was not really the gusto and the energy that the entrepreneur expected from them. Sometimes they just refuse the terms of entrepreneurship. One of the schemes in the office was to sign up for a class to enhance yourself – maybe learn French or a new skill – but nobody ever signed up for these classes. The entrepreneur was always very confused because he’s giving them an opportunity to be ‘neoliberal’ in a sense, to remake themselves as flexible people in a market that’s constantly requiring new skills, and yet, people are just refusing that. I understood it almost as a refusal of that idea that one should constantly work to better themselves and produce more profit for the company. Women wanted to have fun at work – they wanted to learn things – but they were also very emphatic about living in the moment and not extending themselves toward a future that they could not anticipate or control.

Other times, they’d plunge into the worlds that they were expected to plunge into, like, literally, when they went on an off site trip of water rafting. None of them knew how to swim and they were terrified but they came together, held on to each other, and participated in this ‘team building’ activity.

There weren’t uniform ways in which women entered these worlds of startup capitalism, but their refusal to follow a coherent, linear pathway is a sort of undoing of its terms.

These ‘techno-entrepreneur’ figures, what privileges do they hold?

I should have more emphatically called it tech bro culture. At these startup meetings and festivals, it was very much the tech bro, this figure who was always available, who was centered. At this incubation lab I visited, six companies pitching an idea were telling a venture capitalist why their company was amazing. To one of them, the venture capitalist asked how their company was ready for the market, and the person said they have people working around the world, they’re always online, they’re 100% here. Even at three in the morning. This is one example of how the expectation around startup entrepreneurs was that they would always be present, and I compared that with some of the women entrepreneurs who worked very different hours. They weren’t up at three in the morning – they were up at six to pack lunches for husbands and children, then maybe do some work for the startup before others left for office and school, and then attend networking meetings, which were almost never possible for many women.

Even just being able to participate in a startup meant being able to meet collaborators, to network, to meet venture capitalists – but that physical mobility through the city was shaped by where they were in their lives and who they were. Middle-class married women tended to be literally and figuratively on the outskirts of the startup economy. On the other hand, the tech bro figure, or the figure who was ever available to participate in startups, was often a young or middle-aged man, who didn’t have care responsibilities and could really be mobile, both in terms of time and space. The city itself was shaping who could access these dense worlds of startup networking and who couldn’t.

Can you talk about the ‘cyber coolies’ you write about in the book? And the presence of caste ideas and dynamics in entrepreneurial spaces?

When it comes to entrepreneurs or most forms of middle class work in India, nobody wants to talk about caste. It’s almost like you’re embarrassing them by bringing up something that is so ‘archaic’. There’s this belief that India is moving into a future that doesn’t rely on caste differences – a future that is prized on merit and what people can do.

In my conversations, people didn’t bring up caste but there were explicit references to this figure of the ‘cyber coolie’. People would say they were interested in “ideas” or “management perspectives,” not in being cyber coolies— a coolie, in that imagination, is someone who is told what to do. Historically it’s a figure of racialised labour who is sent to plantations to perform a certain kind of work. Some even connected it to visa regimes like the H1B visa, under which many Indians get to travel to the US for work but cannot then leave the company and work elsewhere unless the company sponsors your ability to be mobile. So as entrepreneurs, they wanted to distinguish themselves from labour that is attached to a higher power – they wanted to say we are capable of being mobile and independent of the racialised regime of technology work.

There were constant hierarchies being assembled between work that was creative and intellectual, and work that was assigned to you, task-based, and what they considered menial. These hierarchies are themselves expressions of caste-based labour because they distinguish and assign different values to the labour of the mind from the labour of the body.

Even though people didn’t explicitly talk about caste advantage or disadvantage in differentiating work according to these hierarchies, these caste imaginations that galvanised work were considered valuable.

Why did you shift to telling the story of startup capitalism from the point of view of workers?

There’s always an excitement around entrepreneurship, that we can reshape historical age-old inequalities if we use this technology correctly. We pay a lot of attention to TV shows that celebrate and reward entrepreneurs and the million-dollar-idea. That misses the very next step: what happens after you have this brilliant idea? In a public sphere and a realm where people are constantly exhorted to do what they love, be innovative, take risks, it’s upon us to ask that second step-question.

India is seen as a superpower due to its technological prowess and convenience of getting things delivered within 10 minutes, but the question is who’s bringing this convenience, what kinds of labour are being recruited, what is the precarity required for this market to function?

Globally, we hear these ideas of “do what you love” and work hard from tech CEOs but these sit uneasily with the fact that India’s IT and ITeS sector has witnessed its steepest layoffs in decades. How do we reconcile these two realities?

Those things are fundamentally linked – the more precarity there is, the more entrepreneurship is emphasised. Once people no longer have the ability to imagine stable jobs, entrepreneurship becomes one way to sidestep questions of ensuring basic benefits or job security, or of a State that provides for people who might not be in informal or long-term employment. The more there is precarity, the more there is austerity, the more the State retreats from providing essential goods and services at scale, the more there’s the expectation that people will just innovate or be entrepreneurs to support themselves. In the absence of there being more stable infrastructures of work, it’s this speculative idea that you will just make it big that really takes hold in people’s minds.

Entrepreneurship and platform work celebrate “flexibility” and “independence” but these ideas conceal a deeper feminisation of labour where people are working more in precarious roles but earning less. How has the language enabled this shift?

It’s making workers precarious, and there is a sort of responsibilisation also. In my research, workers were often referred to as ‘travel consultants’, and it wasn’t clear to me why they were consultants when they were employed by the company. These terms became really important in taking away people’s identity as workers or as employees; there was no shared understanding that they were participating in conditions of work.

By thinking of themselves in these sort of privatised terms of ‘partners’ or of ‘consultants’, there was also an understanding of one having control over one’s labour. A consultant is someone who can decide the terms of the consultancy, what they earn, when and how they work. None of these were things that people could actually do at work but by calling them consultants, companies passed the circulating belief that you are in control of the terms of your work, in a way that you’re actually not.

If ESOPs are a form of recognising labour, what do you make of unions demanding ‘worker IPOs’ and a say in decision making?

Concrete material redistribution has always been a fundamental worker demand. If people are invited to participate as equals, they should literally have a say in how the company is run, give feedback on management structures, and share in monetary benefits as well.

In the companies I studied, people are told to take ownership of their ideas, but they are never given extra money for having taken on that role; it’s just a sort of a nebulous idea that the company is yours and you should participate in it.

If a company wants to have egalitarian norms, it will have to let people participate not just in decision making, but also in sharing profits. And to recognise that in order for an entrepreneur’s ideas to hold, people have to take great risks.

Do you see any continuities or disruptions in these histories in the present moment of global AI labour?

I’m curious to know whether my arguments hold in this new moment of data annotation work. In Bangalore and other parts of India, we’re seeing experiments with impact sourcing and ethical AI work. Companies are working with self-help groups to assign data work to marginalised communities and there are efforts in thinking about redoing the terms of data work in that some returns go to the communities and that the conditions of work are not exploitative. These promises are very compelling, especially coming from the Global South, but we’re still figuring out if these projects can really materialise.

What feels continuous with the histories I wrote about is this very visible realm of entrepreneurs who are interested in wanting to ‘remake’ AI work as something that’s well paid and dignified. I’m thinking of initiatives like Karya [which claims to be the world’s first ‘ethical AI company’ that offers digital work to rural Indians] and others who have been foregrounding ethical AI – they are literally visibilised in terms of a TIME magazine cover but also very visible in the public sphere.

But when you look closer at the broader realm of AI work being sent to Asia and Africa, we have to ask: do we know about the actual conditions of this world and how these promises hold in practice? If we begin to ask who does this data work, we’ll find it’s people who are caste disadvantaged, people who have been laid off from agricultural work because of over-extracted land, people who don’t have much choice in the conditions of their work. We might find a very similar dynamic to the book, where there is a tension between the visibility of entrepreneurs who speak about remaking work and the invisibility of the racialised and gender labour that make this work possible.

There’s a lot of potential for this work to be better paid but there’s never that second-level question: does this change the conditions of work, does it offer protection, and who bears the environmental costs of these data infrastructures?

What is the role of the city in realising these tech futures?

I have different responses to this question.

At the 4S conference recently, Ashique Ali Thuppilikkat, Aditya Nayak and Aditi Vashistha presented a paper on how even as platform economies grow and are distributed across the city, there is still a need for massive dark stores to stock produce. But where are these dark stores located? It’s very difficult to find them – they’re not explicitly labelled, they’re not on Google maps. The authors tracked these stores using various forms of geospatial mapping and found them often on the edges of peripheries of low-income neighbourhoods. It raised interesting questions about how even as we imagine that platform work is moving to not be site-based, it still requires urban spaces in order to cite certain key logistical points, which would then influence the rental economies in those areas. It plays a role in how workers might access the space and where they might live and how they are priced out of the city so to speak.

The second has to do with how workers move through the city. In my research on beauty platform workers in Bangalore, I found that even though there’s not necessarily an office or a central site, your movement through the city becomes significant to how much you can make and where you need to live to have access to the right kinds of clients. The city’s role in labour is constantly changing. In Bangalore, in earlier moments technology work happened at the outskirts of the city in IT Parks. Now the work, rather than being concentrated at a particular building or in a particular site, is distributed across the city. Even as workers move through the city during traffic, rain, and extreme heat, the city still retains significance as a site of everyday work. It’s no longer the same kind of site, but it’s the streets and the buildings and the rental economies that shape how workers have access to work.

