AI Disruption & India’s IT Services Reset
India’s $300-billion IT services industry, built over three decades on the bedrock of labour cost arbitrage, is facing a defining test. As AI rapidly automates routine coding, testing, and maintenance, the traditional model is splintering. For thousands of freshers and junior engineers, the entry-level jobs that were once the bedrock of the sector are among the most exposed.
The Magnitude of the Shift
The disruption is structural, not cyclical. TeamLease, a leading workforce solutions provider, has reported a 60-70% decline in headcount demand for traditional coding, development, and testing roles over the past two years. This shift has sent shockwaves through the stock market, with the Nifty IT index declining approximately 20% in 2026, wiping out tens of billions of dollars in investor value.
Veteran venture capitalist Vinod Khosla has issued a stark warning: “India’s IT services industry will be gone,” arguing that AI agents will soon take over most of the cognitive and repetitive work that Indian IT companies currently outsource to human workers. The displacement, in his view, is not a question of if but when. Even as headline figures grow—with the industry on track to cross $300 billion in revenue in FY26—growth has slowed to around 5-6%, and hiring has cooled across major companies.
The New Value Proposition: From Code Writers to System Orchestrators
The most immediate shift is in how engineers spend their time. Agentic AI systems—which do not merely respond to prompts but plan, reason, and execute multi-step tasks with minimal human intervention—can now handle a growing portion of software development work: scaffolding repositories, generating boilerplate, writing unit tests, and reviewing pull requests.
In practice, Indian engineers are shifting from being primary producers of code to orchestrators of systems that produce code. This demands a fundamentally different skill set:
| Traditional Role | Emerging Role |
|---|---|
| Writes code from scratch | Prompts, configures, and evaluates AI agents |
| Follows specifications | Frames problems and translates business needs |
| Specialises in syntax and algorithms | Demonstrates systems thinking and architectural judgment |
| Works in isolation | Orchestrates human-AI collaboration |
Ganesh Gopalan, Co-Founder & CEO of Gnani.ai, captures this shift: “We’re no longer hiring for task execution; we’re hiring for problem framing, adaptability, and the ability to learn new systems at velocity”.
The Skills Gap: A 43% Shortage in Critical AI Roles
Despite the disruption—or perhaps because of it—the demand for specialized AI talent is outpacing supply. A report by Quess Corp reveals a 43% skills shortage in AI, data, and analytics roles across India’s Global Capability Centres (GCCs). Other critical areas facing significant talent gaps include:
- Platform engineering: 38% skill gap
- Cybersecurity (particularly zero-trust architecture specialists): Elevated demand
- Cloud infrastructure (FinOps automation and cloud-native governance): Growing shortages
Kapil Joshi, CEO of Quess IT Staffing, notes that supply shortages ranging between 18% and 43% across these domains underscore “the pressing need for accelerated upskilling, stronger functional mobility, and deeper Tier II ecosystem maturity”.
The Two-Track Response: IT Services vs. GCCs
India’s IT landscape is responding to agentic AI in two distinct ways:
Global Capability Centres (GCCs) are accelerating adoption with relative speed. As engineering arms of parent companies rather than billing by the hour, GCCs have a direct incentive to maximise engineer output. An engineer who can ship three times the features with AI assistance is three times more valuable. GCCs in BFSI, healthcare, and e-commerce are already investing in internal platforms that bring agentic tooling into standard engineering workflows.
Traditional IT services firms face a more structurally complex transition. When revenue is tied to headcount—to the number of engineers billed to a client—productivity improvements from AI create a paradox: an agent that halves task completion time can simply halve billable hours. Services firms are navigating this by shifting pricing models toward outcomes and expanding into higher-value consulting.
The Reskilling Revolution: One of the World’s Largest Upskilling Exercises
To bridge the skills gap, IT firms are undertaking one of the largest workforce reskilling exercises in corporate history. Over 2 million professionals have been upskilled in AI, including 200,000-300,000 in advanced AI.
TCS has articulated a clear ambition to become the world’s largest AI-led technology services firm. Today, more than 217,000 associates are deeply skilled in AI, reflecting a deliberate, enterprise-wide approach. For freshers, TCS has shifted from a technology-led training model to a business-aligned, AI-first framework that blends AI immersion with hands-on project experience.
Infosys has reimagined fresher training to be AI-first, embedding foundational AI fluency through structured pathways such as AI-Aware, AI Builders, and AI Masters. More than 90% of its employees are now AI-Aware, enabling them to work with AI as a co-pilot across coding, testing, and problem-solving.
Wipro is training 10,000 employees on Anthropic’s Claude models over 18 months and has established a new Centre of Excellence for applied AI at its Bengaluru hub. This follows a similar announcement by TCS, which will equip 50,000 associates with Anthropic’s Claude and jointly develop AI solutions for regulated industries.
Cognizant has already upskilled more than 330,000 associates through over 1,000 learning programmes. Its GenAI and agentic AI programmes aim to help associates enhance productivity, decision-making, and business outcomes.
The Mojo Has Moved: Cognitive Arbitrage
The industry’s competitive advantage has been permanently reset. The consensus among industry leaders, captured in MIT Sloan Management Review, is that “cost arbitrage is what got us here; cognitive arbitrage is what gets us to the next decade”. Firms that win will stop measuring themselves in headcount and start measuring themselves in value delivered.
India’s engineering talent remains its greatest asset, but the demands on that talent have transformed. The entry-level engineer is no longer expected to be just a developer—they are expected to be “an architect in training,” combining domain knowledge, critical judgment, and the ability to work alongside intelligent systems.
The question is not whether India’s IT workforce can adapt—it is how quickly. As one industry leader put it: “India’s moat in services was never labor costs, but the ability to deliver complex, accountable work at a scale and consistency no one else could match. AI doesn’t erode that moat; it relocates it”. The relocation is from process execution to cognition arbitrage, and the race is on to build the skills that will define India’s next chapter in global technology leadership.
