Jevons Paradox suggests AI is an upgrade, not an extinction event, for Indian IT

For over a year, Indian IT’s titans have sold a comforting narrative: that AI is a tool, not a terminator. Infosys Ltd. Chairman Nandan Nilekani assured investors of a “new wave of growth,” while Tata Group’s N. Chandrasekaran positioned TCS to lead an “AI-led” future.

But last week, the markets challenged this poise. The catalyst was the arrival of Anthropic PBC’s new plugins that execute multi-step professional workflows, and Palantir’s claim that AI can compress years-long ERP (enterprise resource planning) migrations into weeks.

As US software stocks shed billions, and the Nifty IT index plummeted, it looked like the “all is well” illusion of Indian IT had finally cracked. The fear is that the $250-billion services model, built on a pyramid of billable hours and entry-level freshers, is facing an extinction event.

Yet, this market panic feels like a classic overreaction, one we have seen too often before.

To understand why Indian IT might actually thrive in this new environment, one just needs to look at Jevons Paradox. Historically, when the cost of a resource drops, demand doesn’t shrink; it explodes.

• In the 19th-century coal industry, James Watt’s efficient steam engine made power so cheap that coal consumption tripled as factories scaled up.
• The aviation industry followed suit: as jet engines became 75% more fuel-efficient, ticket prices plummeted, triggering a huge surge in global flying.
• Even LED lighting fell into the same cycle as ultra-efficient bulbs made light so inexpensive that we simply used more illumination.

In every instance, the savings were immediately reinvested into higher consumption.

As AI makes building systems cheaper and faster, global enterprises won’t stop buying. Instead, they will finally greenlight massive projects that were previously too expensive to touch. Indian IT isn’t looking at a shrinking pie, but a vastly larger one that requires massive scale to manage.

Furthermore, the agentic revolution assumes a level of autonomy that may not survive reality of a regulated boardroom. The dream of unsupervised agents accomplishing complex tasks without handholding remains a demo-day fantasy. In the enterprise world, the cost of an AI mistake is far higher than the cost of human oversight. Sure, the thousands of engineers at the base of the Indian pyramid face the threat of being replaced. But not if they evolve from builders to navigators and auditors.

This human element is equally critical when it comes to the interface. Software is never just cold logic; it is a bridge to messy human behaviour. You cannot pre-design a perfect interface because business requirements are iterative and change the moment a user touches a screen. While AI can generate a functional UI (user interface), it cannot navigate the internal politics or the cultural nuances of how a procurement officer in London or a banker in New York actually uses a tool. That bridge-building is a service, not a product, and it is exactly what Indian firms have spent thirty years perfecting.

Critics often ignore that the true value of these firms has never been just the syntax of code, but deep domain expertise. AI acts as an accelerator for this knowledge, not a replacement. Indian IT companies, with their deep institutional memory of Fortune 500 workflows, are better positioned to wield these AI accelerators than any Silicon Valley startup starting from scratch.

Admittedly, the industry faces a cannibalisation trap where companies must deploy AI that effectively reduces the billable hours they can charge. However, Indian IT survived the cloud revolution by pivoting. Already the best companies are moving from selling man-hours to selling outcomes—a transition that rewards efficiency rather than penalising it.

Strangely, the real hurdle may be more cultural than technical. In this AI-driven world, English is becoming the ultimate programming language. For decades, Indian techies excelled in the rigid, logical syntax of Java or C++. Now, the code is natural language, nuanced, contextual, and often culturally specific. Mastering the precision of the spoken word to command machines is the true challenge for the Indian IT workforce.

The market is currently repricing the industry based on fear, but it is likely underestimating the sheer volume of work a cheaper tech world will create. Instead of confronting termination, Indian IT might be facing its most significant upgrade yet.

In Company Outsider, Sundeep Khanna distills more than three decades of his experience writing on India Inc. into a thousand words of context and insights that few can bring to the table. Want this newsletter delivered in your inbox? Subscribe here.