Do these spaces hold the possibility of collective bargaining?

Anthropologist Lilly Irani’s work shows that when we speak of innovation, there is a certain scale at which we’re imagining an idea to be able to move. She examined that for small local ideas – like keeping water cool in a clay pot – to achieve the status of innovation, it was required to scale and travel out of a local circuit in order to be recognised by the government and international consortium.

That expectation that innovation must scale, grow and multiply is inherently in tension with an idea of democratic participation or opening up something to a larger number of people. If to innovate is to celebrate and prioritise a scale, feminists might ask: what about something that is small, local and meaningful but not necessarily something that everybody wants, or that 1.2 billion people need?

Entrepreneurs in Bangalore often say a good idea is one people may not need but they may begin to want it. This combination, of innovation being necessarily at scale and it being something that can be made desirable, to me those two things sit in tension with our democratic imagination: to enable diverse voices, abilities, levels of participation to be seen and recognised.

We would love to hear about your experiences and perspectives of the IT and ITeS sector. Write to us at contact@behanbox.com.

Leave a comment

What does it mean to provide care, and who counts as a care worker? Even official definitions, which speak of goods and services essential for “sustaining individuals and households”, fall short in capturing the spectrum of labour essential for human survival. We see care everywhere — homes, institutions, market spaces — related to age, disability and illness. Some of it, done by say nurses and ASHA workers is paid but involves unsafe, devalued, and exploitative working conditions; some by homemakers and community mobilisers go unrecognised. Care and reproductive labour by its very nature involves multitasking and different levels of skills, and is responsible for 7.5% of the GDP.

Care is difficult and skilled labour that is valourised but we have been socially conditioned to disregard its value, says Neetha N, a scholar and economist at Centre for Women’s Development Studies. Patriarchy peddles it as women’s work, and caste shapes who does what care labour. But without urgent investment, the world faces a global care crisis with the most marginalised women hit hardest, so the State has to step in, Neetha tells Saumya Kalia in an interview. Edited excerpts:

What first drew you to the study of care work?

I came to Delhi in 1998 to work at the National Labour Institute (NLI). Every morning, I’d see a group of women coming to sign in at the gate and enter the apartments nearby for part-time domestic work. It was something I wasn’t used to. In Kerala, some homes had domestic workers but you wouldn’t see workers there in this kind of condition; because of the trade union movement and social movements there, there’s a certain dignity attached to any kind of work. But here, the women were stopped at the gate every day, made to sign, they were always running around. I was fascinated and also very troubled by the way they were treated. I kept wondering: what’s the system like here? How much do they earn? What kind of work are they doing?

At the NLI, I was asked to specialise in women and work, and later in 2000, I proposed a research project on domestic workers in Delhi. This was also the period when more women were entering domestic work, many migrating from rural areas, especially Bengali women, and [there was] an emerging trend of feminisation of paid care work. That research itself was very revealing. I started learning about the different systems — part-time, full-time, live-in, live-out — and realised there were almost no studies on domestic work in Delhi at the time. I spoke to more organisations and my interest in women and work grew. Domestic work became my entry into understanding care work more broadly.

Before arriving in Delhi, did you see examples of care work within your own family?

Yes absolutely. There were six of us, and my mother, a housewife, often found it difficult to manage all the housework and care. My father never helped her, he was a traditional husband, the kind who expected everything to be ready when he came home, food on the table, everything in place. We were two sisters, two years apart, and we used to feel irritated and agitated at the inequality we saw in our homes. My mother was working all the time, whereas my father had a lot of leisure. As children we both wondered why it is that only mothers had to do it all.

Slowly we also started to realise that no one saw the housework or carework women do at home. Everyone spoke of her as if she was not doing anything, whereas she was the one working the most – we hadn’t fully developed these questions but this inequality disturbed us. So we would pitch in to help my mother, but never the brothers – they were told it wasn’t their role. That gendered division of work was so clear.

Share

You have written about how care is highly demanding work but the myth is that it is unskilled and innate to women. Can you explain why and how gender becomes a form of labour training?

This myth is very naturalised in our context because women are socialised to do it from a young age. In my home, no one objected to girls performing these tasks; when my younger brother was unwell, it was assumed that the women and girls would pitch in and take care, but a similar expectation was not put on men. It’s a repeated messaging that you have to do it, this is your work, this is women’s work, and this separation is reinforced over time. It has gone so deep into people’s minds that because all women are doing it without any complaints or concerns, there’s no skill in it.

Care is highly skilled work, not everyone can do it. Of course, people can be trained, but that doesn’t mean the work is unskilled. Yet societies and families and communities have always devalued it, treating it as if it’s no work at all.

What place does care work have in India’s cultural imagination?

It differs from location to location. In Delhi, people think of domestic workers because of how it has become part and parcel of every upper-middle and lower-middle class household. In rural areas, people may not always think of the identity of a care worker because this concept is very academic and technical, but if you explain to them, they will immediately talk about ASHA workers and Anganwadi workers.

A category people readily identify are nurses – they interpret ‘care’ in the literal sense, as something that is offered in hospital, in an institutional setup.

Missing here is the idea of unpaid care – what housewives do for their husbands or parents or children – it’s not very instinctive to most of us. In every family, there is a woman who will take care of all the housework and care work. If it is not one, then there will be another; the family will always find women within its system to do it. The government or capital doesn’t want to acknowledge the problems with that assumption of gendered families.

When you ask women about care work, especially as a form of labour or work, they don’t think of themselves as workers or of contributing to the family income. Care is seen as a family thing, and it is so personal, they don’t identify themselves as care workers.

We also see this with women in domestic work, textile industry, or other types of casualised work opportunities like platform work. Is it language or social conditioning?

Language is an issue but in most cases, it’s a result of the condition in which they have come to understand their care or housework. When we did time use surveys, women would trivialise their work, they don’t want to even look at these tasks as work; they would say women know what we do and didn’t feel the need to spell it out. You had to repeatedly ask. That is where the problem is. It’s been seen by women also as something which is not worthy of sharing. Of course, things are evolving — with more awareness and also with education — but it’s not a major change.

Also patriarchy and gender norms play a role. Women are not supposed to say ‘I’ve done this for my son’ or my daughter or husband, women’s work is to keep them happy and it’s not considered socially acceptable to talk about it.

Subscribe now

The way we have the concept of care has in itself that affective and moral dimension. And that’s one reason why some feminist academicians now like to use, instead of care, social reproduction. Care is important for reproducing labour power, and it is so embedded in the economic system. This gets sidelined when we speak of ‘care’ which has more of a moral, affective tone to it and is a limiting concept in terms of understanding the labour involved. Social reproduction as a concept legitimises care as labour. It doesn’t see care as purely emotional and affective, it detaches from moral ideas to acknowledge other dimensions.

Care work is also often described as ‘invisible’.

Across different care sectors, you will see invisibility in the literal sense and invisibility in terms of how that work is valued. It’s not about just seeing a person but also about seeing what they do, how much time it takes, and what it is worth. Invisibility comes when the work happens inside homes, where people don’t consider them as contributing labour. This is an issue with domestic workers who work in individual, private homes where employers don’t talk about them as workers. Even when some forms of care work are visible, people don’t understand or appreciate the underlying skill. This is the case with women who are doing housework and care work, they don’t articulate it as work.

Care work is invisible even in economic frameworks. You’ve written about how time use surveys and other valuation instruments have a narrow view of care work, care givers and their relationships. Can you elaborate on that?

The moment we talk about ‘valuing’ care work, we turn to market tools like wages to measure work. But that kind of evaluation will always have problems: whose wages are you going to use in the evaluation exercise? The other method of valuation based on opportunity cost is also problematic. If I am an educated, highly skilled person, can my labour be valued more than someone who is an unlettered person, even if we are doing the same housework and care work?

All valuations get tricky because this work is happening in two different realms, the market and households. The market reflects the social understanding and skill that goes into each work and ends up reproducing those inequalities in its calculations. Since society devalues both housework and care work, the market ends up devaluing domestic workers’ [labour] also, which is why their wages are often lower than some informal sector workers.

You can never arrive at any calculation which will give a sense of what contribution women are making to this economy and society – either you overvalue it or you may devalue it. Valuations can be used for framing policies but I don’t think it moves the needle on care work or on highlighting issues of sharing or reducing care work.

What do you make of how courts have responded to this problem of valuing women’s care labour?

Court cases have valued women’s work either based on the time spent working or how their unpaid care enabled men to earn livelihoods, but there is no uniformity in any of these judgments.

Some have offered compensation amounts, not wages, to women for housework. Many state governments have also introduced this. The problem with this approach is these are small arbitrary amounts – there’s no reason why it should be Rs 1,500 or 2,000 – and disbursal is contingent upon when they announce it, say during elections. These may bring economic relief immediately but in the long run, they reinforce the existing labour market discriminations. If as feminists we are arguing for equal sharing of housework and care work, then these policies do not help much.

Share BehanBox’s Substack

What role did India’s neoliberal reforms play in shaping care conversations?

Conversations about wages for housework were in many ways about valuing women’s contribution to the economy. The underlying idea was that men could participate in the labour market only because women sustained families through unpaid care. That’s where the concept of a ‘family wage’ came in, the idea that wages should be high enough to support the household and women should have access to benefits and state facilities.

Another argument also focused on women as mothers, in charge of taking care of children and their education. They were in that sense nurturing and reproducing the next generation who will be the labour for the capitalist system. If all of women’s work is contributing to the economy, to capitalist profit, there has to be more sharing of this profit.

