Artificial intelligence may be lifting productivity across Indian enterprises, but its costs are not being shared evenly, and the country’s entry-level workforce could bear the sharpest edge of the shift. That is the central warning of a new whitepaper from the Centre for Responsible AI (CeRAI) at the Indian Institute of Technology Madras.

Titled “How AI Is Reconfiguring Work in India: A Task-Based Framework for Understanding Augmentation, Displacement, New Roles and the Changing Structure of Jobs”, the paper moves away from the blunt question of whether AI will “take jobs”. Instead, it breaks jobs down into individual tasks and examines which of those tasks AI augments, which it displaces, and where entirely new roles emerge.

The entry-level pipeline is the weakest link

“The entry-level career pipeline is especially at risk,” the authors note. Junior roles have historically worked as informal apprenticeships: routine work such as password resets, basic contract reviews or boilerplate coding gave freshers the reps they needed to build judgement for senior positions. If AI absorbs that routine layer, the paper asks, how will new employees ever acquire that experience?

The question lands hardest on the sectors that have long been India’s largest graduate recruiters: IT services, business process outsourcing (BPO), financial services and professional services. One option the paper outlines is redesigning junior roles around supervising and correcting AI output rather than doing the routine work itself. Whether companies will actually redesign career paths, or simply hire fewer juniors, remains an open question.

Uneven adoption: big firms versus MSMEs

The whitepaper maps clear divides in how AI’s impact plays out: between entry-level and experienced staff, formal and informal workers, urban and rural workers, and between large companies and micro, small and medium enterprises (MSMEs). Large firms can absorb the fixed costs of integrating and customising advanced models, while MSMEs tend to rely on narrower, often subsidised tools.

Language is another fault line. The paper observes that Indian-language AI still performs less well on nuanced and code-switched conversations, which puts regional-language work at a disadvantage compared with English-language workflows.

Automation is also creating new kinds of jobs, from robot operators to AI escalation specialists. But the authors caution that current evidence does not show whether these roles offer stable employment or piecework, a distinction that decides how much of AI’s productivity dividend actually reaches workers.

Surveillance, stress and the job-quality blind spot

Perhaps the most candid admission in the paper is that its framework is “largely silent” on job quality, because Indian evidence on it is thin. It flags wages, workplace monitoring, work intensity, stress and burnout, career progression, the quality of human interaction and accountability for errors as areas that urgently need study.

The authors raise pointed scenarios: BPO agents being scored on the harder calls left over once AI handles routine queries, MSME shop-floor workers tracked through productivity dashboards, and employees left managing a more demanding stream of exceptions after automation strips away lighter tasks.

The informal workforce is still largely unmeasured

India’s informal sector employs far more people than the formal jobs the framework examines in detail, and the paper acknowledges this gap. AI could still reach informal workers indirectly, through prices, consumer demand, platform management and business survival. Job losses in formal IT and BPO hubs, for instance, could reduce the local spending that sustains informal service providers. AI-based credit scoring could widen access to formal credit for some, while shutting out those without a digital transaction trail.

The paper does not attempt to quantify these effects. Instead, it calls for India-specific research on AI’s labour-market impact and for structured pathways that help workers transition as their jobs change.

Why it matters for CIOs and enterprise leaders

For technology leaders scaling generative and agentic AI, the CeRAI framework is a useful reminder that responsible AI is not only about model bias or data privacy. It is also about workforce design: how talent pipelines are rebuilt, how AI-driven monitoring is governed, and how productivity gains are shared. As Indian enterprises accelerate AI deployment, these questions are likely to move from research papers into boardroom agendas, and into the national AI governance conversation.

Be a part of Elets Collaborative Initiatives. Join Us for Upcoming Events and explore business opportunities. Like us on Facebook , connect with us on LinkedIn and follow us on Twitter.

"Exciting news! Elets technomedia is now on WhatsApp Channels Subscribe today by clicking the link and stay updated with the latest insights!" Click here!