Indian IT services firms are laying the groundwork through investment in AI centers of excellence (CoEs), dedicated budgets, and enhanced data standardisation efforts
| Photo Credit:
As Indian IT firms embed AI across service lines and delivery models, most client deployments remain in pilots or limited production. Analysts warn that AI will influence revenues deal by deal — initially shoring up margins and renewals — before translating into visible topline growth closer to 2027.
A Nomura report highlights that almost every India IT services company is gearing up for investments in AI. The key strategy revolves around using AI internally, infusing AI in service lines, helping clients in their AI journey, and investing in ecosystem partnerships. Clients are also gradually moving from proof-of-concept (PoC) projects to standalone implementations of AI. Due to this momentum, bigger revenue pools for India IT service providers should emerge when enterprise adoption of AI happens, likely to gather pace in the next 12-18 months.
Investments
Biswajeet Mahapatra, Principal Analyst, Forrester, highlighted that Indian IT services firms are laying the groundwork through investment in AI centers of excellence (CoEs), dedicated budgets, and enhanced data standardisation efforts. However, the majority — 75 per cent to 92 per cent — remain in pilot or early deployment stages, with only 8 per cent to 25 per cent having fully operational production systems.
Key barriers include fragmented data, legacy system integration, domain-specific talent shortages and weak governance frameworks. Scaling requires a shift to AI-first operating models, combining internal and external teams under leadership oversight and stronger backend data infrastructure.
Globally, around 40 per cent of GenAI proofs of concept are advancing into production, and in India, this trend is accelerating: 40 per cent of enterprises now run multiple GenAI use cases in production, with 30 per cent still in pilot phases.
“2027 would be the tipping point when AI would start showing measurable results as organisations move from pilots to scaled implementations. The timing will depend on factors like enterprise readiness, integration with existing systems, and service providers delivering clear business outcomes,” he added.
According to Greyhound Research, enterprises are moving into production cautiously and in defined pockets, with production deployments that are assistive rather than autonomous. Clients are no longer funding AI as experimentation, but production scale only follows once data ownership, risk posture and audit requirements are addressed. Where those conditions are missing, PoCs continue to linger. The shift is deliberately conservative rather than aggressive.
“AI is already influencing revenue, but its earliest impact is defensive rather than expansive. It is helping Indian IT firms protect renewals, expand scope within existing accounts, and remain competitive in large deal evaluations. Net new logo-driven AI revenue will take longer. Over the next 12 to 24 months, as clients move from enablement to sustained production and managed AI services, revenue attribution will become clearer. That is when AI begins to show up more visibly in topline numbers,” Sanchit Vir Gogia, Chief Analyst at Greyhound Research, explained.
The inflection will unfold deal by deal as AI becomes embedded into long-running programmes, rather than sold as a standalone capability. Service lines with repeatable, well-governed workflows see the earliest AI gains. Application development benefits from faster coding and testing, though early progress can slow as quality controls reset. IT operations and service desks advance more steadily due to clearer metrics and lower risk. Business process services gain the most when AI is paired with process redesign, while data and analytics accelerate as AI forces data clean-up and platform consolidation. Consulting benefits when tied to execution. In short, disciplined workflows compound AI’s value; unstable ones amplify confusion.
AI-related projects
The analysts noted that AI-related projects are already reshaping pricing models, increasingly geared toward outcome-based and consumption-based contracts, enabling providers to charge premiums for proven business value.
In the near term, AI is helping providers deliver the same work with fewer errors, less rework, and shorter cycles, improving the margin quietly. Clients generally expect those gains to show up as faster delivery and better reliability rather than higher rates. Pricing power emerges only when AI is tied to outcomes and accountability, especially in regulated or complex environments where execution capability is scarce. Over the next 12 to 18 months, the most successful firms will not advertise AI premiums, but redesign delivery, stabilise quality after early friction, and allow margins to expand naturally.
Early data indicates a 200 to 400 basis points margin uplift in some AI-enabled engagements. While traditional fixed-price and FTE billing remains under pressure, with some rates falling 1 per cent to 20 per cent, AI services tied to specialized IP, platforms, and automation frameworks are seeing better margin resilience.
Published on December 14, 2025
