Has the AI Bubble Burst? What It Means for the Indian Stock Market

A few weeks ago, artificial intelligence was driving markets worldwide. Money poured into AI-linked companies. Valuations climbed. Countries at the centre of the AI boom became investor favourites.

That trade has now cooled sharply. And the question closer to home is simple. Could a slower global AI trade actually help the Indian stock market?

What changed in global markets

The shift began around 22 June 2026. Asia took the hardest hit.

South Korea's Kospi hit a record high of 9,385.59 on 19 June. It then fell to about 5,520 by late July. That is a drop of roughly 40% at the low. As of 7 August, the index sits near 6,259, still about 31% below its peak.

The Nasdaq Composite peaked at 27,190.21 on 1 June. It closed at 24,442.94 on 29 July. That marks a 10.1% fall and a second correction for 2026.

Japan, Taiwan and mainland China also corrected over the same weeks. Hong Kong was the clear exception and gained ground.

The AI leaders fell hardest. Chip stocks lost more than $1 trillion in market value during the July selloff. The iShares Semiconductor ETF fell 23% from its high in a month. South Korea's SK Hynix dropped about 35% in July alone. Alphabet was roughly flat over one month. Meta is down close to 10% in 2026.

One trigger stood out. A report suggested OpenAI may delay its listing to 2027. That deepened worries about how much value now sits in a handful of AI names.

Why India is drawing fresh money

India held steady while AI-heavy markets fell. The Nifty 50 and Sensex gained over 2% in July. Both ranked among the better global performers that month.

Capital flows explain much of it. For months, foreign investors favoured markets tied directly to the AI boom. As that enthusiasm faded, money started to move.

Foreign portfolio investors bought a net $2.05 billion of Indian equities in July 2026. That was the highest inflow in Asia. NSDL data shows FIIs bought over Rs 15,000 crore of Indian shares that month. It snapped a selling streak that had run more than four months.

The outflows elsewhere were far larger. South Korea saw $30.5 billion leave in June alone. That was its worst month in over 25 years. Taiwan saw an $18.3 billion exit in the same month. In July, Taiwan lost another $9.8 billion and South Korea another $7.7 billion.

One number deserves care. India itself lost more than $27 billion in the four months to June 2026. So July marked a reversal, not a long run of strength. Investors tracking this should learn to read daily FII and DII data rather than react to single months.

Is this another dot-com moment?

Comparisons with the late-1990s dot-com bubble are growing.

That crash wiped out close to $5 trillion in market value. It also killed many companies. Yet it produced Amazon and Google, which went on to shape the modern internet.

Today's cycle differs in one key way. Most leading AI firms have real revenue and strong cash flows. Many internet firms of 1999 had neither. So a price correction is not the same as a broken technology.

The real debate is about valuation, not usefulness. The question is whether current prices justify the enormous capital being spent. SAHI's breakdown of AI capex at SpaceX shows how heavy that spending has become.

India is not immune to the AI trade

India benefits from the rotation. But it carries its own AI risk.

Indian IT is the direct exposure. In February 2026, foreign investors sold $1.85 billion of Indian IT stocks. The Nifty IT index fell 19.5% that month. It was the sector's worst month since September 2008. The index later staged a July rally as the AI trade reversed.

The Reserve Bank of India Governor has warned publicly on this. A sharp fall in global AI-linked technology stocks could spill into Indian markets. That is a supervisory view, not a forecast.

The revenue risk is measurable. Motilal Oswal estimates AI could remove 9% to 12% of Indian IT industry revenue over the next four years. Other analysts see about 2% to 3% annual deflation in traditional IT services. Sector growth is expected to stay muted through FY27 and FY28.

What this means beyond stock prices

Jobs are the second channel. Stanford's Digital Economy Lab studied US payroll data in its paper "Canaries in the Coal Mine". Employment for software developers aged 22 to 25 has fallen nearly 20% from its late-2022 peak. After controlling for company-level shocks, the drop was about 16%.

There is a possible upside too. If large US technology firms turn cautious on spending, more high-end work could move to India. That pattern followed earlier global downturns.

India has recognised the stakes. The Cabinet approved the IndiaAI Mission in March 2024 with an outlay of ₹10,371.92 crore over five years, as recorded by the Press Information Bureau. Execution is the gap. Only about ₹400 crore had been released two years in. Money alone will not build research depth.

Mistakes investors make in a rotation like this

• Reading one month as a trend. July's $2.05 billion inflow followed $27 billion of outflows. One month proves little.
• Confusing a price fall with a failed technology. AI adoption and AI valuations are separate questions.
• Buying fallen AI names for the discount alone. A 35% fall after a 10x run is not automatically cheap. Weigh it against India's own AI-linked stocks first.
• Treating India as a safe haven. Indian IT earns heavily from US technology budgets.
• Ignoring leverage. Korea's crash was driven by margin debt, not only by AI doubt.

What to watch next

The cooling AI trade does not mean the end of AI as an investment theme. It may simply mark a change in where global money goes next.

Three things will decide whether the shift lasts. Whether FII buying in India continues past a single quarter. Whether US technology capex holds up. And whether Indian IT earnings stabilise through FY27.

For now, India looks better placed than most Asian peers. That position still has to be earned quarter by quarter.

Sources: Press Information Bureau (IndiaAI Mission approval), Reserve Bank of India, NSDL foreign portfolio investment data, Stanford Digital Economy Lab ("Canaries in the Coal Mine"), Motilal Oswal, and exchange data for the Kospi, Nasdaq Composite, Hang Seng, Nifty 50 and Sensex. Index levels as of 7 August 2026.