Indian IT firms need to step up, take clients forward in AI world: Experts

Industry leaders say Indian IT firms must evolve from implementation partners to orchestration partners in the AI era, even as AI-driven revenue is projected at $10-12 billion in FY26

The Indian IT services sector, expected to touch $315 billion in revenue in FY26, has come a long way from being dismissed as the ‘back office’ of the world. Yet today, a new label is emerging — that of a ‘plumber’, plugging in tools being released by global artificial intelligence (AI) firms.

 

For a sector that employs close to six million people and remains a major contributor to the country’s gross domestic product (GDP), it may be time to reset the narrative — and move beyond being seen merely as back-end support for global multinationals grappling with the AI juggernaut.

 

The industry, however, is approaching the shift with caution rather than chest-thumping. At the Nasscom Technology Leadership Forum (NTLF), senior leaders and chief executive officers (CEOs) reiterated that new-age AI tools will not upend decades of accumulated expertise overnight.

How can Indian IT reset its role in the AI era?

Asked how Indian IT services can reset the narrative at a time when it risks being stereotyped again, Rajesh Nambiar, president of Nasscom, said the biggest shift underway is from being an “implementation” partner to becoming an “orchestration” partner.

 

Nambiar also stressed the fundamental difference between enterprise IT deployments and the capabilities demonstrated by new AI model releases. “I do not underestimate the power of these tools. With every new release, we see new things being built. But remember, most of this innovation is still concentrated in the software development layer,” he said during a panel discussion at the NTLF.

 

Krishnan Iyer, chief growth officer at LTM, agreed that the industry needs to articulate a stronger point of view.

 

“Nobody is denying that AI is disruptive. Both our customers and we are bullish on the technology. But we are also cautious while deploying it in enterprise environments to ensure there are no pitfalls. Even today, most AI systems are probabilistic in their outcomes, not deterministic. And the gap between probabilistic and deterministic systems remains wide. I believe that gap will narrow over the next three to four years,” he said.

What is the scale of AI-driven revenue in FY26?

To the industry’s credit, even as AI adoption has accelerated at what executives describe as “Formula 1 speed”, AI-driven revenue is expected to reach $10 billion–$12 billion in FY26. For a technology that emerged barely three years ago, that is a significant milestone — especially in an enterprise ecosystem where adoption cycles typically span five to six years.

 

On Tuesday, when Anthropic announced that its Claude Code tool could automate the modernisation of COBOL systems, global technology stocks reacted sharply. IBM alone reportedly lost nearly $30 billion in market capitalisation. In India, IT firms — many of which have built billion-dollar businesses maintaining legacy mainframe applications — also saw selling pressure.

Will AI-led modernisation shrink or expand opportunities?

Yet industry leaders dismissed the selloff as overhyped.

 

Srikanth Velamakanni, co-founder and CEO of AI-native firm Fractal Analytics, said it is true that AI will significantly assist in modernising mainframe systems and tackling the massive technical debt that exists across enterprises.

 

“The first-order reaction is that if project sizes shrink, the opportunity shrinks. But the flip side is that it exposes the entire global base of COBOL implementations to modernisation. Earlier, a chief technology officer (CTO) or chief information officer (CIO) would hesitate to undertake a billion-dollar modernisation project — it was too risky and too core to the business. If it failed, careers were at stake,” he said.

 

He added: “The total addressable market of technology expands massively in such a scenario. It presents a once-in-a-lifetime opportunity to retire technical debt. And the beneficiary is the broader tech ecosystem.”

 

For those prematurely writing the industry’s obituary, there is another reality to consider: leading AI model companies are actively seeking partnerships with IT services firms. The reason is straightforward — only IT services players possess the deep enterprise relationships and execution muscle needed to embed these tools meaningfully within complex business environments.

 

For instance, Anthropic is collaborating with Infosys to develop and deliver advanced enterprise AI solutions to companies across telecommunications, financial services, manufacturing, and software development. The collaboration will begin in telecommunications with a dedicated Anthropic Center of Excellence to build and deploy AI agents tailored to industry-specific operations.