India's IT sector is not collapsing, but AI is forcing investors and founders to ask a harder question: how much of the old outsourcing model still survives when software can do more of the work itself?
India has spent three decades turning skilled manpower into export wealth. Now the same model is being tested by AI agents that can write code, review documents, answer customer queries and sit inside enterprise workflows without needing a large delivery team behind them.
That is why the latest market wobble matters. According to Reuters, India's IT shares fell to a three-year low on May 12 after OpenAI launched a new enterprise deployment company and agreed to acquire Tomoro, an applied AI consulting and engineering firm. The Nifty IT index dropped 3.6 percent to its lowest level since May 2023, with Tata Consultancy Services, Infosys, HCL Technologies and Wipro all falling between 2.5 percent and 4 percent.
Reuters Breakingviews pushed the argument further on May 14, with Aimee Donnellan and Una Galani framing India as an early stress test for AI-driven job disruption. That is a fair place to look. If AI is going to pressure any large labor market first, it will probably show up in a country where millions of people are tied to coding, maintenance, support and outsourced business processes for global clients.
The basic fear is simple. For years, global companies saved money by sending work to large Indian IT vendors that could hire, train and manage vast teams at scale. The model worked because software projects were complex, Western talent was expensive, and clients needed armies of people to keep systems running.
AI changes the calculation. OpenAI's new Deployment Company is not just selling access to a model. It is building a services layer around AI, backed by more than $4 billion in initial investment and helped by Tomoro's roughly 150 forward deployed engineers and deployment specialists. That puts frontier AI companies closer to the kind of enterprise work that Indian outsourcing firms have historically owned.
Anthropic has been pushing in a similar direction with Claude Code, which has already sharpened investor worries about software development work becoming less people-heavy. The point is not that every programmer disappears. The point is that a client may need fewer junior developers, fewer testing hours and fewer support seats when agents can handle first drafts, routine fixes and operational tasks.
For a founder, this is not an abstract market story. It affects hiring plans, service margins and the price of building a company. A startup that once needed a ten-person offshore team for product maintenance may soon ask whether three strong engineers with agentic tools can do the same job faster. That creates opportunity for some companies and pressure for others.
The numbers are not all bleak
The alarm needs balance. Nasscom's latest sector projections show India's technology industry reaching about $315 billion in FY26 revenue, growing 6.1 percent. The workforce is expected to rise to 5.95 million employees, with roughly 135,000 net jobs added. That is not the picture of an industry being wiped out.
But those figures also show the next phase will not look like the last one. Revenue can keep growing while hiring slows. Companies can sell higher-value AI services while reducing the need for large entry-level benches. The stress will fall unevenly, with routine delivery roles exposed first and workers who can combine domain knowledge, engineering judgment and AI fluency becoming more valuable.
This is where India has a real opening. Its biggest IT firms have deep enterprise relationships, process knowledge and global delivery discipline. Those things still matter. If TCS, Infosys, Wipro and HCL can turn AI into managed transformation work, they can defend part of their position rather than simply watch frontier labs move downstream.
The harder task is cultural. A manpower-heavy services company is built to bill time, manage utilization and keep large teams productive. AI services reward reusable tools, faster delivery and smaller groups that can solve bigger problems. That shift can be painful because it changes incentives inside the business, not just the technology stack.
Founders should read the signal carefully. The risk is that parts of the outsourcing market become less reliable as a source of cheap, abundant labor. The opportunity is that AI lets small teams compete with companies that once needed scale to win. In practical terms, the best startups will hire people who know how to direct AI systems, audit their output and connect automation to real customer workflows.
India will not stop being a technology labor force because OpenAI or Anthropic builds better tools. But the market is already asking whether the next dollar of IT revenue needs as many people behind it as the last one did. That is the question to watch, because it will shape wages, startup costs and emerging-market tech wealth long before the debate about AI jobs is settled.
Also read: Bambu Lab's open-source fight is now a startup trust problem • Claude just made lost Bitcoin recovery look like a real market • Coinbase backs the CLARITY Act as crypto rules reach markup.
