NextFin News - India's software exporters are being positioned to outperform their AI-exposed global peers by the widest margin in years, as the sector's valuation resets to levels below every historical median while earnings expectations trough. The Nifty IT index trades at a price-to-earnings ratio of 17.45 as of early October 2026, roughly 36% below its five-year median of 27.11 and about 46% cheaper than the 32.20 multiple carried by the S&P 500 information technology sector. The setup is a bet that the worst of the artificial-intelligence disruption panic has already been priced in - and that the companies most feared as AI's victims are cheap enough to beat the very stocks investors bought to own the AI revolution.
The Setup: A Sector Priced for Extinction
The numbers behind the reset are severe. The combined market capitalization of India's five largest software exporters - Tata Consultancy Services, Infosys, Wipro, HCL Technologies and Tech Mahindra - fell more than 46% to 18.15 lakh crore rupees in July 2026, down from a record 33.71 lakh crore in August 2024. The Nifty IT index has been the worst-performing sector of 2026, and in late April shed roughly $26 billion of market value in a single week after earnings from TCS and Infosys disappointed investors. Foreign portfolio investors, who had been net buyers of Indian equities, pulled about 110 billion rupees ($1.21 billion) out of IT stocks in the first half of February 2026 alone, following a record 750 billion rupees of net selling across 2025.
The trigger for the rout was a growth story that broke. TCS posted its first annual revenue decline in more than two decades. Infosys, HCLTech and Wipro trimmed their fiscal 2027 revenue-growth forecasts. The National Association of Software and Services Companies still projects the broader Indian technology industry will grow 6.1% to $315 billion in fiscal 2026, but analysts expect the large listed exporters - which derive more than half their revenue from the United States - to manage only about 3% to 4% near-term growth. The sector had last reported double-digit revenue growth in the March 2023 quarter.
The valuation gap that opens from this damage is the heart of the thesis. At 17.45 times earnings, the Nifty IT index sits below every historical median window tracked - one-year (21.01), three-year (27.52), five-year (27.11), seven-year (27.12), ten-year (25.13) and fifteen-year (21.53). Against the S&P 500 information technology sector's 32.20 multiple, Indian IT trades at roughly a 46% discount. That is the kind of spread that, historically, has preceded mean-reversion rallies even when the fundamental backdrop remains imperfect. A sector does not need to return to glory to reward investors who buy it at a 46% discount to its peers; it only needs to stop getting worse.
Why the Panic Overshot: Cyclical Demand Freeze Meets Structural Margin Pressure
The first question is whether the selloff reflects a cyclical demand pause or a permanent impairment of the business model. The evidence points to both - and that distinction is what makes the trade work.
On the cyclical side, clients froze discretionary technology spending amid macroeconomic and geopolitical uncertainty. The U.S., which accounts for more than half of revenue at most large Indian IT firms, saw softer deal pipelines, while uncertainty around immigration policy and tariffs delayed long-term technology commitments. Deal cycles stretched to 12 to 18 months. This is classic cyclical behavior: spending deferred, not cancelled; projects paused, not terminated. The tell is in the bookings data. Wipro's large-deal bookings rose 12.9% sequentially to $1.63 billion in the first quarter of fiscal 2027, and total bookings reached $3.37 billion. Accenture, widely viewed as a bellwether for Indian IT, forecast stronger-than-expected annual revenue growth in early October and reported record bookings of $22.1 billion - a signal that enterprise technology spending is stabilizing rather than collapsing. The Nifty IT index rose roughly 1.6% on October 5 on that read-through.
On the structural side, the threat is real but smaller than the panic implied. Generative AI is automating tasks that Indian IT services firms historically billed by the hour, shifting pricing from effort-based models toward outcome-based contracts. Brokerage ICICI Direct estimates AI-led revenue deflation of 2% to 3% per year in traditional IT services for fiscal 2027 and 2028. That is a real headwind, but it is not an extinction event. Meanwhile, the same transition opens an incremental AI-led total addressable market estimated at $300 billion to $400 billion by 2030, against a current Indian IT services industry of roughly $280 billion.
The critical data point: AI revenue is still small.
Revenue from AI is growing at a fast pace, but it's coming off a very low base and is hardly 5% of total revenue,said Piyush Pandey of Centrum Broking, noting that AI was weighing on pricing particularly in legacy contracts. Infosys disclosed that AI-related revenue made up 8.2% of its total in the June quarter, while TCS reported annualized AI revenue of $2.6 billion. The companies are not being displaced by AI - they are selling it, albeit from a small base.
The Second-Order Read: What the Market Has Not Priced
The consensus view - that Indian IT is cheap because growth is broken - is already reflected in the 17.45 multiple. The second-order question is what happens when the earnings trough confirms and the multiple has nowhere to go but up.
Consider the mechanics. A stock's return decomposes into earnings growth plus multiple change. At 17.45 times earnings versus a 27-plus historical median, the multiple leg alone offers roughly 55% of upside if the sector merely normalizes - without a single rupee of earnings growth. If the large exporters deliver even 3% to 4% revenue growth while AI revenue scales from its 5% to 8% base, the earnings leg adds to that. The asymmetry is that the downside is capped by how cheap the stocks already are, while the upside re-rates the entire earnings stream.
