OPINION | AI’s social costs India cannot ignore

AI thrives in systems that are clean, digitised and interoperable.

Climate change, once the dominant policy drumbeat, has quietly slipped from dinner table debates and boardroom strategy sessions. In its place stands artificial intelligence, spoken about with equal parts awe and casual certainty. The shift has been so swift that it should worry us.

Our own misplaced sense suggests AI will solve everything from productivity gaps to climate action. That is comforting marketing. It is also dangerously incomplete. Anyone who has spent time with climate-tech founders knows the harder truth. Building scalable products is only the first mountain. Driving adoption across fragmented Indian markets is the Himalayan range beyond it. Technology announcements are easy. Behavioural change across a billion people is not. Of course, eventually a climate-tech using AI could possibly be deployed at scale.

The Boardroom AI Fever

In many high-powered boardrooms today, every strategy conversation carries a fashionable suffix. Everything is suddenly something dot AI. The enthusiasm is understandable. Even executives who privately admit they are still learning the technology liberally deploy jargon in discussions that should demand sober trade-off analysis. AI will solve A, B and C, we are told. Rarely does the conversation dwell on what it may simultaneously unsettle.

This omission is not trivial. Commercially available combinations of automation, IoT and AI are already capable of running large manufacturing facilities with single-digit human intervention. The technology exists. The economics increasingly work. The real question is political economy.

Are promoters and boards of India’s large listed manufacturing companies prepared to absorb the social backlash if such efficiencies translate into large-scale workforce rationalisation? Would any brand willingly trade decades of goodwill for a near-term margin expansion driven by labour compression? The technology debate is incomplete without confronting this human arithmetic.

A Young Nation in an AI Economy

India’s AI conversation is also occurring in a country that is not yet affluent. We have islands of extraordinary wealth that often distort the macro narrative. But the lived reality is more complex. A nation of over 1.4 billion people with a massive youth bulge cannot afford to treat labour displacement as an abstract future risk.

More than 800 million Indians are under the age of 35. They will remain economically active for decades. The central policy question is brutally simple. Do we have enough jobs, enterprises and productivity pathways ready for them in an AI-heavy economy?

Signals from recent public discourse are mixed and instructive. Industry bodies note that the technology sector continues to expand and may cross 315 billion dollars in revenue, even as firms embed AI into service delivery. At the same time, others caution that AI will certainly disrupt India’s services model even if it does not entirely derail it. Leaders within the outsourcing industry argue the shift will create new roles rather than eliminate them.

Each of these views contains a slice of truth. Together they reveal the policy fog. AI will create jobs. AI will destroy some jobs. AI will change the nature of most jobs. The distribution of pain and gain, however, will not be symmetrical. That is where India’s real vulnerability lies.

The Social Arithmetic of Disruption

The social costs of an accelerated AI transition in India should be central to policy design. Unlike advanced economies with deeper safety nets, India’s labour market remains highly informal, geographically uneven and socially sensitive.

Even modest displacement in sectors such as entry-level IT services, back-office processing, retail intermediation or routine manufacturing could have multiplier effects on household incomes, urban migration patterns and consumption demand. The risk is not merely job loss in a narrow sense, but the quiet hollowing out of first-rung employment that has historically enabled social mobility for millions of young Indians.

If efficiency gains accrue faster than new opportunity creation, the country could face a widening gap between headline productivity and lived economic security. That is the social arithmetic policymakers must now confront with urgency rather than optimism.

Even within government thinking there is visible caution. The Economic Survey has warned against both exaggerated fears and naïve optimism, noting the uncertainty around AI’s labour market impact. That uncertainty should trigger policy urgency.

Unprepared Skills and Institutional Pipeline

Consider the education and skills pipeline. India continues to struggle with full-scale execution of the National Education Policy. Curriculum reform, teacher readiness, digital infrastructure and state-level implementation remain uneven. Yet the AI narrative assumes a workforce that will seamlessly upskill into higher-order roles.

Meanwhile, early tremors are already visible in knowledge work. Traditional IT services and consulting models are under pressure from generative AI tools that compress effort hours. Even industry leaders acknowledge productivity gains when AI is systematically deployed.

That efficiency dividend will not remain neutral. It will reshape hiring patterns, billing models and career ladders across the white-collar economy.

The old guard in many sectors still behaves as if this is a cyclical technology wave that can be absorbed through incremental adaptation. That reading may prove dangerously complacent. Globally, companies are already restructuring around AI-native operating models, sometimes with significant workforce reductions.

There is another uncomfortable dimension that deserves attention. India’s ease-of-doing-business narrative has improved on paper, but everyday frictions remain stubborn in many parts of the economy. Urban civic infrastructure continues to lag aspiration in most cities. Administrative capacity is uneven. Regulatory co-ordination across sectors is still evolving.

AI thrives in systems that are clean, digitised and interoperable. It struggles in environments where data quality, process discipline and institutional trust are weak. Before celebrating AI-led transformation, policymakers must confront these foundational deficits.

Policy Priorities for an AI Transition

AI cannot be treated as a standalone technology policy. It must be embedded within a broader employment, education and industrial strategy.

Three policy shifts are urgent.

First, India needs a national workforce transition framework that goes beyond generic skilling slogans. This must include sector-specific job mapping, mid-career reskilling pathways, portable social security mechanisms and incentives for companies that demonstrably redeploy labour rather than simply eliminate it. Without such guardrails, the adjustment burden will fall disproportionately on younger workers entering the labour market.

Second, regulatory capacity must catch up with technological ambition. AI deployment in finance, healthcare, mobility and public services will raise complex questions around accountability, data governance and systemic risk. Fragmented oversight will create both innovation bottlenecks and trust deficits. India needs co-ordinated regulatory sandboxes that allow experimentation while maintaining institutional credibility.

Third, the country must invest seriously in broad-based human capital rather than narrow elite capability. Sovereign AI models and frontier research are important. But the larger economic dividend will come from millions of workers productively augmenting their roles with AI tools. That requires foundational digital literacy, domain depth and continuous learning ecosystems that extend beyond metropolitan India.

The deeper risk is psychological. Nations often fall in love with the promise of new technologies while underestimating the discipline required to absorb them. India has seen this movie before with earlier waves of digitisation and infrastructure build-out. But policy must now focus on the future that is arriving faster than our comfort allows.

The youth bulge will not wait. Neither will the algorithms. 

(Srinath Sridharan is Author, Policy Researcher & Corporate Advisor, Twitter: @ssmumbai.)

Views are personal, and do not represent the stand of this publication.