◉Signal & Seam
Analysis

AI deflation is now hitting services pricing

AI productivity gains colliding with legacy services pricing models
2026-04-24source quality: highaiit-servicesinfosyshcltechearningspricingenterprise-aistrategy

Infosys and HCLTech both reported meaningful AI momentum, yet both delivered cautious FY27 growth outlooks and drew sharp market reactions. The signal is that AI is compressing legacy project economics before it cleanly expands top-line services growth.

If you only looked at the AI headlines, you would assume large IT services firms should be in a near-term growth boom.

If you looked at this week’s numbers and guidance, you would assume the opposite.

Both readings are partially true—and that’s the point.

Infosys and HCLTech just gave us a clean view of an uncomfortable transition: AI is arriving as a deflationary force on legacy services economics before it fully arrives as a growth engine in reported revenue.

The contradiction is real, but not confusing

On paper, both companies can show real AI progress.

So AI traction exists. But markets still reacted harshly, and Reuters coverage across both names repeatedly emphasized the same pressure points: delayed discretionary spending, slower deal ramping, and concern that AI can do parts of traditional project work faster and cheaper.

This is not a paradox. It is transition math.

What’s being repriced: labor-heavy work

For years, large IT services growth depended on a familiar loop:

1. Enterprise starts a broad modernization or transformation program. 2. Scope expands across multiple towers. 3. Revenue scales with multi-quarter staffing and execution.

AI changes the slope of that loop.

Clients are increasingly asking harder questions before committing to the old project shape:

From the vendor side, this creates a temporary but painful dynamic: capability excitement rises while monetization visibility falls.

That is exactly the kind of setup equity markets discount aggressively.

AI revenue growth can coexist with weaker headline growth

This is the part many hot takes miss.

It is possible—likely, even—that AI-specific revenue lines grow quickly while total growth remains muted.

Why?

Because the denominator (legacy services model) is huge.

If AI accelerates delivery productivity, some of the old spend pool gets repriced downward before new high-value AI work scales enough to offset it. That creates a “J-curve” in services economics:

We are still mostly in Phase 1.

HCLTech’s and Infosys’s FY27 ranges suggest management teams see the same thing, even while they continue investing and signaling confidence in AI positioning.

Why guidance is the key signal, not only Q4 beats

Quarterly beats matter less in transition periods than forward guidance quality.

Reuters noted that Infosys beat quarterly expectations on revenue/profit, yet attention snapped back to FY27 guidance and sector caution. HCLTech likewise reported substantial AI momentum but still guided below optimistic expectations and cited a fluid demand environment.

That tells you what investors are buying now:

In other words, markets are asking for proof that AI capability translates into commercial architecture, not just technical demos.

The new battleground: contract design, not model branding

If my read is right, the next durable winners in services will be the firms that move fastest on three fronts:

1) Outcome-linked pricing

Vendors need pricing structures that capture value from speed and quality gains instead of being penalized for them. If AI makes delivery 30% faster, but contracts only reward labor volume, economics break.

2) AI-native managed operations

One-off pilot work is easy to announce and hard to scale. Recurring, governed operations around agentic systems are harder to sell initially—but better for durable margins and renewals.

3) Delivery governance as product

As clients shift from experimentation to operational AI, governance (security, auditability, reliability, change control) becomes billable value, not overhead.

The firms that package this layer well can turn AI from a deflation force into a platform force.

My take: this is healthy pain, not terminal weakness

It is tempting to read weak near-term guidance as “AI hype has failed.” I think that read is lazy.

What we are seeing is a normal (and necessary) repricing of a mature services model under a new production technology.

The old model optimized for labor scale and program duration. The new model optimizes for automation, output, and controlled execution.

Those models cannot coexist indefinitely at the same pricing logic.

So yes, this is painful in the middle. But middle phases are where strategy quality shows up.

What I’m watching next

Over the next 2-3 quarters, I’d focus on five indicators:

1. How quickly AI deals move from pilot into run-state contracts (especially in regulated industries). 2. Whether guidance quality improves without margin collapse. 3. How firms describe pricing frameworks (effort-based vs. outcome-based mix). 4. Large-deal quality, not just TCV totals (renewal resilience, scope durability, cross-tower expansion). 5. Evidence that AI productivity gains are retained by the provider rather than entirely passed through as discounts.

If those indicators improve, this year will look like an awkward transition year. If not, AI deflation pressure will keep outrunning AI growth narratives.

Bottom line

The most important signal this week is not that AI demand is missing. It is that AI monetization is being renegotiated in real time.

In IT services, that renegotiation hits pricing first, then reported growth.

So the question for the next year is straightforward: who can redesign the commercial model fast enough that AI stops looking like discount pressure and starts showing up as durable earnings quality?

---

Source trail

Primary - U.S. SEC, Infosys Form 6-K (Apr 23, 2026): https://www.sec.gov/Archives/edgar/data/1067491/000106749126000014/index.htm - U.S. SEC, Infosys Exhibit 99.2 (IFRS USD press release): https://www.sec.gov/Archives/edgar/data/1067491/000106749126000014/exv99w02.htm - HCLTech, *FY26 revenue up 3.9%, led by increasing demand for Advanced AI* (Apr 21, 2026): https://www.hcltech.com/press-releases/hcltech-fy26-revenue-39-led-increasing-demand-advanced-ai

Secondary - Reuters (via WHBL), *India’s HCLTech drags IT index as weak outlook, Q4 miss highlight pressure across sector* (Apr 22, 2026): https://whbl.com/2026/04/21/hcltech-falls-as-weak-outlook-q4-miss-highlight-pressure-on-indias-it-sector/ - Reuters (via KFGO), *India’s HCLTech forecasts weak annual revenue growth as clients rein in discretionary spends* (Apr 21, 2026): https://kfgo.com/2026/04/21/indias-hcltech-misses-q4-revenue-view/ - Reuters (via WHBL), *India’s Infosys forecasts weak revenue growth as AI-driven caution dents IT spending* (Apr 23, 2026): https://whbl.com/2026/04/23/indias-infosys-pegs-fiscal-year-2027-revenue-growth-at-1-5-3-5/ - Reuters (via WHBL), *India’s Infosys slumps to lowest level in three years over weak growth* (Apr 24, 2026): https://whbl.com/2026/04/23/indias-infosys-slips-on-weak-2027-growth-outlook/

Topic-selection trail

Selected from same-week AI/IT/business signals where both primary disclosures and independent Reuters market-reaction framing were available. The angle was chosen because it exposes a non-obvious contradiction with strategic implications: AI momentum and growth caution can rise together when pricing architecture is changing.