Free tools Windows power users keep installed
One-click scans. No signup required.
The $40 trillion figure is a hypothetical loss, not a forecast: in an October 2025 Forbes article, contributor Peter Cohan applied a 78% Nasdaq decline—the scale of the dot-com bust—to a later market level. He did not disclose the starting market-cap calculation. Investors should treat the number as a stress scenario and focus instead on whether AI spending is turning into lasting customer demand, revenue and cash flow.
What does the $40 trillion figure actually mean?
Cohan’s Forbes article dated October 15, 2025, says a Nasdaq decline of 78% could erase $40 trillion in market capitalization from the level he was considering. The 78% drawdown is the article’s historical comparison to the dot-com collapse; the $40 trillion is Cohan’s conditional extrapolation, not an independently verified estimate, market consensus or prediction that a crash will happen. The article does not show the starting market-cap baseline or the calculation behind the $40 trillion figure. Cohan also cites a $3.6 trillion Nasdaq market-cap loss during the dot-com collapse; that figure, too, is attributed here to his article. Read Cohan’s Forbes scenario and assumptions.
Is AI in a bubble?
There is no simple yes-or-no answer in the cited analysis. The concern is that expectations and investment in chips, data centers and AI services may run ahead of what customers will pay for and what companies can ultimately earn. A central test is whether adoption and usage become durable revenue and profit—not simply whether firms announce large investments or AI products.
Why some analysts see bubble risks
Cohan points to high expectations, data-center construction partly financed with debt, and investment or financing arrangements linking suppliers and customers. Those arrangements can make reported demand harder to interpret if companies are funding one another rather than attracting independent buyers. INSEAD faculty likewise identify circular financing as a warning sign when it obscures whether end-user demand is real. These are risks to examine, not evidence that a collapse is inevitable.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Why the dot-com comparison has limits
INSEAD’s February 23, 2026, discussion distinguishes today’s public AI infrastructure leaders, many of which already have substantial earnings and cash flow, from some private AI startups that are burning cash. That difference matters: established profitable businesses are not interchangeable with speculative firms that depend on new funding. It does not make their stock prices immune to overvaluation, a pullback in spending or a tightening of financing. INSEAD finance professor Ben Charoenwong put the distinction cautiously: “By this framework, the current AI market is not obviously a bubble – yet.” Read the INSEAD faculty analysis.
What could happen if the boom slows?
Cohan outlines three possible paths in his October 2025 article. The percentages below are his personal scenario estimates, not objective probabilities or a consensus forecast.
| Scenario | Cohan’s estimate | What it describes |
|---|---|---|
| Continued boom | 40% | AI investment and growth continue rather than ending in a bust. |
| Soft landing | 35% | Valuations ease without the sharper funding cascade described in the downside scenario. |
| Funding or bankruptcy cascade | 25% | A failure to raise capital by a major AI company—Cohan specifically discusses OpenAI—spreads financial strain. |
The scenarios are useful as a map of possible outcomes, not as a timing tool. Even without a dramatic failure, valuations could decline gradually if growth, margins or customer spending disappoint. A sharper shock could arise if firms that depend on continued financing cannot raise it, or if spending commitments weaken across connected suppliers and customers.
What should investors monitor?
Look for evidence that AI investment is converting into independent customer demand and sustainable earnings. INSEAD highlights adoption, revenue, willingness to pay, utilization, pricing power and profits; Cohan adds company funding and financial vulnerabilities to the watch list.
- Adoption and willingness to pay: Are businesses and individuals using AI in ways that lead to recurring paid demand, rather than experimentation alone?
- Revenue and profit conversion: Are AI-related sales growing, and are those sales producing durable profits and cash flow after the cost of computing, infrastructure and development?
- Utilization and pricing power: Are data centers and other infrastructure being used effectively, and can providers maintain prices as capacity expands?
- Capital spending and financing: Is investment supported by expected customer revenue, or does it depend on borrowing and continued access to outside capital?
- Company-specific exposure: Cohan suggests watching OpenAI’s cash burn relative to revenue growth, venture funding, Nvidia customer concentration and CoreWeave debt coverage. These are indicators he identifies, not a complete checklist or a recommendation to trade particular securities.
In February 2026, INSEAD reported point-in-time valuation figures including a Shiller P/E around 40 and a Nasdaq-100 trailing P/E just over 33. They describe the period covered by that article, not current multiples; valuation snapshots should not be treated as live readings.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What can an investor do to prepare?
A headline scenario is not a personalized portfolio instruction. The practical response is to understand how much risk your finances can bear and avoid basing a plan on a single boom-or-bust prediction.
- Check concentration. Review how much of your portfolio depends on a small number of AI-linked companies, technology stocks or funds with overlapping holdings. Broad diversification can reduce dependence on one company or theme, though it cannot prevent market-wide losses.
- Match risk to your time horizon. Money needed soon generally has less time to recover from a sharp decline than long-term investments. Keep near-term spending needs separate from money exposed to market volatility.
- Review your plan before markets move. Decide in advance whether and how you will rebalance to your intended allocation. Avoid making a large change solely in response to a dramatic headline or an uncertain crash prediction.
- Use evidence, not scenario odds, to reassess. Track customer demand, revenue and cash generation alongside spending and financing conditions. Cohan’s probability estimates are his own, not a basis for assuming a particular outcome.
- Get advice for your circumstances. A qualified financial professional can help account for your goals, taxes, time horizon and tolerance for loss; no general article can determine a suitable allocation for you.
Consumer-finance coverage has also repeated chatbot-generated portfolio directions, including changing exposure across asset categories. Such generic suggestions are not individualized financial advice and should not be treated as expert consensus.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
