Indian IT is again facing the type of question that markets dislike the most: is this only a cyclical slowdown, or is the business model itself getting disrupted? The Nifty IT index is now trading close to 19 times trailing earnings, well below its roughly 25x decade median. On the surface, this looks like a standard valuation correction. But the reason behind the correction is more interesting: investors are worried that artificial intelligence could reduce the need for human software engineers, thereby weakening the traditional billable-hours model that powered Indian IT for more than two decades.

This is not the first time the sector has faced such fear. The last comparable episode was the 2001-03 period, when the dot-com crashed and the end of Y2K demand pushed Indian IT valuations down to nearly 12-15x earnings. At that time, investors believed the sector had lost its biggest growth engine. Yet, the same downturn eventually created a larger opportunity: global outsourcing. Today, the market is asking a similar question in a new form. Could AI become the next demand engine, or will it permanently compress the revenue pool?

Exhibit 1: Nifty IT Index - Historical PE Ratio (2001-2026)

Source: NSE, Bloomberg consensus estimates, analyst estimates. PE data for 2001–2004 approximated from individual company multiples; Nifty IT index PE series from 2004 onward. Current = May 2026.

The last big trough: what happened after the dot-com burst

The 2001-03 derating was not a normal market correction. Indian IT had entered the 2000s after two powerful demand tailwinds. The first was Y2K, where companies globally needed to fix old software systems before the year 2000. The second was the dot-com boom, where every internet company and large Western enterprise wanted new software systems built quickly. Indian engineers became extremely valuable because they could deliver these projects at scale and at a much lower cost.

During that boom, valuations moved far ahead of fundamentals. The IT index traded at 40-60x earnings at the peak, while companies such as Infosys delivered exceptionally high profit growth. But when the Nasdaq collapsed, global clients froze discretionary technology budgets. New projects were postponed, billing rates came under pressure, and the demand that had looked permanent suddenly disappeared. The sector's PE multiple compressed sharply to 12-15x by FY2002-03.

Why the recovery was stronger than expected

The recovery after 2003 did not come because Y2K returned or because dot-com spending revived. It came because Western companies, under cost pressure, started to treat offshore outsourcing as a mainstream board-level decision. A software developer in the US could cost several times more than an equivalent offshore resource in India. For large enterprises facing weak growth and margin pressure, shifting IT work to India became a practical cost-saving lever.

This changed the nature of the Indian IT opportunity. The sector moved from one-time projects to multi-year outsourcing contracts with large global companies. BFSI, manufacturing, telecom and retail clients began shifting application development, maintenance, back-office work and support functions to Indian vendors. The addressable market became much larger and more durable than the temporary Y2K/dot-com demand pool. That is why the sector not only recovered, but also entered a long structural growth phase.

Currency also helped. Indian IT companies earned revenues largely in dollars while a meaningful part of their cost base was in rupees. A weak or stable rupee improved margins and gave the sector an additional cushion during the recovery. As earnings visibility improved, the PE multiple moved back from trough levels towards the low-to-mid 20s by FY2006-07.

Today's anxiety: AI is the new structural question

The current situation looks similar in sentiment but different in mechanism. In 2001, the fear was that demand had disappeared. In 2026, the fear is that AI could change the supply side of software development. If AI tools allow one engineer to do the work of two or three, clients may demand productivity benefits, lower pricing or smaller teams. That directly challenges the old revenue model where growth was closely linked to headcount addition and billable effort.

This explains why the valuation correction is more than a macro slowdown story. The sector is not facing a balance sheet crisis. Large IT companies still have strong cash positions, healthy dividends and strong client relationships. Deal wins have not collapsed in the way project demand collapsed during the dot-com burst. Yet the PE multiple has compressed because the market is unsure whether AI-led productivity will expand the opportunity or reduce the need for traditional services.

The parallel investors should focus on

The closest parallel with 2001-03 is not the exact PE level; it is the nature of uncertainty. In both phases, investors were forced to value the sector before knowing what the next demand pool would look like. In 2001, the replacement pool turned out to be global outsourcing. In the current cycle, the potential replacement pool could be AI implementation, model integration, data engineering, governance, cloud modernisation and workflow redesign.

Large enterprises may not be able to simply plug AI into old systems and achieve productivity overnight. They will need to clean data, rebuild workflows, integrate AI tools with legacy systems, manage compliance and train users. These are complex, multi-year transformation projects — exactly the kind of work Indian IT companies are built to execute. Therefore, the bull case is not that AI will have no impact. The bull case is that the implementation opportunity may offset, or eventually exceed, the productivity compression in traditional services.

What is different this time

However, the comparison should not be stretched too far. AI is a different kind of disruption. Outsourcing expanded the use of human engineers; AI may reduce the number required for the same output. That makes revenue-per-employee and pricing trends more important than in earlier cycles. If clients capture most of the productivity gains, revenue growth could remain muted even if project activity improves.

The valuation starting point is also different. The 2001-03 trough happened at 12-15x on depressed earnings. The current 19x PE is lower than the sector's recent average, but not as distressed as the dot-com trough. In other words, the market is already pricing concern, but not complete structural destruction. This makes the next few quarters important. Investors need evidence on whether AI is creating new deal pipelines or only causing revenue compression in existing work.

Key takeaway

For investors, the main conclusion is that low valuation alone is not enough. A sector trading below its historical average can stay cheap if earnings visibility remains weak. The more useful question is whether a new growth engine is forming. In 2003, that engine was outsourcing. In 2026, it may be enterprise AI transformation — but the proof must come through order wins, revenue conversion, stable pricing, margin protection and improving commentary from global clients.

The PE chart therefore should be read as a sentiment map. At around 19x, Indian IT is not at a crisis valuation, but it is clearly trading with structural doubt. History shows that such phases can create attractive opportunities if the feared disruption opens a larger demand pool. It also shows that investors must be patient until the earnings cycle confirms the narrative. The next rerating in Indian IT will not come merely from saying AI is an opportunity; it will come when companies show that AI-led work can become a scalable and profitable revenue stream.

2001-03 vs 2025-26: the clean comparison