Among feminist economists, the objective of these conversations was also to show that women’s contribution to the economy is substantial, and governments need to change macro policy to address women-specific questions, like providing access to clean water or other infrastructure requirements that will unburden or reduce the burden on women. The idea was to make a case for why the state should have a policy specific to women even though they are not in the formal GDP framework. We had to take this approach not entirely because of the market but the way the State’s role had changed after neoliberalism reforms.

Care work is often done by the most marginalised women. How does this decide who does care work and how that work is valued?

Social identities are central, especially in the context of domestic work, they come from the most marginalised communities – they’re poor, migrants, and from lower castes – which adds to their exploitation and invisibility. Even in the case of more formal care workers like nurses, who come from a wider range of backgrounds, there’s still a lack of recognition and invisibility in that sense. The care they perform in private hospitals and small clinics, is their work valued? This is a larger question of labour laws and how privatisation has deteriorated working conditions.

Caste also plays a big role in how care work is divided. You rarely see domestic workers from Dalit or Adivasi backgrounds doing certain tasks like cooking, they can only do cleaning work which is seen as physically demanding and culturally impure. Cooking also earns you higher wages in comparison to manual work like cleaning work, because one is considered skilled, the other not. These perceptions are shaped by caste-based understandings of who can do what. Even ASHA workers, Anganwadi workers and midwives can be seen as doing work – like going inside homes and travelling all the time – that society may not find acceptable for ‘good women’. Everything is connected.

You and others have written about ‘care drain’ and how emotional labour gets displaced in global care supply chains. Can you elaborate on the role of migration in the Indian context?

A lot of women migrated internally from rural to urban especially after the 1990s due to rural distress. Simultaneously, in urban areas there emerged an upwardly mobile middle class, increasingly employed in the service sector, which made it possible to hire cheap labour and in turn create a demand for domestic workers.

In those initial years, many women migrants who entered into domestic work never revealed their status in the city [back home] because their work – going to other homes and performing housework – is not socially accepted in rural areas. Even though they were earning, their work wasn’t socially accepted. Things have changed now, at least in Delhi, partly because wages are much higher now than what they used to be.

Many women also arrived as part of family migration, with the understanding that men will also find some job here. However men could only find casual, irregular work, so women had to take on the responsibility of supporting the family. This was a big thing, that family migration was dominated through the earnings of the women.

Now we’re seeing women’s singular migration also, especially as live-in domestic workers, as they migrate to cities and support their families back in villages. This is also what is happening in terms of international immigration.

What is the emotional cost of migration on women’s ability to provide care at home?

We speak about it, but the fact is, women need this money to provide basic necessities to their children – economic distress dominates their approach to child care. When you speak to them, many of them are actually happy they are able to support their education and send money.

But for women whose children are living with them in cities, there is constant anxiety. It’s not about lack of care that they can give but about safety. When I was studying domestic workers in Delhi in 2000, I went to Nizammudin, in a locality where many domestic workers lived, and I saw some children, all below six years of age, tied to unused railway tracks with a saree. The women said they couldn’t leave them alone at home and they couldn’t afford to skip work, but here, by tying them to this unused track, they can play and they won’t move beyond this.

Women find different ways to manage their childcare and other care demands. In some cases where domestic workers migrate or are unable to provide care at home, the eldest girl child is in charge of caring. Then what happens to that girl? Her education, freedom, understanding of leisure are all taken away. This entire cycle of care is very generational and influences gender empowerment.

Unpaid care work is seen as being beyond the scope of State intervention, the idea being that the home cannot be not a site of an economic transaction. What do you make of this?

Unpaid care work is more complex than market work. The labour that unfolds inside homes is more intense and deeply habitualised, you cannot decide whether or when to stop it, there’s no choice. In our context, it is also very clear that it is not about the work but how that work burdens women through a lack of agency in terms of how much to do, what to do, when to do. There are no choices for women. In the market if you don’t want to do a job, you can decide to leave and work elsewhere. But how many women can just leave the house and walk away?

You can only ensure safe and dignified work conditions through reasonable wages and working conditions; there have to be leave policies and some form of work regulations. Right now they are compelled to work in seven-eight homes and they are running from one house to another – there’s no regulation which adds to their burden of employment.

Scholars like Alyssa Battistoni note capitalism is disinterested in care work due to low profitability. Do you agree?

What does it mean to say the market ‘undervalues’ care work? Take nurses, the market is making a lot of money through hospitals yet nurses remain underpaid nurses’ work. That is also happening in schools: teachers are counted as care workers but on their backs private schools are making huge profits. So I don’t think the issue is whether capitalism can make profit from care – it is how the market perceives the skill involved in care work and who performs it. The assumption often is women are performing these roles so you don’t need to pay them fairly. If care work is going to be low paid, the system ensures it’s done by women from the most marginalised groups. The market knows how to use these existing social divisions — of gender, caste, class — to its advantage.

The market is also currently defining value and has also come to a point where it has access to cheap labour. Even in factory work, wages are kept low, not because the work lacks value, but no regulation compels them to pay more. This is the larger problem of how markets are not regulated that helps them escape accountability. Of course, the market may not be interested in certain sectors, like investing in creche systems, but that’s where the State has to pitch in. There are areas the private sector will never invest in unless forced to. The State has to tax capital, mobilise enough resources, and prioritise care at social and community level. But resources have to be marshalled, and the State will have to push back and use the market’s capacity to fund this infrastructure.

Share

If you were to value care differently, what would that look like?

Valuing care is not about assigning numbers or wages – it’s about understanding how important care is for the society, how it keeps the social fabric together. Valuation here is about equal sharing, recognising housework and care work as central to the economy and for running everyone’s life. Care workers should be given importance, acknowledged and supported. It should also not be gendered, why should it be? If anyone can do it, then let us socialise everyone to do it.

And this is where the State can enter with policies like reducing, redestributing, and revaluing care work. The family is a site of patriarchal relations, that character has to be shaken, and no one can shake that except the State.

How do you understand care work and what does it look like for you? We would love to hear about your experiences of care work. Write to us at contact@behanbox.com.

Read here about our Care Vertical and why we’re centering care as the future of work.

Leave a comment

For at least two years, Mitanins, the frontline health workers of Chhattisgarh, have been campaigning to be formally absorbed into the National Health Mission (NHM). The programme was, for two decades, implemented by the State Health Resource Centre (SHRC), a state-civil society partnership organisation. Mitanins had hoped that integration into the NHM would bring them a step closer to formal government employment.

So, last August, when the NHM announced that it would take over the operations of the Mitanin programme from the SHRC, the workers believed that it would pave the way for a smooth transition. But the euphoria was short-lived. Soon after, the programme faltered: Workers told Behanbox that payments have become erratic and there is little clarity about how the new system will work. And within months of the transition, the government roped in an NGO to run the programme, creating fresh apprehensions in the cadre.

“Earlier, we used to get all our dues on time but since the NHM has taken over, the payment system has been a mess,” said Lamiya Yadav, a Mitanin from Kawardha’s Pandariya block. In the first few months, the NHM used to send messages providing details about the remuneration, she said, but these stopped after three months. “Now we get random amounts and that too sporadically,” she said.

All the Mitanins and union leaders that Behanbox interviewed reported the same issue. The last payment Kamleshwari, another Mitanin from Kawardha district, got was around Rs 4,000 in July. “I don’t know on what basis they have sent it but I am yet to receive any payment since,” she said.

The Mitanin programme that predates the NHM’s ASHA programme has a distinct structure–women from the cadre are promoted to supervisory roles and also train other Mitanins. These coordinators, who were part of SHRC, also verify and approve Mitanins’ incentives. ASHA workers, on the other hand, are supervised by Auxiliary Nurse Midwives (ANMs), who are employed by the health department.

Workers are now raising questions about the new order that remain unanswered: Will it adopt the NHM practice for ASHAs and make the Mitanins accountable to ANMs or other health department staff? Will the existing system of reporting continue? Could many of them lose their jobs for not meeting the recruitment criteria for ASHA workers as health officials have reportedly told some Mitanins?

Ahead of the 2023 state elections, the ruling BJP, had promised in its manifesto that it would press for Mitanins to be fully incorporated into the NHM and that their incentive would be hiked by 50%. Two months ago, thousands of Mitanins and trainers marched to the capital city of Raipur pushing that the promises made by the party in its manifesto be fulfilled.

However, not all unions participated in this protest–one led by the Mitanins was among these and its representative maintains that the movement should be led by workers, who form the majority of the workforce, and not by trainers and coordinators, fewer in number. While the movement has split into multiple factions, all of them are demanding that the entire system, as is, be incorporated with no mediating organisation involved.

Mitanins, Unlike ASHA

The 72,000 Mitanins (meaning a ‘woman friend’ in Chhattisgarhi) form a crucial link between the community and the state healthcare system. The programme envisioned these workers as the ‘spokespersons’ of communities and their task is to push for their right to health, and help them access healthcare and services. While the state government funded the programme, it created the SHRC to implement it and provide additional technical capacity to the health ministry.

In 2005, when the ASHA programme came into effect at the national level under the NHM, Mitanins were included as ASHA workers with the existing structure of governance and supervision.

The Mitanin programme is different from the ASHA programme in other ways too: unlike an ASHA who serves a population of 1000 population, a Mitanin is selected for every hamlet, irrespective of its population.

By the current projected population of 30 million, on an average, there is one Mitanin for every 400 people. Also, Mitanins are not expected to be literate as ASHAs are. A Mitanin must be a married woman from the local community who is selected by the Gram Sabha and endorsed by the local panchayat. An ASHA worker, by contrast, is expected to be a literate woman with formal education up to 8th class and is selected by the Gram Sabha from three candidates shortlisted by facilitators identified by the Block Nodal Officer.

