The use of the word Gen AI spiked in November 2022 with the release of ChatGPT. Adoption went viral with 100 million users in two months. By the last quarter of 2025, 62% of global CEOs mentioned AI at least once in their earning call. On the back of the Gen AI frenzy, the Dow Jones moved up from 33,712 to 49,920 by January 2026.
Equity valuations are stretched, and the cyclically adjusted price-to-earnings ratio of the S&P 500 is hovering near 40, reminiscent of the dot-com boom of 2000. American household savings have moved into equity at a dizzying pace — up from 27% to 52%. The magnificent seven — Apple, Microsoft, Amazon, Nvidia, Alphabet, Meta, and Tesla — account for roughly one-third of the index’s total market capitalisation. OpenAI, for instance, is estimated to be valued at $500 billion, but has only $10 billion in revenue and reported $15 billion in losses in 2025.
In the first two months of 2026, the first signs of market nervousness became apparent when the market wiped out $1.25 trillion of market capitalisation when five of the magnificent seven announced a combined investment of over $680 billion for 2026. The contraction is greater than the GDP of over 175 countries in the world, including the likes of Switzerland, Taiwan, Sweden, Singapore and Israel.
The issues go beyond the mismatch between investment and return. Many of the new technology firms share the same entities as investors, suppliers, and customers — a tightly interlinked structure where a single shock could rapidly cascade across balance sheets. Another red flag is the growing use of special purpose vehicles (SPVs) by both big-tech and new-tech firms. While marketed as flexible financing tools, SPVs often obscure true risk exposure and allow leverage to build out of sight. At an ecosystem level, the imbalance is even starker: Global spending on AI-related data centres exceeded $400 billion in 2025 — more than six times the roughly $60 billion in direct AI services revenue. Clearly, expectations, not fundamentals, dominate pricing.
We do not doubt the deeply transformative promise of Gen AI, but that is almost beside the point. Gen AI is here to stay. It will fuel a productivity revolution beyond easy imagination; only its current valuations have run ahead of the fundamentals. History shows that revolutionary technologies, from railroads to the internet, often generate bubbles before their true economic value is realised. In the short run, technology is overestimated. In the long run, it is underestimated. With GenAI, the acceleration curve is steeper than prior waves. The “long run” may arrive faster. The real risk is not whether GenAI creates value. It will. The greater risk lies in how governments, companies, and investors respond when the market turns. The initial winners and losers will change. There will be a big, short-term impact. And the short term will need to be lived through.
Today, risk has migrated beyond traditional banking. Post-crisis regulations have tightened bank lending, but lightly regulated private credit and equity instruments have filled the gap. The investor base for these products increasingly includes retail-exposed entities such as insurance companies, pension funds, and even 401(k) plans. Adding to the concern, data-centre debt is now being sliced into asset-backed securities and sold to retail investors — an uncomfortable echo of the financial innovations that preceded the global financial crisis.
History offers sobering lessons. The dot-com crash, often described as mild in hindsight, still triggered a prolonged bear market. Between 2000 and 2002, the S&P 500 fell nearly 50%, capital expenditure collapsed, and unemployment rose from 3.9% to over 6%. The Federal Reserve cut rates aggressively — 11 times in 2001 alone — yet it took years for confidence to return. The global financial crisis was far worse, wiping out 57% of the S&P 500’s value and pushing US unemployment to 10%. Globally, GDP contracted for the first time in decades, with Europe and Russia suffering deep recessions.
Countries and companies must prepare proactively for this eventuality. If the market were to correct by 20%, then the likely scenario can be a sharp cutback in US consumer spending, given the extent of household exposure to the market. The dotcom bubble burst, and the global financial crisis both saw retail investors exit equity markets and flow funds to safe assets like bonds/ money market funds. Even when markets recovered, retail investors burned from the crisis, did not aggressively buy the dip.
If the correction were to happen, central banks have a more complex macro backdrop to deal with. Inflation remains structurally higher than in the 2000s. Central banks may not have the same flexibility to cut aggressively without risking renewed inflationary pressures. If rate cuts are constrained, traditional “flight-to-safety” dynamics may not operate as cleanly. If cuts do occur, they may coincide with credit contraction and a temporary pause in AI investment.
Both countries and companies should prepare alternative scenarios for the market. Companies with strong balance sheets will be able to take advantage by picking up AI assets and talent cheaply. But for this, they will have to be ready. They will need to pick from their AI experiments and scale some of them, and start a broader upskilling of their organisation. It will not be a time to go back into a shell but to move forward with purpose, prudence and confidence. Past crises show that firms willing to invest during downturns — particularly in productivity-enhancing technologies — always emerge stronger. The winners will be those prepared with talent, capital, and conviction to embrace disruption when others retreat.
At the recently concluded India AI Impact Summit 2026, India established itself as the Global South leader on AI policy. However, Summit pronouncements need to translate into actions. A correction will offer India an opportunity to attract talent and strengthen its economic resilience by building centres of excellence. It should continue to accelerate regulatory reform and manage the correction with prudence to attract some foreign capital and invest in infrastructure. Asia weathered the last two major crises better than the West; a stronger, more prepared India could do even better next time.
Every major transformation has a messy middle. The mistake will be to misread the noise to be the signal.
Janmejaya Sinha is chairman, India, BCG and Neha Gupta is managing director and partner, BCG. The views expressed are personal
