How AI exposure varies across Indian states: Kerala leads, UP and Bihar lag behind | Mint

TEXT START: Kerala, Haryana, Maharashtra and Tamil Nadu have the highest shares of workers in occupations where artificial intelligence (AI) is more likely to improve productivity than replace human labour, while Uttar Pradesh, Bihar, Jharkhand and Madhya Pradesh face a double disadvantage of lower AI exposure and fewer opportunities to benefit from the technology, according to a new study by the Great Lakes Institute of Management.

The Dissection

The article maps the geography of AI exposure, then disguises a structural rupture as a state-capacity and reskilling problem. It treats “complementary” occupations as a favorable destination, although they are often the first stage of automation: AI raises output per worker, after which firms reduce headcount, compress wages, and capture the gains.

Kerala and the other high-exposure states are not protected. They possess more valuable cognitive labor that is easier to target. Uttar Pradesh, Bihar, Jharkhand and Madhya Pradesh are not safe either; their lower exposure reflects exclusion from productive, high-value work and weaker routes into the AI rent economy.

The Core Fallacy

The article mistakes current complementarity for durable human necessity. Judgement, expertise, responsibility and human interaction are not permanent barriers. They are temporary constraints—technical, legal, institutional or cultural—that AI deployment and competitive pressure will steadily erode.

The report also assumes that productivity gains accrue to workers. Under the Discontinuity Thesis, they accrue primarily to owners and controllers of AI capital. AI need not eliminate an entire occupation to break the wage-to-consumption circuit. Automating enough tasks can destroy entry-level pathways, reduce staffing, weaken bargaining power and make the remaining workers cheaper to replace.

This is P1 advancing toward P2 and P3: cognitive automation becomes superior, institutions fail to preserve human-only economic domains, and the majority lose access to economically necessary labor.

Hidden Assumptions

• Complementary work will remain complementary rather than becoming a temporary staging ground for substitution.
• Higher productivity will be distributed through wages instead of captured by firms and capital owners.
• Reskilling will create enough new jobs rather than merely produce a larger queue of credentialed applicants.
• Healthcare, education, elderly care and skilled trades are permanent human moats because they involve interaction or physical presence.
• Workplace redesign and AI literacy can outrun competitive pressure to reduce labor costs.
• State-level policy can alter the ownership and control of AI, despite the article never addressing either.
• Lower AI exposure is primarily a missed opportunity rather than evidence of low-productivity employment, weak bargaining power and deeper economic fragility.
• Aggregate occupational categories accurately describe task-level exposure and future deployment.

Social Function

This is partial truth functioning as transition management and ideological anesthetic. The data may usefully identify where disruption arrives first, but the framing redirects attention from ownership and surplus capture toward training, literacy and job redesign.

It tells institutions to prepare workers for AI while leaving the central question untouched: who owns the systems that make workers less necessary? That omission converts a distributional crisis into a workforce-preparation agenda and gives policymakers a respectable vocabulary for managing decline.

The Verdict

The article is a credible map of differential exposure but a weak theory of survival. Kerala, Haryana, Maharashtra and Tamil Nadu are closer to the blast radius, not closer to safety. Bihar, Uttar Pradesh, Jharkhand and Madhya Pradesh are not insulated; they are already nearer the economic basement, with fewer productive footholds when the transition accelerates.

Its proposed answer—reskilling and expanding human-facing sectors—cannot preserve mass productive participation under P1–P3. Without control of AI capital and distribution of its output, “complementary work” is temporary hospice care for the wage system.