Short answer: Tata Consultancy Services has publicly argued that AI is adding to its business rather than eating into existing revenues. The surprising part is not the claim itself. It is the logic behind it, and that logic applies to any company selling or buying software and services.
What is the AI cannibalisation fear?
Cannibalisation means a new product shrinks the sales of an older one. In IT services, the worry is simple: if AI coding assistants and agents let one engineer do the work of several, clients may pay for fewer people-hours, and billing models built on headcount could erode.
TCS management has pushed back on this view, saying AI is a growth driver and not a threat to existing revenues. Treat that as a company position, not a proven law. It is a useful hypothesis to test in your own business.
Why It Matters Now (2025–2026 Context)
Large enterprises moved from AI pilots to production planning during 2025. Boards now ask two questions at once: how much can we save, and what new revenue can we open? Service providers are being asked the same questions by their own clients.
This matters for founders and CTOs because vendor pricing is changing. Fixed-fee, outcome-based and platform-led deals are replacing pure time-and-materials contracts, and buyers who understand the shift negotiate better.
How AI Is Changing This
The growth argument rests on three mechanisms. First, AI creates new work: data foundations, model integration, security reviews and change management all need to be built before any agent runs in production. Second, cheaper delivery makes previously unaffordable projects viable. Third, clients who see results buy more, not less.
The contrarian insight: productivity gains rarely shrink total spend when the cost of a capability falls and demand for it is elastic. Cheaper software projects usually mean more software projects. We call this the Elastic Demand Test: if lowering your price by 30% would double requests, AI efficiency is a growth lever, not a threat.
Real-World Examples
TCS is one of the largest IT services firms in the world, and its leadership has framed AI as an expansion of its addressable work, covering areas such as AI-ready data platforms, legacy modernisation and managed AI operations. Other large providers have made similar arguments, though results differ by company and quarter.
A realistic scenario for an SME: a 40-person software agency adopts AI coding assistants and cuts delivery time on routine modules. It could pocket the savings, or reprice and take on twice as many maintenance and integration contracts. The agencies that choose the second path are the ones that treat AI as growth.
Practical Insights / Actions
Founder mistake: measuring AI only by hours saved. Hours saved is a cost metric. Growth shows up in new offers, faster launches and higher win rates, so track those too.
Hidden opportunity: the work that AI makes cheap is often the work you declined before, such as small custom integrations, internal tools and data cleanups. Re-quote those and see which become profitable.
A simple action list for decision-makers:
Strong opinion: if your AI plan has a headcount-reduction target and no revenue target, it is a cost project dressed up as strategy.
Future Outlook
Expect pricing to keep moving toward outcomes and platforms. Providers that bundle AI operations, governance and data work will look different from those selling pure labour. Some revenue lines will shrink, so the honest framing is net effect, not zero effect.
Buyers should watch for vendors who cannot explain how AI savings are shared with the client. Transparency on that point is a good signal of a long-term partner.
Conclusion
TCS's position is a reminder that AI economics depend on demand, not just efficiency. Whether AI grows or cannibalises your revenue depends on what you do with the savings. If you want help mapping AI to revenue and not just cost, RP SoftTech offers an AI strategy audit for SMEs and growing teams. Start with the guide above, then book a conversation when you are ready to scope a pilot.