Subscribe now

However, just like ASHA workers, the role of the Mitanins too has expanded considerably over time. But only two years ago, the state government started paying them a fixed honorarium of Rs 2,200 per month in addition to the task-based incentives from the NHM.

India’s 1 million workforce of female community health workers have been instrumental in improving access to healthcare for millions of people and saving lives. Despite denied fair wages and protections, these workers are cementing the healthcare system and ensuring care reaches those who need it most. To mark their 20 years of contribution, Behanbox has launched ‘The ASHA Story’, a special project documenting their lives, tracing the breadth of their work, the politics of their resistance, the economics of care work, and geographies of their rich inner lives.

Growing Workload

The Mitanin programme was introduced soon after Chhattisgarh achieved statehood in 2001 as the newly carved areas had half the number of sub-centres than required and 40% of the posts for Auxiliary Nurse Midwives (ANM) were vacant, according to T Sundararaman, the first director of SHRC who played a key role in designing the Mitanin programme. The purpose was to create a community health workforce but only as an interim measure, he added.

For the first three years of the programme, Mitanins were not compensated for their work for fears that financial incentives would give more control to the state’s bureaucracy and would also defeat the public spirit of the initiative–to improve health literacy and access to services through health activists. This was also meant to dissuade influential people in the community from interfering in the selection process. But the government also recognised that this practice could be exploitative and unsustainable and so the onus of payment shifted to the communities.

A November 2001 government order stated that ‘no honorarium or pay shall be paid to the Mitanin by the Government; however the village community itself can compensate her for her efforts by pooling money or grains’.

In 2006, after the launch of NHM, task-based incentives were introduced. These incentives changed the nature of Mitanins’ work. Activities with readily achievable targets like immunisation and sterilisation took precedence over promoting breastfeeding, maintaining cleanliness and sharing knowledge on better health practices.

“Mitanins were so successful that the government started roping them in for more programmes. Soon, they were given tasks for all national and state health schemes. While this improved Mitanins’ skillset, it also increased their workload,” said Chandrakant, national joint convenor of Jan Swasthya Abhiyan (JSA) from Chhattisgarh.

Share BehanBox’s Substack

While the earliest norms required Mitanins to work 8-10 hours a week, a 2020 representative study of 660 Mitanins found that they actually put in 24-33 hours every week, 25% of it on data collection and reporting. Only 20% of the time was spent on core tasks that provided most incentives–linking the community with health services.

As the workload increased, Mitanins unionised and demanded fair remuneration and other benefits. It was only in April 2023–seven months before the state elections–that the state government announced a fixed honorarium of Rs 2,200 per month in addition to the task-based incentives.

When the BJP government won the elections that year, it gave the Mitanins a reason to fight for its demands. “For over two decades, the SHRC had been operating the programme. But after the new government came into power, they said that Mitanins have complained to them against SHRC,” said Chandrakant.

There were multiple issues around SHRC, said Amulya Nidhi, also a joint convenor of JSA. “One of them was that now that the government has changed, they [the government] want to work directly on the programme and not through NGOs,” he said.

Lack of Clarity

The Mitanin programme was structured in a way to provide the workers with autonomy. All of the supervision and training is provided by Mitanin trainers and coordinators at block, district and state level, who are selected from the same cadre.

Additionally, some Mitanins are selected as Swasth Panchayat Samanwayaks (health panchayat coordinators) to provide support to Mitanins and work with the Village Health Sanitation and Nutrition Committees on social determinants of health and some are also selected as Help Desk Facilitators in district and block hospitals.

While the SHRC that operated the programme had representatives from the health department in its governing board, most of its members, including the chair, belonged to civil society organisations and did not come under direct control of the health department. This gave the SHRC a high degree of autonomy.

However, after the NHM was introduced and Mitanins began to receive task-based incentives, coordinators got the additional responsibility of verifying these tasks as well. This system brought about a dual structure of control- the NHM and the SHRC.

Under the NHM’s support mechanism for ASHAs, there is a provision for ASHA Resource Centres to be set up at a state level. “In Chhattisgarh, this centre was located in SHRC since the beginning of the programme and thus all supervisory cadres like coordinators were part of this resource centre and SHRC. The training, regular monitoring and facilitation of all Mitanin work was being undertaken by SHRC under the guidance of the Department of Health and NHM,” said Vikash R Keshari, the executive director of SHRC till March 2025

Now, once the NHM took over operations from SHRC, this structure was affected, creating confusion amidst the workers. “Every programme, every system has a structure. If you completely change that, then the whole system will be disrupted,” said Chandrakant.

As we mentioned earlier, since the transition, Mitanins and trainers are facing problems with their payments. “Earlier too payments have been irregular. Sometimes we would get every month or every 2-3 months but we always got what we were due,” said Suman Thakur, a Mitanin trainer and part of Mitanin Pratikshik Sangh, adding that the payment have been random and erratic in the last year since the NHM has taken over. “Some are getting Rs 3,600, others are getting Rs 8,000. We have no idea who is paying us and how they are calculating it,” she added.

Moreover, the transition has also created apprehensions about how the programme will be operated in the future. Will the NHM continue with the same existing structure or will it adopt the NHM structure of Mitanins and trainers reporting to ANMs? The government has not provided the workers with any clarity.

“During the transition last year, the government had verbally assured the SHRC staff that everyone would be incorporated into the NHM,” said Vikash.

But within months, the government put out a tender and hired Society for Action in Community Health, an NGO, as a partner agency of the State Health Systems Resource Centre (SHSRC) to operate the Mitanin programme. (SHSRC is distinct from SHRC, the latter is an organisation based on partnership between the government and civil society, SHSRCs are resource centres set up in each state under the health mission to provide technical support.)

While the Mitanins that Behanbox spoke to had knowledge of this, they are yet to receive clarity on the role and function of the NGO. Suman, a Mitanin, wondered if this meant that the workers would have to now report to the NGO.

Who Is In/Out?

Lamiya was one of the thousands of Mitanins who took part in the protest in August in Raipur. “Our demands are simple. We do all the work on the field but these contractors [meaning the NGOs] who sit in offices get paid more than us. We do not want the contract system anymore. When we are doing the work, we should get the money,” said Lamiya.

The hope, pointed out Amulya, was that working directly with the government meant scope for permanence.

The 32-day protest ended with the government promising a 50% increase in honorarium but the workers are yet to receive the increased amount. No decision has been made yet about complete integration into NHM. “We protested for 32 days but no one came to listen to us. The NHM director met us but told us that Mitanins cannot be included in the mission because many of us are illiterate,” said Suman.

“Both literate and illiterate Mitanins have been doing the same job so everyone should be included. So many women have spent 10-20 years doing this work, they should not be just cast aside. They should get their dues,” said Lamiya.

The current government is trying to take the Mitanin programme and mould it into the guidelines of the ASHA programme, said Chandrakant. “For the longest time, Chhattisgarh has not had the same party in power as that in the Union government. So each government has tried to strengthen this programme to showcase its work. But now that BJP is in power in the state, they want to just stick to the national scheme and implement that here,” he said, adding that this could mean dismissal of Mitanins and other support workers who don’t fall within the NHM guidelines.

“If they implement these criteria, 70-80% of Mitanins will not be eligible,” said Asha Vaishnav, state president of the Chhattisgarh Mitanin Jagruk Nagrik Manch.

This anxiety has fragmented the fight for regularisation. Asha and others from her union did not attend or support the protest led by Suman and Lamiya’s union. “There are two other groups, both led by trainers and block coordinators. There are 72,000 Mitanins but trainers and block coordinators combined make up for only 5,000 people. So, to build numbers, they have had to rope in many Mitanins. But if they give a letter to the government to regularise themselves, the Mitanins who do the actual ground work will be left behind,” said Asha, underscoring the need for a union for Mitanins to be led by Mitanins.

These differences and quarrels, said Chandrakant, only help the government push back against the workers’ campaign.

Leave a comment

In the early 2000s, Amazon executive Venky Harinarayan created the Amazon Mechanical Turk (AMT), the earliest model of global outsourcing of work. Its name was inspired by the 18th hoax of the Mechanical Turk, wherein a chess-playing machine was maneuvered by a human hidden inside. Amazon’s AMT too was marketed as automated but it was actually powered by armies of human labour performing unseen tasks.

The illusion persists even today. ChatGPT, Instagram, and the internet are not entirely the work of automatons. It takes labour – often invisible, always low-paid to make them work efficiently. A worker in India may be curating social media feeds for people living in Montreal, while another could be tagging images of disfigured people and diseased organs to train AI models to identify tumour locations. In Kenya, people work for less than $2 per hour (about Rs 175) to make ChatGPT less toxic. Elsewhere, people could be labelling data for a military software doing targeted killings in Gaza.

The lives of these data workers who power global tech supply chains are as much a mystery as the work itself. Described as “ghost workers” or “invisible slaves”, workers operate inside secretive supply chains and remain absent from policy conversations on gig work. Google last month fired more than 200 AI workers who made chatbots like Gemini intelligent, after they complained of poor pay and precarious work conditions.

Remote digital work has never been a part of the Indian platform work story, even though ‘crowdwork’ has existed globally for at least two decades, says Aditi Surie, an urban and platform work researcher at Indian Institute of Human Settlements. Complex AI systems – operating globally between suppliers, platforms, and human labour – are hard to trace, and harder to fit into national or international regulatory frameworks. Our understanding of gig work is also one that is visible, physical, mobile, and mostly performed by men. “This world of click work is very different from that of physical gig work,” Aditi says.

