AI now shapes hiring at 85% of Indian firms. Only 35% report new early-career roles or skill needs

TEXT START: About 85 per cent of Indian organisations say AI or agentic AI is transforming their hiring processes in FY26, up from about 79 per cent a year earlier.

The Dissection

The text documents the first visible layer of automation: AI is tightening the employment gate before it has visibly erased the job titles behind it. Recruitment becomes cheaper, faster and more selective while graduate roles remain superficially intact. The article correctly identifies the crucial asymmetry: companies are automating access to work faster than they are redesigning the work itself.

But its framing stops at the threshold. It treats the transformed funnel as a market “re-sorting” rather than an early mechanism of labor displacement. More screening capacity, internship filtering and precise matching allow firms to process more candidates while committing to fewer marginal workers. The gatekeeper is not merely changing its criteria; it is acquiring the power to deny entry at machine scale.

The Core Fallacy

The central error is measuring disruption by job-title creation or immediate headcount retreat. Under the Discontinuity Thesis, AI does not need to eliminate the title “business analyst” to destroy the graduate labor bargain. It only needs to let one analyst, supported by AI, perform the work formerly distributed across several entrants.

The article mistakes continued hiring, rising recruitment budgets and improved retention for evidence of durable demand. Those figures can coexist with shrinking labor intensity. A company may spend more to identify a smaller number of higher-yield workers. That is not an expanding employment system; it is a more efficient sorting machine.

The 85-to-35 gap is therefore not reassuring. It is the warning. The selection infrastructure is scaling before replacement becomes legible in official role categories. Human institutions record the corpse by occupation title long after the economic function has been automated.

Hidden Assumptions

• That if early-career roles still exist, productive participation remains broadly available.
• That unchanged job titles imply unchanged labor demand.
• That increased campus budgets mean increased opportunity rather than more expensive exclusion.
• That internships and selective conversion preserve a pathway into work, rather than turning the pathway into an unpaid or underpaid audition funnel.
• That AI literacy and specialized skills will diffuse opportunity instead of concentrating it among a narrow technical elite.
• That credential hierarchies can remain meaningful as the underlying volume of entry-level work contracts.
• That low single-digit campus pay growth is a normal compensation lag rather than evidence that workers are being asked to acquire more capability without receiving proportional bargaining power.
• That firms will continue hiring graduates at roughly the same structural scale once AI-enhanced incumbents and smaller elite teams can absorb more output.
• That “new skill requirements” represent new productive demand, when they may simply represent additional screening requirements for a shrinking number of seats.

Social Function

Classification: partial truth, transition management and ideological anesthetic.

The article performs a useful diagnostic act by admitting that the gate is changing faster than the job. That is the partial truth. Its anesthetic function begins when this asymmetry is presented as a “new entry-level bargain” and as re-sorting rather than as the opening phase of productive participation collapse.

It gives institutions a tolerable narrative: hiring is becoming smarter, skills are evolving, retention is improving, and talent is merely being differentiated. This language converts exclusion into precision and labor contraction into better matching. The young worker is told to become more capable while the system quietly reduces the number of economically necessary beginners.

The credential data exposes the harsher reality. AI is being layered onto an existing hierarchy, not democratizing access. The likely result is a narrower Sovereign-adjacent elite, a smaller pool of indispensable Servitors, and a widening population screened out before employment becomes possible.

The Verdict

This is a competent early-warning report wearing the clothes of a labor-market adjustment story. It correctly sees that AI first captures the gate, but fails to follow the mechanism to its terminal implication: once firms can evaluate and augment workers more cheaply, the number of workers worth admitting declines.

The graduate job has not escaped automation. It is being surrounded by it. The apparent stability of entry-level employment is lag-weighted evidence, not structural safety. The gatekeeper is becoming AI-native because the system is preparing to need fewer people on the other side.