Indian IT Can Finally Measure AI Revenue, But the Numbers Don't Stick

India's largest IT services companies have begun putting hard numbers behind their artificial intelligence businesses.

TCS says AI has notched a $2.6 billion annualised revenue run rate for Q1 FY27. Infosys says AI now contributes 8.2% of revenue. HCLTech's Advanced AI business grew 62% year-on-year in the June quarter, reaching $171 million.

Yet the same earnings season largely brought weaker guidance, sluggish revenue growth, and cautious client spending.

That widening gap between expanding AI businesses and slowing overall growth is raising doubts over whether AI is creating enough new demand to compensate for the disruption it is causing across traditional IT services.

While TCS, Infosys, and HCLTech chose to quantify AI revenue in Q1, Wipro and Tech Mahindra—though they spoke extensively about AI-native platforms and AI-enabled operating models—did not attach a number, a divergence that itself suggests the industry has not agreed on what counts as "AI revenue" even as every CEO is now under pressure to show it in black and white.

Numbers Behind the Narrative

At TCS, the AI story is as much about changing business economics as it is about growth. During the earnings call, CEO K Krithivasan said AI engagements differ fundamentally from the company's traditional outsourcing contracts, as they are typically short-duration projects lasting one or two quarters rather than multi-year managed services deals. That means the company has to continually compete for follow-on work instead of relying on long-term recurring revenue.

Even as TCS said its AI portfolio was scaling in double digits, its operating margin slipped 130 basis points following annual wage hikes, underscoring that a growing AI business has yet to translate into stronger financial performance.

Infosys offered the most aggressive framing of AI as a demand driver.

“We saw a strong acceleration in our AI business with AI revenues for the quarter at 8.2%. This is growing at double-digit QoQ over the last several quarters. With this momentum, we see long-term relevance of our services for our clients," Chief Executive Officer Salil Parekh told analysts, pointing to $3.6 billion of large deal wins in the quarter, with 61% of those wins representing net new business. He also credited the company's Topaz platform and partnerships with leading AI model-makers. 

Parekh also disclosed that more than 80,000 Infosys employees are now using AI coding tools internally. 

However, at the same lines, Infosys narrowed its FY27 revenue growth guidance to 1.5–3%, from 1.5–3.5% earlier, affected by a client programme termination, softer volumes through the quarter, and a bigger-than-expected hit from macroeconomic conditions.

Meanwhile, HCLTech's C Vijayakumar split the business into what he called "AI-native and AI-amplified" opportunities, which HCLTech wants to capture disproportionately, and "AI-disrupted services," where the company has openly conceded ground. 

The management told analysts they expect AI to shave 2–3 percentage points off traditional revenue even as the company tries to grow faster than that decline. 

HCLTech backed the rhetoric with capital, approving a ₹3,500-crore AI data centre investment and describing its ambitions, in Vijayakumar's words, as "not a colo (colocation) business" but a full-stack play spanning GPUs, models, and managed services.

On the other hand, while Wipro's Srini Pallia said, "AI disruption is expanding the market, not shrinking it", it delivered the weakest numbers of the big four. 

Its IT services margin fell to 16%, profit was flat, and the company's own guidance implies revenue could shrink again next quarter. Wipro is betting on a newly formed AI-Native Business & Platforms unit and an Applied AI Centre of Excellence built with Anthropic, but has so far declined to disclose what share of revenue that bet actually represents.

Sceptical Stance

Sell-side reaction to the quarter has been considerably more sceptical than the language in the earnings calls. 

According to a recent report by brokerage firm ICICI Securities, while Infosys’s AI revenue (8.2% of overall revenue, up double-digit QoQ) and net new deal TCV component (up 6.4% YoY) are growing at a healthy pace, the moderation in large deal TCV (down 4.3% YoY) and growing disconnect between TCV to revenue conversion suggest continued pressure from AI-led deflation.

Other industry voices frame the disconnect more in terms on the timing of disruption than the disruption itself.

Gaurav Vasu, Chief Executive Officer of research firm UnearthInsight, tells AIM that though AI is "helping sustain growth, improve productivity, and protect margins," it is "not yet driving the kind of high single-digit or double-digit growth that can offset weakness in the traditional IT services business." 

AI can become a genuine growth driver, in his view, only once enterprises begin "allocating incremental technology budgets specifically for AI-led transformation rather than funding AI from existing IT spending", and he believes the industry is not there yet. 

Most AI work today, according to him, is "still tied to improving productivity, modernising applications, or enhancing existing workflows," which reinforces the existing services model rather than opening new revenue streams. 

Vasu flags a structural limit on the opportunity as it stands: current AI spend for services companies is concentrated in migration, implementation, integration. and consulting—"an important opportunity" but one that "captures only a small part of the broader AI value chain," with much of the value still accruing to AI-native platforms and software companies embedding AI directly into their own products.

Praveen Bhadada, MD and CEO of consulting firm Neovay Global, describes the present AI deals as "lumpy because clients are still consolidating vendors, running proofs of value, and demanding productivity pass-throughs on legacy contracts." 

AI, in his reading, is "simultaneously cannibalising traditional T&M revenue through pricing pressure and seeding the next growth cycle"—inflicting near-term pain but also giving a long-term opportunity. 

Bhadada puts the inflection point "at least 12–18 months out," once AI moves into full-stack transformation programmes backed by multi-year commitments.

Cautious Guidance

Away from the AI-hype cycles of the last two years, IT firms are stepping carefully into the next quarter, bracing for AI headwinds and possible contraction. 

This doesn’t read like an industry confident that AI has already become a growth engine. Rather, it is buckling under investor pressure while quietly hedging the growth story with guidance that assumes the pressure continues.

“I think it's really important that IT services firms understand that the economics of how you deliver software is going to be significantly different,” remarks Navin Budhiraja, Chief Technology Officer at enterprise AI services firm Hang Ten Systems. “Your sales teams have to evolve, where they are not selling multi-year deals, but selling these shorter deals, so they have to stay, in some sense, constantly aware of what is the next thing that can be done”.

According to industry executives, for now, AI appears to be playing two roles simultaneously—creating new implementation and consulting opportunities while compressing the economics of traditional IT services through automation and productivity gains. The story can only change if and when enterprises begin allocating fresh budgets for AI-led transformation rather than funding it from existing technology spending.