In this, the first part of a three-part series, BehanBox documents the nature and history of the “invisible” data work that makes possible the internet and AI and powers transnational economies.

The World of Digital Labour

Part-time online work that guarantees ‘easy money’ with ‘zero investment’ can be commonly found on online platforms such as YouTube, LinkedIn and Reddit. But there is no clarity on the specifics – the nature of the job, designation, working hours, payment system, or the end customer, says Priyam Vadaliya, a researcher at Aapti Institute, a digital think tank that has researched data work.

“You can do it from your home, from your mobile phone,” says an influencer in an auto-dubbed voice, promising Rs 1 lakh a month for part-time work. And what’s better, anyone can do it – a student, a fresher, a housewife. She points to third party ‘AI solutions’ platforms like Toloka for labelling data – which has also helped Russian surveillance companies train facial recognition software – and Appen for transcription. Another influencer has done a series on the ins and outs of being an MTurker in India. Scroll down the rabbit hole, and one finds a YouTube canon dedicated to the world of crowdwork.

Once crowdwork meant work done collaboratively but independently by a large number of people, but today, it represents work performed using digital labour platforms. The International Labour Organisation divides this work into three categories: physical gig work such as delivery services, work done on online web platforms, and that done remotely.

Labels such as “crowdworkers”, “microworkers”, or “online freelancers”, describe how often the work is done or what the employment relationship is. People doing repetitive data labelling are ‘microworkers’, while content moderators can both be online freelancers or microworkers depending on who employs them.

At any given moment, millions are estimated to be engaged in this liminal workspace. Here, work is broken into smaller, repetitive tasks – scrub social media, clean excel sheets or label audio/video to experience Instagram without glitch – and routed digitally to a dispersed labour force.

In India alone, roughly 1 million data annotators may be active by 2028, contributing to a data annotation sector projected to grow more than $8 billion globally. Other data work includes curating, labelling, content moderation, AI rating, transcribing, conducting surveys. If ‘AI is like a child’, as data annotators told Fifty Two the humans in the loop ensure it is trained to be precise.

An Invisible Industry

Production chains in AI data work operate like a digital factory. Big Tech companies own the product, a series of third-party vendors act as supervisors, and data workers repeat monotonous tasks to keep the assembly line running.

Workers, experts say, have little idea of who the final product goes to, or who they interact with at times. Third-party companies present themselves vaguely – as platforms doing “data solutions” work, offering opportunities to “contributors” or “freelancers” for tasks on a “variety of AI projects” that “contribute to a safer internet,” or “improve identification technologies”. Workers don’t know if the work they are doing is for Meta or TikTok because they report only to the contracted employees of some intermediary.

Applicants are hired to do ‘customer service’, but it is only after contracts are signed and training is done that they realise that this is about cleaning up somebody’s internet feed – with graphic or violent material sometimes – with no upfront disclosure, says Priyam.

Workers are also made to sign Non-Disclosure Agreements (NDAs), barring them from speaking about their work. When former BPO worker in the Philippines Renso Bajala described his job to the digital publication Rest of World, his employer accused him of breaching his NDA, he said in a recent panel discussion organised by the Tech Global Institute. Indian women working with AMT, Telus and Microworker also said they had signed NDAs and could not disclose specifics about their contract agreements.

Share

The Indian BPO sector has also caught up with the crowdwork model of work, says Aditi. Data labelling work, which previously happened on an AMT, is now being done in traditional BPO companies like Accenture that could, in turn, be doing it for Meta. “It’s no longer remote, but it’s still data work,” Aditi notes. These new models made it hard to classify new arrangements and pin accountability on an ‘employer’. Work conditions of a remote mTurker, as they are called, are not the same as one working for an Accenture, inside an office with clear managerial structures. Their contracts may call them ‘employees’, but who is the employer?

Third party vendors have also started “labour hedging” — hiring thousands of workers for ‘ghost jobs’ to pad their numbers and look scalable financially to win contracts from BigTech companies, while workers sit without pay and work for weeks, if not months. Even when work comes, wages are still not promised and regular. A client may refuse to pay for cleaning, say, an Excel sheet with 500 rows of data if one entry is wrong, but without grievance redressal systems, the worker has no bargaining power.

“It all comes down to how you see the worker,” says Priyam of Aapti. Tedious work performed in a vacuum has the effect of alienating workers and distancing Big Tech companies from their obligations.

In the film Humans In the Loop, Geeta Guha’s character Nehma is told: “AI bacche ki tarah hain, galat sikhayenge toh galat seekh jaayega (AI is like a child, if you teach it wrong it will learn wrong).” Nehma later is reluctant to label a caterpillar a ‘pest’ because she had seen it protect the plant. “A pest for one person is not a pest for another person… Is AI really a clean slate as people think it is, or is it sort of like a descendant of our biases, weaknesses and knowledge systems?” director Aranya Sahay said in an interview.

Finding Humans In the Loop

Nighat, a tech policy analyst at Aapti, says it is a challenge to document the size or shape of this industry. An oft-quoted Niti Aayog study puts the size of the gig workforce at about 8 million. And of this, nearly half a million are online platform workers, including home-based data annotation workers. “No measurement to that scale has come in again. It still doesn’t go down to the level of gender, caste, what types of tasks they do, what are the popular sectors or departments,” says Nighat.

This opacity makes any measurement of this industry nearly impossible, she adds.

India’s Periodic Labour Force Survey uses the National Classification of Occupations which accounts for “data entry clerks” in traditional roles, overlooking modern data workers and rendering them “statistically invisible”, noted researcher Neha Arya.

Researchers instead rely on proxies like industry growth statistics — NASSCOM data from 2019 show home-based workers in India are amongst the largest in number working for these global platforms. Over 90% of data annotators are also from Tier 2 and 3 cities, such as Ranchi, Shillong, and Vishakhapatanam. India’s supply of online labour also rose from 26% to 34% from 2018 to 2020.

The demand for crowdwork went down during Covid-19 as economies struggled but the supply – the number of people seeking to do crowd work – increased. This trend was pronounced particularly among women and young workers, said ILO economist Niall O’Higgins in an interview. The lockdown, and loss of traditional jobs, both increased an appetite for remote work. And livelihoods like nursing, childcare, elderly care, education, and sex work – which primarily engaged women and gender minorities – experienced disruptions, worker shortages or transitioned to online platforms. The platformisation of these feminised sectors normalised the idea of earning through digital interfaces, drawing many women into forms of precarious “clickwork” and data labour.

Share BehanBox’s Substack

An ILO consultation from 2023 also showed the median age of people is about 26, comprising mostly women working over smartphones and laptops, and for about 64% of them, microtasking was the main source of income. Many are graduates — half with STEM degrees — but turn to data work amid high unemployment, informalisation of labour, and the search for flexible options to balance caregiving. “Some may be new mothers who want part-time work to make additional income or who would want some mobility,” Aditi explains. Other women do this work for usual reasons: the lack of opportunities, commuting problems, safety or cultural barriers.

The ILO study added that workers initially believed the promise of co-creating with AI was attractive, but in reality, they found the jobs led nowhere, and demanded more of them physically and cognitively. There is a mismatch between education qualifications and the work women end up doing in particular, Aditi says. And as with traditional sources of employment like sewing, embroidery and so forth, women didn’t think of themselves as ‘workers’.

From Dotcom Boom To AI Labour

The story of data work begins before platforms, at a time of change. It was the dot-com boom of the 1990s. The world wide web came about, and became free and open in 1993; about 50,000 companies (funded publicly or by venture capitalists) rushed to commercialise the internet. Soon came giants like Yahoo, Amazon and eBay.

These newly established internet regimes are headquartered in the US and the Global North, looking to outsource work to countries with low-labour costs like India, Pakistan, and the Philippines. These destinations also have colonial histories with huge populations of English speakers and skilled workers to shoulder ‘customer service’. At the same time, liberalisation opened India’s economy to foreign capital. States offered tax exemptions and infrastructure offerings such as tech industry-related economic zones. Here, in crisis-ridden economies, emerged the Business Processing Outsourcing (BPO) sector, coordinating global diffusion of work at an equally grand scale.

By the early 2000s, India’s crises of mass unemployment and recession had given way for the information technology, telecommunications, and service (ItES) industries to take off. This also coincided with a rise in casualisation of labour – contract work increased from 12% of total manufacturing work in 1985 to 24% in 2008.

Between 1998 and 2005, a telecom and tech industry boom gripped India. Internet population – predicted to increase fivefold from 5 million users by 2005 – had reached at least 27 million users by some estimates (and 38.5 million by others). More users meant more data, which was plentiful and cheap for industries to mine and monetise. Data equaled profit, and companies began to design platforms to extract data, analyse consumer preferences, and offer targeted products in the early days of machine learning and AI systems.

AMT entered this landscape in 2005. From its very start, crowdwork platforms were aligned to perform tasks like transcribing or tagging images that were too complex for computers, but too voluminous for a traditional workforce. It was the problem that Amazon’s Venky Harinarayan wanted to solve at the turn of millennium, when the Amazon marketplace was expanding beyond books and basic products. Part of this process required painstaking cataloguing and removing thousands of duplicate products on its website. As we said, Harinarayan proposed the idea of solving this data processing problem through a “hybrid machine/human computing” arrangement that would farm out cataloguing tasks to a network of human workers. Jeff Bezos, when launching mTurk, called these workers “artificial artificial intelligence”.