The cross-asset transmission channel runs through global AI leaders. For two years, capital rotated out of Indian IT and into U.S. AI infrastructure and software names - the trade that made the S&P 500 technology sector's 32.20 multiple sustainable. As AI spending shifts from experimentation to deployment, the beneficiaries broaden beyond chip and platform vendors to the system integrators who actually implement AI inside enterprises. That is Indian IT's core competence. The rotation does not require Indian IT to win the AI race; it only requires that the market stop pricing them as if they are losing it entirely.
There is also a currency cushion the panic ignored. Analysts had expected a falling rupee to boost sector revenue by roughly 10%, and while that did not fully materialize, a weaker rupee still flows directly to margins for exporters who earn in dollars and pay costs in rupees. Wipro's operating margin fell 130 basis points sequentially to 16.0% in the first quarter - a trough-like print that gives margins room to recover even before any pricing power returns.
The Strongest Case Against the Trade
The bear case is not weak, and it deserves its weight. Morgan Stanley downgraded TCS to equal-weight, arguing that the stock's premium to Accenture had risen above 40% and that valuations for the broader group remained at risk. The structural argument runs deeper than a 2% to 3% deflation estimate: agentic AI systems can increasingly perform the coding, testing, and maintenance work that has been the backbone of Indian IT services for three decades. If the deflation accelerates to 5% or more annually, no multiple is cheap enough. The cost-arbitrage model that built the sector is being leveled by the very technology that made automation cheap.
Global capability centers - captive units that multinational corporations set up inside India - compound the pressure by insourcing work that used to go to vendors. The sector employs about 5.9 million people, and any meaningful job reduction would feed back into lower domestic consumption and political pressure. A bear who sees agentic AI compressing billable work by 5% a year has a coherent case that 17.45 times earnings is not a floor but a value trap.
The answer to the bear case is timing and price. Even if the structural headwind is real, a 46% drawdown in market value and a 36% discount to the five-year median multiple prices in a far worse outcome than the 2% to 3% deflation that analysts model. The trade is not a bet that AI is harmless; it is a bet that the harm is known, quantified, and already reflected in prices that assume the model never recovers.
For Indian IT services, elevated booking growth and continued revenue conversion, particularly in outsourcing, provide a positive read-through on enterprise IT spending,PL Capital noted after Accenture's results.
What to Watch: The Signals That Decide the Call
The second-quarter fiscal 2027 earnings season is the first real test. TCS reports October 8, HCLTech on October 12, and Infosys on October 23. Three metrics matter more than the headline numbers. First, deal conversion: whether the strong bookings reported by Wipro and Accenture are turning into revenue. Second, AI revenue disclosure: whether the 8.2% share reported by Infosys is rising toward double digits while total revenue stabilizes. Third, guidance: whether fiscal 2027 growth forecasts are cut again or held.
The falsifying signal is specific. If TCS and Infosys report AI revenue above roughly 10% of total while overall revenue still contracts, and the Nifty IT multiple expands back above 22 times earnings while organic growth remains below 3% for two consecutive quarters, the thesis breaks - that combination would confirm that AI is a net revenue destroyer, not a transition cost. Until then, the risk-reward favors the cheap side of the trade.
Conclusion: A Cyclical Re-Rating on Trough Earnings, Not a Return to the Old Model
The forward view splits cleanly by time horizon. In the short term, sentiment and liquidity dominate; the sector can remain oversold if foreign selling continues or if U.S. discretionary spending stays frozen. In the medium term, fundamentals should improve as bookings convert and the earnings trough confirms - this is where the multiple re-rating plays out. In the long term, the sector's growth model is structurally different: lower headcount growth, higher AI-enabled productivity, and a shift from effort billing to outcome pricing. Investors who expect a return to the double-digit growth of the March 2023 era will be disappointed; investors who price the transition correctly can still capture a cyclical re-rating.
The base case is a gradual recovery: 3% to 4% revenue growth in fiscal 2027, margins stabilizing in the mid-teens, and the Nifty IT multiple drifting back toward its 21 to 24 range as the AI panic recedes. The upside case requires deal conversion to accelerate and AI revenue to scale toward 10% of the mix without further pricing damage - a path that would push the index well beyond its historical median multiple. The downside case is the bear thesis realized: AI deflation accelerates beyond 3%, guidance is cut again, and the 17.45 multiple proves to be a value trap rather than a floor.
Who benefits and who is exposed is asymmetric. The large-cap exporters - TCS, Infosys, HCLTech - offer the cleanest re-rating play because they carry the deepest valuation discounts and the strongest balance sheets. Mid-tier firms with stronger digital and AI exposure, such as LTIMindtree and Persistent Systems, may outperform on growth but carry more execution risk from acquisitions. The exposed are the investors who sold at the bottom after pricing in an extinction scenario that the earnings data does not support.
The market has spent two years treating Indian IT as the losing side of the AI trade. The numbers now say the price of that fear has become the opportunity - and the widest valuation gap in years does not need a perfect recovery to pay, only a less-catastrophic one than the one already priced in.
Explore more exclusive insights at nextfin.ai.
