AI's S-Curve in Indian IT: Assessing the Infrastructure Bet

The technological inflection is here. For Indian IT, artificial intelligence is no longer a side project or a cost-saving tool. It is rapidly becoming the core infrastructure layer, rewriting the industry's fundamental playbook. This is the start of a new S-curve, where the scale of embedding, the speed of adoption, and the depth of disruption signal a once-in-a-generation shift.

The scale of embedding is already profound. Across Indian tech companies, over 40% of software development work is now handled by artificial intelligence. This isn't a niche experiment; it's a mainstream operational reality. The report from NASSCOM and Indeed shows AI's reach extends beyond code, with 20–40% of work across HR, IT, and administration already automated. For the industry's largest player, the commitment is staggering. Tata Consultancy Services now has 217,000 employees with "higher order AI skills," up from 180,000 barely a month ago. This isn't just training; it's a forced, large-scale re-skilling of a workforce of over 600,000, signaling that AI fluency is now a baseline requirement for survival.

This speed of adoption is the hallmark of a paradigm shift. The industry is moving from a model built on people and processes to one driven by intelligence and data. As one analysis notes, the people- and process-driven model that defined traditional outsourcing is being rewritten by automation, intelligence, and data-native workflows. The disruption is generational because it attacks the core value proposition: the billable hour. AI automates the very tasks humans once performed, compressing project timelines and margins. The result is a new competitive landscape where AI upstarts are emerging with smarter, faster solutions, challenging the incumbents' decades-long dominance.

The bottom line is that AI is the new infrastructure. It's the foundational layer upon which all future IT services will be built. The industry's trajectory is clear: Almost all (97%) expect work to be done by teams made up of humans and AI by 2027. This isn't a future scenario; it's the immediate operational mandate. For investors, this means looking past the current service margins and focusing on which companies are building the rails for this new paradigm. The S-curve has begun.

The Monetization Gap: From Pilots to Production

The industry has built the rails. Now it must learn to run the train. The central uncertainty for Indian IT is no longer about capability, but about monetization. Despite the deep embedding of AI into operations, the translation into scalable, measurable revenue remains the critical disconnect. This is the gap between the S-curve of adoption and the S-curve of profit.

The budget commitment tells the story. There is a striking paradox: 95% of Indian companies spend less than 20% of their IT budget on AI, even as 47% use multiple GenAI applications daily. This isn't a lack of conviction; it's a pragmatic, cautious approach. Executives are prioritizing speed and tangible ROI over large-scale bets, opting for quick wins with APIs and incremental tools rather than massive, long-term investments. It's a rational response to the need for immediate impact, but it also caps the potential for exponential growth.

This caution has defined the recent past. 2025 was another year of pilots and incremental gains – not enterprise-scale monetization – for most of the industry. The focus has been on proving concepts and demonstrating value on a small scale, not on building the new, outcome-based business models that will drive future revenue. The industry is stuck in the early, slow phase of the adoption curve, where the promise is clear but the payoff is not yet systemic.

The scale of current monetization underscores the challenge. Even for the leader, the numbers are still modest. Tata Consultancy Services' annualized AI services revenue is $1.8 billion. While that represents a 17.3% quarterly growth and a clear strategic priority, it is a tiny fraction of its total $264 billion business. For the industry as a whole, the path from this promising start to a dominant revenue stream is the unproven variable.

The bottom line is that the infrastructure layer is being built, but the monetization engine is not yet running at full throttle. The industry's central uncertainty is now operational: it has the talent, the tools, and the vision. The next phase is about execution at scale, moving from isolated pilots to integrated, platform-driven services that capture the full value of the AI paradigm. Until that happens, the exponential growth story remains on hold.

Building the New Rails: Strategic Partnerships and Talent

The industry's bet on AI infrastructure is now being built with concrete partnerships and a fundamental rethink of its workforce. This is the strategic layering required to move from pilots to production at scale. The moves by the largest players are about securing the compute foundation and the human capital that will drive the next paradigm.