“What AMT did was take global outsourcing work and put it in on a platform, so you’re able to do it from the comfort of your own home instead of having to go to somebody else’s office to do it,” says Nighat.

AMT then diversified into market research, data entry, data cleaning, and other low-wage tasks to support artificial intelligence. A wave of web-based platforms (like Telus, Clickworker, Microworkers) followed, trying to make smarter machines over time. Platforms enabled a new model of traditional ‘outsourcing’, giving way to crowdsourcing, where work was fragmented into micro-tasks distributed worldwide.

The result, researchers argue, is the rise of “digital sweatshops” and “global digital factories” without oversight and beyond geographies.

Vague Policies

In recent years, unions from Latin America and East Africa have ushered movements to bring digital labour industries to account. But Indian gig work policies are still occupied with visible, ‘on-demand’ forms of labour, where issues of poverty, vulnerability, physical harm, and discrimination are easier to track. The Code on Social Security, which defines work beyond traditional employer-employee relationships, differentiates between ‘gig work’ and ‘platform work’. But Aditi points out a problem.

“Within the current framework, there is potential to interpret a home-based worker to be an online worker… But in official communication from the Ministry of Labour and Employment, the current imagination of people who need social protection is physical gig workers,” she notes, partly because others lack collective organisation.

Several informal or feminised sectors have historically been hidden in plain sight. It took the Covid pandemic, for instance, for ASHA workers’ contribution to public health to become visible. Much like microtasks in data labelling, women engaged in garment industries also did work that was fragmented, invisible, remunerated on a per-piece basis, and operated in transnational economies. Their visibility grew when NGOs and unions documented exploitative supply chains.

More documentation could similarly help reveal the hidden reins of this industry in India, says Nighat. Seeking answers about who these workers are, where they live, and what they do would help tailor policies, from subsidised internet and maternity benefits to mental health support for content moderators exposed to violence.

In East Africa, worker collectivisation has created awareness at a global level, and people are slowly seeing the links of AI supply chains spread across countries like India too, says Priyam. “It will get to a point where you can’t ignore this work, and these workers, anymore.

Further reading:

1. This IT For Change study documents stories of women working with AMT during the pandemic.

2. A study linking the erasure of human labour with techno-utopian narratives of AI hype.

3. A participatory project where data workers from five continents report on their workplaces.

4. How Pulitzer Centre investigated the scale of human labour.

5. A report from Kenya about the experiences and demands of digital data workers.

Want to share experiences or questions about the world of data work? Write to us at contact@behanbox.com.

Leave a comment

A new tariff regime – ranging from 25% to 50% duties – imposed on Indian exports by the United States will have a devastating impact on key export driven, labour-intensive industries that employ a high volume of women workers.

The tariffs that came into effect from the midnight of August 27, while seeing a sharp 43% decline in exports– from the current USD 86.5 billion to USD 49.6 billion in 2026 — will impact the competitiveness and employment in the related sectors, as per a detailed sector wide analysis by the Global Trade research Initiative. These include textile, gems, auto parts, leather, agricultural products and marine exports, particularly shrimp.

At least 30% of India’s exports to the United States are in industries that rely heavily on women’s labour, often in low-paying and at the bottom of the supply chain, our analysis of export data revealed. In 2024 alone, India exported electrical, textile, leather and fish and marine products worth USD 23 billion (Rs 675 lakh crore) to the US, all of which will have significant proportions of women workers.

On July 31, US President Donald Trump announced a 25% “reciprocal” tariff on Indian goods exported to the US. On August 6, he passed another executive order, announcing an additional 25% “penalty” tariff tied to India’s continued purchase of Russian oil, bringing the total duty to 50%, one of the steepest levies, alongside Brazil, imposed on any trading partner.

India has called these tariffs “unfair, unjustified and unreasonable” and maintained that it will not “bow down” and asserted its right to protect national interests. Prime Minister Modi urged citizens to purchase only Indian products and has announced GST slab reforms as a measure aimed at bolstering domestic consumption. There are also plans to develop other international trading markets.

Share

Tariffs: A Feminist Issue

In his second term, Trump has introduced a sweeping global programme of tariffs to reduce the country’s trade deficit–the gap between the value of exports and imports. These include 15% tariffs on exports from Japan and South Korea to a high 50% tariffs on Indian and Brazilian exports.

Tariffs and taxation in general are a feminist issue, argue feminist economists and those advocating for tax justice. ‘National and international taxation are gender blind and tend to see taxation only from a functional revenue-raising potential rather than from a redistributive perspective,’ argues this brief from the Global Alliance for Tax Justice, a south-led global coalition for tax justice.

On the supply side, countries facing the highest reciprocal tariffs are major exporters of labour-intensive industries such as apparel and textiles to the US, where 80% of the workforce in the industry are women in low-paying jobs. The high tariffs are already forcing factories to shut down their US driven production lines, as we reported from Tiruppur and Dindigul in Tamil Nadu, disproportionately affecting women workers with reduced shifts and halved salaries, forcing them into debt and poverty.

From a consumer perspective, tariffs impact incomes of poorer households, as they tend to spend a larger proportion of their income on imported goods than those with higher incomes. Tariffs are considered regressive taxes, when applied uniformly, regardless of the price of goods or incomes of consumers. The Yale Budget Lab estimates that the present tariffs could lead to a 2% spike in inflation and cost American households between $3,400 and $4,200 per year. Since women make up 80% of all consumer purchases, are 35% more likely to live in poverty and represent two out of three of workers earning at or below minimum wage, and head 81% of single income households, any price increase will hit them harder, says this brief from the Feminist Majority Foundation. This is on top of an already existing “pink tax” where women’s versions of consumer products like razors, shampoos, clothes etc are priced higher than men’s items.

The Cato institute has dubbed this as the Trump administration’s most ‘anti-women’ economic policy that penalises goods and appliances like dishwashers that have freed women’s time.

Share BehanBox’s Substack

Tariffs And India’s Women Workers

India’s exports have seen a consistent increase– from USD 18 billion in 2007 to USD 80 billion in 2024, with a 65% increase in the post-pandemic years. (full dataset on Dataful here).

This increase has been largely driven by electronics and ready-made garments, which constituted 15% and 10% of the exports in 2024 respectively. While leather (0.9%) and fish products exports (2.4%) are comparatively lower by proportion, the US is a significant market for the export-driven industries. Over 80% of leather products in India are exported, close to 22% of which are sent to the US. Frozen shrimp accounts for 66% of India’s seafood exports with 34% of the revenue coming from the US market.

Women, often from the most marginalised communities, make up for a significant proportion of the workforce in these industries. The leather industry is one of India’s biggest non-farm employers with 4.5 million workers, 40% of whom are women.

In the frozen shrimp industry, women constitute over 70% of the 8 million workforce, and they perform low-end processing jobs such as deheading, peeling and sorting shrimp in cold processing plants. In apparel and textiles, they make up almost 70% of the 45 million workers, while the electronics industry is known to largely hire young women because electronics manufacturing needs “small and soft hands for small pieces”.

In these export-driven sectors that are part of a global supply chain and where labour conditions are often dire and exploitative, economic shocks like high tariffs will cascade into unforeseen short and long-term impact, say researchers and labour economists. This will push them into debt, lower their employment opportunities and reduce wages in the long run.

“Across global supply chains, the US tariffs will make women workers vulnerable. Women are concentrated in labour-intensive, low paying, precarious jobs, whether it is processing or assembly line and men dominate more capital-intensive, technical and leadership roles in these spaces,” said Ashmita Sharma, executive director at Society for Labour and Development, a labour rights and support organisation with a focus on global supply chains, labour migration and gender justice.

In these industries, women are paid on a per-piece basis or low monthly salaries with tight deadlines in conditions that have adverse health impact, as we have reported. In the leather industry, women are mostly home-based workers earning Rs 8-10 per pair of shoes. Typically, they earn Rs 1500- Rs 3000 a month after stitching around 8-10 shoes a day. Few work in factories and earn between Rs 7000-9,000.

Share

In the shrimp processing industry, a March 2024 report by the Corporate Accountability Lab unmasked the exploitative working conditions where most low paid women workers, often paid per kilogram of shrimp peeled or cleaned, are “ghost workers” with no contract, health insurance or social security.

Reduced demand from the US market due to high tariffs will lead to immediate cutback of vulnerable workers, particularly women, as men are engaged in capital-intensive roles like transport, machine operations and maintenance and handling of exports, said Ashmita.

Women in textile and electronics industries earn around Rs 10,000 and Rs 15,000 respectively while those processing shrimps are paid a monthly wage of around Rs 10,000 or daily wage rates of around Rs 300, depending on the location and the factory.

Impact of the Tariffs

While the full impact of tariff increases will be felt in the coming days, early reports show a developing slowdown in export driven industries and employment of workers, especially women.

The impact is visible in the most adverse ways in the textile industry. Tamil Nadu’s textile hubs, where the effective tariff for certain kinds of knitwear has reached 64%, our report showed how the costs to the producers are being pushed down the supply chain. Saraswati, a 35-year-old garment worker in Dindigul, and other workers in her factory were put on rotational shifts, i.e. three days a week, instead of the usual six days. Her wages were halved–from 12,000 a month to Rs 6,000. Companies are also delaying monthly wages, pushing women workers further into debt.

This came after months of women working overtime to complete the orders as the tariffs were initially announced on April 2, and then paused for a 90-day period. “Since there was so much uncertainty, a lot of the brands were asking the factories to fulfill their orders quickly,” said Deepika Rao, executive director of Cividep, a social impact organisation working in the labour space.

