India’s IT outsourcing industry is going through a sharp and increasingly visible reset, with AI now emerging as the central force behind both market repricing and job anxiety. The human impact is already clear in accounts cited by the report: Shiv, a 52-year-old engineer who spent 14 years at Oracle in India before being laid off, has made it a habit to send out at least five resumes every day since April. Another worker, 25-year-old Priyanka, learned of her dismissal through an email while preparing for the gym, and is now relying on savings to keep up with installment payments and continue living in Bengaluru.
In financial markets, that stress has translated into one of the clearest sector-level trades tied to the “AI replaces white-collar labor” theme. The Nifty IT index, India’s main listed benchmark for technology services, peaked at 46,089 on December 13, 2024. By the end of June 2026, it had fallen 43% from that high. In the first half of 2026 alone, the index dropped about 30%, making it the worst-performing major sector in India, compared with a roughly 9% decline in the broader Nifty 50.
Markets are repricing India’s IT sector around AI disruption
India’s top IT names have suffered deep drawdowns. TCS, Infosys, Wipro and LTIMindtree have each fallen roughly 50% from their highs, while the country’s ten largest IT companies have collectively lost about INR 19.28 trillion in market value, equivalent to more than $200 billion. TCS alone has slipped below the INR 10 trillion market capitalization threshold. Bloomberg data cited in the report shows that the combined weight of the five biggest IT companies in the Nifty 50 has fallen below 7.6%, the lowest since 2002.
The pace of selling has also been notable. Several sharp declines in Mumbai appeared to coincide with major announcements from U.S. AI firms. On February 4, Anthropic unveiled a new coding tool and said it could automate much of the exploration and analysis involved in legacy system modernization. That matters because COBOL modernization has long been one of the core revenue streams for Indian outsourcers. After the announcement, the Nifty IT segment entered a selloff that eventually erased more than 15%, wiping out about INR 5.08 trillion in value.
In May, OpenAI said it would spend more than $4 billion to build a team of “forward-deployed engineers” working directly with enterprise customers to redesign workflows around AI. Investors quickly read the implication: high-value consulting, deployment and transformation work might increasingly bypass Indian service companies. The Nifty IT index then fell to its lowest level since May 2023. In June, Accenture plunged nearly 18% in a single day, its biggest daily drop since listing. The next trading day in Mumbai, Nifty IT slid another 6%, while Infosys fell 8.19% to a five-year low, erasing INR 1.35 trillion in just one session.
Sell-side analysts have turned more cautious as well. Jefferies warned that, in a worst-case scenario, valuations of Indian IT stocks could still fall another 30% to 65%. Citrini Research projected that contract cancellations at TCS, Infosys and Wipro would continue accelerating through 2027. Domestic brokerage Nirmal Bang downgraded TCS from buy to sell and cut its target price from INR 3,046 to INR 1,693. Taken together, these moves suggest that global investors are no longer treating India’s IT sector as a stable compounder, but as a business model exposed to direct substitution risk.
India’s outsourcing model depends on selling engineering hours
To understand why India appears especially vulnerable in the AI era, it helps to look at what the industry has historically sold. At its core, the model has been simple: billable engineering labor. Since the late 1990s, India built a massive global delivery machine in which clients sat in New York or London while software work was completed in Bengaluru or Hyderabad. The value proposition was labor arbitrage. Similar tasks could be done for a fraction of Western costs, and over time that formula grew into a $283 billion industry.
That model did far more than create export revenue. It also produced one of the largest white-collar middle-class expansions in modern India. TeamLease Digital CEO Neeti Sharma, as cited in the report, summarized the logic this way: borrow money for an engineering degree, join TCS, Infosys or HCLTech, and your life is effectively set. The article points to an engineer named Pooja as a representative example. She grew up in a cramped suburb near Kolkata, started as a programmer in 2005 with a monthly salary of INR 7,056, and now earns INR 3.5 million a year at a leading IT company.
The spillover into the wider economy was substantial. A joint study by Nasscom and Crisil found that by 2007, each IT job could support around four additional jobs elsewhere in the economy, including drivers, guards, cooks and domestic workers. Housing finance as a share of India’s GDP rose from 0.6% in 1995 to about 11% today, with 35% concentrated in the country’s southern technology hubs. In cities such as Bengaluru and Hyderabad, the housing market became deeply tied to the payroll of IT professionals.