A prime example is Tata Consultancy Services' new alliance with AMD. This isn't a simple vendor deal; it's a co-development pact aimed at building industry-specific AI solutions from the ground up. TCS and AMD plan to engage on the co-development of industry specific AI and GenAI solutions, targeting sectors like life sciences and manufacturing. The goal is to create tailored accelerators and frameworks that help customers move beyond experimentation. More importantly, the partnership includes a joint investment in talent, aiming to build a deep pool of experts who can co-innovate and deliver next-generation AI solutions. This is infrastructure building: combining TCS's global integration reach with AMD's high-performance computing hardware to create a new, integrated stack for enterprise AI.

This partnership reflects a broader shift in the talent model. Hiring is moving from a focus on today's specific skills to identifying long-term potential and learning agility. The biggest shift we will see in talent acquisition is moving from hiring for today's skills into hiring for long-term potential. Soft skills like problem-solving and adaptability are becoming the new baseline, as AI-driven platforms make reskilling more accessible. This is a critical adaptation for a field where technical specifics will evolve rapidly. The strategy is to build a workforce capable of continuous learning, ensuring the company's human capital keeps pace with the exponential curve of AI advancement.

Yet, this strategic build-out faces a looming constraint: a severe talent shortage. The demand for AI skills is projected to more than double by 2027, from 600,000–650,000 to more than 1,250,000. This creates a clear race between the industry's ability to build its rails and the supply of skilled workers to operate them. The partnership with AMD and the focus on potential are direct responses to this gap, aiming to create a pipeline of experts faster than the market can supply.

The bottom line is that Indian IT is laying the fundamental rails for the AI paradigm. Strategic alliances like TCS-AMD are building the hardware and software stack. The talent model is being retooled to prioritize adaptability over static expertise. But the exponential growth story hinges on closing the widening demand-supply gap. For now, the industry is investing heavily in its infrastructure, but the final, most critical component-the human capital to fully harness it-remains the most uncertain variable.

Catalysts and Risks: The Path to Exponential Growth

The industry stands at a fork. The path to exponential growth is not guaranteed; it will be determined by a few critical inflection points in the coming year. The catalyst is clear: a fundamental shift in how work is delivered and priced. The industry must move beyond embedding AI into existing processes and instead rearchitect its entire delivery model around AI-human 'pods' that are explicitly designed for measurable business impact. This is the operational engine for the new S-curve. It requires dismantling work at the archetype level and rebuilding it with these hybrid teams, a move that can unlock productivity gains of up to 40% in IT and 80% in BPM roles. The parallel shift to outcome-based models and agentic pricing is equally vital. Pricing must evolve from time-and-materials to commitments linked to business results, often reducing costs by 30-50%. This is the monetization leap from pilots to production.

The key risk to this trajectory is the rise of AI-first challengers. These nimble startups are not constrained by legacy structures and can reimagine legacy services with smarter, faster, and more adaptive solutions. They are building new business models from the ground up-AI-enabled services, services built for AI, and pure software-led platforms-that can outperform traditional models on speed, quality, and cost. The incumbents' vast talent pool and client trust are enduring advantages, but they are not moats if execution is slow. The risk is that AI upstarts capture market share by delivering the very outcome-based, platform-driven value that the giants are still piloting.

The critical timeline is now. 2026 should be the year the industry changes its trajectory from pilots to enterprise-scale monetization. Analysts warn that nearly 25% of planned global AI spend could fall short unless vendors demonstrate tangible ROI. The window is closing. The industry has spent the last year building the rails. In 2026, it must prove it can run the train. The success of the TCS-AMD partnership, the scaling of AI-human pods, and the adoption of outcome-based pricing will be the metrics that decide whether Indian IT can achieve exponential growth or be left behind on the old S-curve. The inflection is imminent.