The shrimp processing industry – for which the US is the largest market – saw a 20% price drop in Andhra Pradesh, which accounts for a bulk of shrimp farming, after the 25% tariffs came into effect. Farm-gate prices in West Godavari and Ongole have plunged by Rs 40–60/kg for small-count shrimp. Processing plants in Visakhapatnam and West Godavari are cutting overtime and preparing for shift reductions as exporters report paused or cancelled orders, said the Global Trade Research Initiative report. In addition to the 50% tariff, shrimp from India also attracts a 2.49% anti-dumping duty and 5.77% countervailing duty, making effective tariff on the product over 58%.

Share BehanBox’s Substack

“Our importers from the US have asked us to hold on to our shipments even for existing orders,” said Alex K Ninan, vice president of the Seafood Exporters Association of India.

Importers are shifting to countries like Vietnam, Thailand and Indonesia whose exports are taxed at 19-20%, said Alex. “Other markets have also now realised that we are sitting on a huge inventory. So, in anticipation of [lower] prices, they have also asked us to hold our shipments,” he added.

In Agra, a leather footwear manufacturing hub, orders placed by American buyers for autumn and winters are in the final stages of production or ready for shipment. But buyers are telling the companies to hold shipments or share the losses. Many companies that had increased their capacity hoping to target US markets are effectively unsure of the future.

The impacts of these tariffs will depend on various factors, Deepika said, such as the product in question and the tariff imposed on it in competing countries. “They can differ from region to region and even factory to factory,” she said.

Alex said that seafood exporting companies have not yet taken any measures to layoff workers as they are waiting for relief measures from the government. No such measure has been announced yet.

Most export-oriented businesses maintain a small segment of permanent workers while outsourcing the rest of the work, said Neetha N, professor at the Centre for Women’s Development Studies (CWDS), an autonomous research institute. “So, when the demand for products reduces, companies will retrench these workers,” she pointed out.

The COVID-19 pandemic showed us how these shocks impact the most vulnerable workers. In Karnataka, textile factories illegally forced women workers to resign after a fall in orders, bypassing legal requirements to pay compensation or seek permission for retrenchment. For two months, 1,300 workers led sustained protests and submitted letters to the labour department and succeeded in receiving their dues of nearly Rs 45 lakh.

Weak Bargaining Power

Such collective bargaining, however, is rare and not without its challenges.

Since 2019, Ashmita’s organisation has been facilitating the formation of a women-led collective in the seafood processing industry of Gujarat’s Veraval. Alongside some resistance from employers, a key challenge lay in the limited space available for collective empowerment of women workers, as factories operated under stringent working conditions driven by international production requirements.

“Nearly 15-20% of the women workers are migrant workers who live in employer-provided accommodations and entry into these spaces is difficult,” said Ashmita. So, they started working with women workers who lived in villages around the processing unit and formed village-level committees, which fostered new ways of supporting worker voice and agency, rooted in principles of leadership, empathy and care. Despite caste dynamics between Koli, Kharva, Dalit and Muslim workers creating tensions, they have been able to create a shared space that brings different social groups together.

For home-based workers, such as those in the leather industry, the very nature of their work, which as we reported is not considered as work, reduces their collective bargaining power. “The smaller workshops and home-based units isolate women from freedom of association and collective bargaining. So then women workers become even more invisible, informalised and excluded from the more formal rest of the workforce,” said Ashmita

During economic shocks like the likely fallout of high tariffs, good social security coverage is essential, said Neetha, which will help workers tide through the impact, especially if the shock is short term and the industry is able to find its way back. Workers’ access to these social security systems are dependent on the level of formality and the specifics of the working conditions. While home-based workers do not get any of these benefits, formal workers in textile factories have more access.

With limited support systems, workers are compelled to choose more precarious and exploitative work or seek debt. “Women often have no financial cushion. They absorb the shock personally–by stretching food, taking up more informal jobs. That also pushes women into debt,” said Ashmita.

There are long-term repercussions too. “Fewer jobs leads to increased competition and will, in the long run, impact wage rates,” said Neetha.

As tariffs will impact 0.5% or 1% of India’s GDP by some estimates, the impact will be disproportionately felt by women, said Mitali Nikore, economist.

Leave a comment

Nisha Pawar’s phone is a hive of activity. Her practised fingers dart across the screen, flipping between multiple WhatsApp groups bringing messages from women working as beauticians and spa therapists on platforms like Urban Company and Yes Madam. There are also missed calls, voice notes, SOS messages, and meeting schedules.

On quieter days, Nisha, a leader at the Gig and Platform Service Workers’ Union (GIPSWU) in Maharashtra, acts as an administrator, accompanying complainants to the police station or labour commissioner’s office to file their grievance. At other times she’s a leader, explaining to women how commission slabs erode earnings and the strength of a collective or listening to their challenges. Graver days bring news of a suicide or death, and then she plays investigator, digging up medical records, tracking down family members, gathering evidence to establish that when a worker dies on the job, the platform must be held accountable.

Share

There’s always fear. In the last three years, alleges Nisha, Urban Company’s management has infiltrated these WhatsApp groups and blocked accounts of ‘partners’ found to be a part of unions. “The company wants to surveil every decision we make, every conversation we have, so that they can retaliate and dismantle efforts to build our union,” Nisha says. “We keep motivating the workers, easing their fears. We have to show them they are not alone.”

GIPSWU is one among the dozens of emerging unions and informal collectives supporting millions of scattered workers whose paths criss-cross in an increasingly fragmented and alienating digital economy. How do you build solidarity when there’s no physical workplace or time frame, when algorithms measure the worker’s worth each day, and where everyone is easily replaceable? Where can workers go when customers misbehave, algorithms dictate earnings, and companies block IDs? Should there be separate legislation for gig workers or should they be incorporated into existing labour laws?

In March, the International Trade Union Confederation called for the right to organise for all platform workers. A joint declaration by 33 groups urged the International Labour Conference to adopt a convention safeguarding their collective bargaining rights.

In India, platform companies do not recognise worker collectives or unions, per a Fairworks study, and India’s laws leave workers unambiguously defined and outside the scope of legislations such as the Trade Union Act, 1926. India’s cost of living crisis, high unemployment rate among young people, and the scattered nature of digital work gave platforms outsized power to control and repress workers, activists and experts told BehanBox.

“That’s why unions matter,” says Rikta Krishnaswamy, an organiser with Rajdhani App-Based Workers Union (formerly AIGWU). “We are in a neoliberal age, and with the weakening of the trade union movement, the platforms are shaping the dominant story – that this is flexible work, with lower entry barriers, and full independence. But none of this is true, the lived experience of workers contradicts these claims. We need class and caste organising, and for workers to build these movements to change the narrative.”

In a two-part series, we explore the unique challenges union activists face in creating worker solidarity in platform companies. In this, the first part, we look at the logistical, ideological, and also existential issues. Next, we will examine how gender complicated the challenge even further.

Share BehanBox’s Substack

‘Modern-Day, Digital Mills’

Workers have two fights on their hands – one for daily survival, and the other for long-term rights – while unions move between the two, says Ritika.

Manju, a leader at GIPSWU, begins her day at the union office where workers stream in to report problems – blocked IDs, opaque policies, arbitrary pay cuts, customer abuse, sexual harassment. “I try to be present most times so that I can meet them and register their case,” she says. Other times she meets workers closer to their station hubs.

Manju points out the advantage of a collective body – it can talk to companies, labour commissioners, the National Human Rights Commission – things a single worker cannot do. “When we escalate an issue and [for instance] get an ID unblocked, the workers see that unions are effective. That trust helps,” she says.

Union work rises and falls with the rhythms of workers’ lives. On slow days, they conduct drives on making workers aware about government schemes and plan how to mobilise new groups. “It is hard to mobilise people – they are always in a rush,” Manju says. “We show up at their working points directly and then tell them about meetings. If we try to mobilise 200 people, at least 50 show up.” Sometimes workers turn off their digital IDs to show up to meetings – they say hamare haq ki baat ho rahi hai. (it is our rights that are being discussed).

In gig work companies, workers are labelled “partners”, “independent agents”, “contractors”, a misclassification that strips them of labour rights. Manju tells people about what rights exist on paper and what are missing. “Ab unko ehsaas ho gaya hai ki ab hum kya hain, kaun hain (they are more aware of their identity as gig workers),” she says.

Karnataka, last month, passed a historic legislation to ensure the rights of gig workers and hold platforms or aggregator companies accountable for workers’ occupational health and safety. News of gig worker legislations proposed in states like Rajasthan and Telangana gives workers the hope that one day there may be similar laws for the entire country. Similar discussions are underway in Jharkhand and Maharashtra.

Informal, contractual work has always existed in India, but app networks are the new factory floors. Aggregator companies present themselves as digital market places, where workers are independent agents, but Rikta says they function more like a single, centralised employer, a modern-day mill. Workers rely on these platforms for livelihoods, follow strict controls and supervisions, and spend most of their day performing tasks assigned by the app – four conditions Indian courts have previously used to establish an employer-employee relationship in informal work.

“Our agenda is to tell the workers this is the same work but in a new form, to go to them and map the long fight – and show how small demands and scattered strikes can scale into a larger movement,” she says.

In GIPSWU meetings, discussions shuffle between learning little tricks of the trade, an attempt to negotiate fairer commission or reduced penalties, and how to listen. “A lot of times women don’t speak up against violence. So it is important to hear them no matter how tenable a solution is,” says Manju. Women who go unheard by their families, communities and employers are particularly vulnerable.