The weakness in this system is that much of what it sells has a precise description: repetitive work done by junior and mid-level engineers. Template coding, manual testing, ticket processing and legacy maintenance are all areas where large language models have improved quickly. From the client’s perspective, AI now looks like a lower-marginal-cost, always-on, visa-free junior engineer. India spent three decades becoming the world’s preferred substitute for expensive U.S. programmers. Now it faces a new substitute that may be cheaper than Indian programmers themselves.
Layoffs, weaker hiring and delayed onboarding are hitting the middle class
The labor impact is already broadening beyond a handful of headline layoffs. TCS announced 12,000 job cuts in July of last year, equal to about 2% of its workforce and the biggest layoff in the company’s history. Reuters quoted a 45-year-old employee in Kolkata describing the news as devastating, especially for someone at that age trying to find a new role. At the same time, more than 500 candidates who had received TCS offers with joining dates set for July 2025 were still waiting indefinitely, with many having already left previous jobs.
Hiring momentum has weakened materially. India’s top five IT companies posted a net reduction of around 7,000 employees in the fiscal year ending March 2026, compared with net additions of more than 12,000 in the prior year. Over the previous five years, those firms had hired about 230,000 people a year on average. In FY26, that figure fell to 170,000. TCS reduced its annual fresher hiring plan from an average of 40,000 over the prior three years to 25,000.
According to UnearthInsight founder Gaurav Vasu, roughly 400,000 to 500,000 IT workers could face layoff risk over the next two to three years, with about 70% of them in the four-to-twelve-year experience band that has traditionally formed the backbone of delivery teams. Fund manager Saurabh Mukherjea put the issue in even broader demographic terms. India produces about 3 million engineering graduates a year, of whom around 1.5 million are considered employable. Before 2020, the IT services sector absorbed almost all of them. Over the past three years, he argues, that absorption has moved close to zero. Meanwhile, Azim Premji University’s India Employment Report 2026 shows unemployment among graduates aged 15 to 25 at 40%.
The shock is now feeding back into sectors that once benefited from the IT boom. In the first quarter of 2026, residential home sales in India’s major cities fell 13% year over year, with analysts directly citing IT layoffs as one of the main reasons. Shared apartment occupancy in Bengaluru has weakened, and landlords are increasingly blaming reductions in headcount at tech firms. Mukherjea also flagged a more worrying pattern: workers who fear they may soon be laid off are rushing to apply for personal loans and mortgages before they lose their income, helping drive credit growth through what he calls “doom loans.”
U.S. policy and AI adoption are squeezing India from both sides
External conditions are making the adjustment harder. Roughly 60% of India’s IT industry revenue comes from the United States, totaling close to $135 billion. Historically, the sector had two ways to serve that market: offshore delivery from India and physical mobility through visas that allowed Indian engineers to work in the U.S. That dual channel is now under pressure.
On the demand side, AI gives American companies a more credible path to “re-shore” certain services or at least reduce the volume of work sent to Indian vendors. On the labor mobility side, visa access is getting tighter. The report says that in September 2025, the Trump administration at one point raised H-1B visa fees from $5,000 to $100,000, a twentyfold jump. Two months earlier, Trump had publicly urged Google and Microsoft to stop hiring in India. In 2024, Indians received more than 200,000 U.S. work visas, and Indian companies accounted for 20% of all approved H-1B visas, showing how important that route had been to the business model.
The result is a form of double compression. AI creates a technology path for U.S. corporations to keep more service work at home or automate it outright. At the same time, visa policy makes it harder for Indian engineers to go to the U.S. themselves. In plain terms, the people cannot easily leave, and the work is no longer guaranteed to come in. For a country with a median age of just 28 and millions of young people entering the labor force every year, that raises the stakes far beyond one sector.
What markets are signaling, then, is not just a cyclical slowdown in Indian IT stocks. It is a structural repricing of white-collar labor in one of the world’s largest talent-exporting economies. Shiv is still sending out five resumes a day. Office towers in Bengaluru are still lit at night. But more workers are beginning to ask a question that would have sounded implausible only a few years ago: how long will those lights stay on, and who will they be on for?