“Imagine the cognitive dissonance of an Urban Company worker,” Rikta says. “They’re highly skilled, experienced, called a business partner, and then they’re treated like shit in every aspect of the way.” The union is a space where these contradictions are acknowledged.

Subscribe now

A Young, Unique Movement

The unionising landscape is bustling – some like Telangana Gig and Platform Workers’ Union (TGPWU) are as new as the platform economy; others like the All India Trade Union Congress (AITUC) predate the independence movement. At least 18 collectives are organising across different sectors like transport, delivery, and home-based work.

The Covid epidemic was a catalyst for this burst of union activity, activists say. In the early 2010s, disruptor transport platforms like Ola and Uber had emerged, riding the idyllic wave of digitisation and flexible work arrangements for workers. “It took three-four years for workers to get out of that honeymoon period,” says Aditya Ray, an urban studies and gig work scholar.

TGPWU’s founder Shaik Salauddin led some of the first protests – over worker deaths, exploitative commissions, and the disappointment of drivers who had migrated from traditional cab services.

The collapsing, post-pandemic job market pushed people onto platform work, and the demand for services such as delivery and cleaning saw an uptick. At the same time, those already in this sector felt stifled by unfair pay reductions, unreasonable work hours, and client harassment. This was also the time when the Bharatiya Janata Party-led central government introduced four new contentious labour codes covering wages, industrial disputes, social security, and occupational safety and health. Ambiguous on the definition of a “gig worker”, they excluded millions from social security protections.

As workers buoyed up to protest, platforms got better at suppressing them – union workers speak of blocked IDs, unfair terminations, FIRs, and union busting tactics. “Platforms became more powerful and started showing their real faces after Covid-19,” says Nisha.

Share

The result is a “unique” trajectory of platform unionism, noted Aju John and Aditya Ray in this paper. It closely mirrors that of construction workers’ struggles – beginning with demands for formal recognition, then negotiating strategic compromises for welfare through dedicated legislation.

While participation in the unions has increased in the last five years, these spaces represent less than 5% of app-based workers in India, says Apoorva, president of App-Based Karamchari Ekta Union (AKEM) which represents delivery workers in New Delhi.

‘We Can’t Keep Striking Again and Again’

The post-pandemic collective action has emerged in short, scattered bursts, a factor of how the platform economy is designed, to trap workers in a ‘digital cage’, activists say.

Bijender, 27, joined Swiggy Instamart during the 2020 lockdown, after his father was diagnosed with cancer. His enthusiasm ebbed as the company arbitrarily revised rate cards and earnings became unpredictable. When a revised rate card was quietly introduced last year, Bijendar rallied fellow delivery workers, from Swiggy, Zepto and BlinkIt, shutting down 10 stores in protest.

“It helped that I wasn’t alone, then they can’t block everyone’s IDs,” he says. The strike worked – briefly. After 15 days, the company quietly rolled back its promises. “What could we do? They know we can’t keep striking again and again,” Bijender says.

This is a dispersed sector, with a high turnover of workers who are everywhere but nowhere in particular, floating between apps, creating a churn too rapid for organising. “The traditional idea of going on strike goes out the window,” says Nihira Ram, who has organised with AIGWU earlier. “You need people to come out in numbers, but how do you get, say, Swiggy workers across the country to mobilise on a particular day?”

Unlike factory workers in 1990s Delhi, who gathered at mill gates, or Accredited Social Health Activist (ASHA) workers who meet at health centres, platform workers are dispersed digitally and geographically. Their “workplace” is the street, a client’s home or at the doors of restaurants. And they are rarely at the same place often enough.

This year, contract workers at Maruti Suzuki staged a successful protest in Manesar for fair pay and dignified working conditions because there was a site of work. “They could go again and again to the manager’s office, corner managers who had nowhere else to run. Workers were able to become an annoyance – and an important part of any labour movement is to become a pain in the ass,” says Nihira.

In 2021, women Urban Company workers staged a protest in front of the Gurugram and Bengaluru offices, after which the management agreed to improve working conditions. But since then, Nihira says, many physical offices have shut down and workers struggle to get a hold of the management.

Subscribe now

Fighting Invisible Capital

While platform work continues in the tradition of informal labour like agriculture or construction – where legislations do not apply and where unions have struggled to find a foothold – it introduces a fresh challenge, says Aju John, a scholar of platform work and researcher at Humboldt University, Berlin.

“Informal workers fought against local factories and entities, but workers right now are fighting multinational billion dollar companies backed by hedge funds and venture capitalists,” he says.

Companies are also getting more smart about strategies to preempt collective action. Strikes would earlier happen when companies announced a new rate card price, points out researcher Ambika Tandon, citing the example of the Delhi-NCR wide strike by Swiggy workers. “Now rate cards are not announced but implemented in a phased way. Targets and incentives change weekly, and you don’t know how much you will earn next week.”

Workers are getting increasingly squeezed for time, says Bijendar. If 20 riders handled 100 orders, now 50 riders share the same area, each earning less. “We used to make Rs 300 in two hours. Now it takes four to do the same. The company has slowly taken away our spare time,” he says.

The lack of traditional job opportunities has created an almost inexhaustible source of labour that is educated and unemployed. Companies can no longer cope with the demand for gig work. A 2023 study found more than a third of food delivery workers are educated, and another, focusing on women data workers, showed 94% had a bachelor’s degree and 43% workers held STEM qualifications.

This labour pool is tempted by the promise of freedom in gig work and associated schemes like Zepto’s Rural Mobilisation Programme which guarantee earnings of Rs 35,000 a month. “I couldn’t find any work in Kanpur, so I came to Delhi for the RMP programme,” says Gopal*. He left less than a month later due to poor pay and living conditions and sheer exhaustion. Now he hopes to find an office 9-5 job.

Apoorva believes that the wages are so depressed that workers are doing double shifts and still not able to earn one single shift’s minimum wage. In Delhi, many AKEM workers are Bengali and Bihari immigrant migrants living in Jai Hind camp in South Delhi or JJ clusters in North Delhi, juggling multiple jobs as DTDC workers in the day and delivery riders at night.

This kind of work schedule leaves little time for participation in union activities.

Women are famously time poor, forever juggling unpaid domestic duties and care work. A majority of women in platform jobs are single mothers or the sole earners of the family, and also socially isolated.

“Women are always rushing – from client to client, then running back home to look after the children, then back to work. You try to speak to them and they say they will get late – either for the customer, or for the husband or mother in law at home,” Nisha says.

Subverting Resistance

Platform companies, say labour organisers, wield technological and structural tools that make sustained resistance nearly impossible. These include ID blocks, permanent terminations, transfers and false FIR threats, says Ambika Tandon, a gender and platform work researcher.

For 32 days between May and June, Mahima*, 42, pleaded with Urban Company to unblock her ID. The company alleged Mahima had “defamed” them by sharing a story about a customer who verbally harassed her, and suspected her of union activity. “I told them I want to work here until my retirement. But they didn’t listen to me,” she says.

In some places, resistance was met with violence. In New Delhi’s Prashant Vihar, a BlinkIt store manager reportedly turned off CCTV cameras before assaulting five protesting delivery workers in the presence of police personnel. In Malviya Nagar, Zepto workers striking over reduced payouts were detained and allegedly beaten by officers.

Mahima was allowed to return to the company on two conditions: she would not speak about the harassment, and she had to give up the names of people who she told. She shared the incident with her colleague and friend, later summoned for questioning by the company. “But who else can we talk to about bad customer behaviour except each other?” she says. Mahima’s relationship with her friend soured, but she says she had no choice – she had loans to pay and children to raise.

ID blocks and onboarding policies are profit-making tools, activists allege. Each new recruit pays up to Rs 1 lakh in onboarding and training fee, and it is in the interest of the company to keep blocking, firing, and hiring new people.

In 2023, when a group of women protested against Urban Company, their IDs were restored only after they signed undertakings promising to never protest again. In informal WhatsApp groups, workers were coaxed into adding management members to spot and dismantle union activity, allege activists.

These punitive actions against workers who join unions suggest an active resistance to worker collectivisation, says scholar Sai Amulya Komarraju.

Unlike traditional employment settings, platform companies have no grievance redressal systems, no arbitration, no transparency, or a notice before suspension. As we reported earlier, the labour codes refuse to register cases due in this sector because of the absence of any legal contract proving the employer-employee relationship. The proposed state-level protections in states would demand companies to follow a proper procedure in order to fire or penalise workers for participating in the strike. But legislations like Telangana’s are yet to be implemented, and experts worry they will be ineffective due to weak enforcement and lobbying.

Shaik Salauddin says unions are fighting both invisibility and fear of the algorithm: “We’re pushing for algorithmic transparency as a core demand.”

An analysis by RAWU of ride-hailing and delivery sectors showed workers don’t have access to data – and subsequently proof of unfair pay – about their past earnings or the rate cards, “which is required to understand wage changes and raise their voices against unfair pay and algorithmic volatility”. But, says Chandan, workers are now beginning to understand data.

Labour used to be a tripartite issue discussed between governments, employer bodies, and worker representatives, but has been reduced to a bipartite issue especially after 2014, and worker representatives are sidelined in conversations, activists note.

The biggest challenge is psychological, Rikta says. The company depersonalises the union as an external entity. The right to withhold labour – once a core tool of protest – is being turned against the worker. “It is companies that force you to stop your labour; there are 20,000 other people ready to replace you,” she says.

Leave a comment