When did the dot-com bubble burst? The dates the AI bubble is compared to

On what date did the dot-com bubble burst?

The Nasdaq Composite reached its peak of 5,048.62 on Friday 10 March 2000 (Wikipedia, Dot-com bubble). Most writers therefore date the burst to March 2000, because that was the last day the index stood at its high. Prices then began a long slide, so the peak is the first marker on the timeline and the starting point of every later measurement.

A peak is recognised only afterwards. On 10 March 2000 nobody could prove it was the top, and many buyers saw a pause. The same rule applies to AI: a date becomes the peak when later prices stay below it. A timeline built after the event is a record, while a timeline built during the boom is a hypothesis that later prices confirm or reject.

Which dates mark the dot-com bubble burst from peak to low?

Five dates carry the story. The peak came on 10 March 2000. On 20 March 2000 Barron’s published Burning Up, a warning that internet companies were running out of cash fast, and the same article reports that the Federal Reserve raised rates the day after. On 3 April 2000 the Microsoft ruling cut 8% from the Nasdaq in one session, about 350 points.

By November 2000 most internet stocks stood 75% below their highs, and by October 2002 the Nasdaq Composite had lost 78% from its peak. Counted from March 2000 to October 2002, the slide lasted 2 years and 7 months. Each date adds a different kind of evidence: sentiment, policy, a shock to a leader, and a final low.

Does the dot-com bubble burst have one official date?

Sources choose their own date, and each choice answers a different question. Market historians pick the Nasdaq peak of 10 March 2000. Journalists often pick 20 March 2000 and the Barron’s warning. Lawyers and economists sometimes pick 3 April 2000 and the Microsoft ruling. All three fall within 4 weeks, which is why March and April 2000 form one burst window.

A useful habit is to name the date you use and the reason. A trader asks when prices peaked, a risk manager asks when funding turned, and a strategist asks when behaviour changed. Stating the question first keeps comparisons clean, and it keeps the comparison with the AI bubble honest, because you must make the same choice for AI.

What did the years after the dot-com burst show?

Failures followed the peak in sequence. Wikipedia names Pets.com, Webvan, and Boo.com among the companies that closed, and it reports that most internet stocks had fallen 75% by November 2000, which wiped out $1.755 trillion of value. The pattern is typical of a burst: weaker companies fail first, and the index keeps falling while stronger companies are repriced.

By 2004, 48% of dot-com companies still operated, at lower valuations. Those dates show that a burst is a process of 2 to 3 years, not a single day. Readers who want the index levels can follow the dot-com crash and the Nasdaq, which tracks how far prices fell and where the slide stopped.

Which dates does the AI bubble get compared to?

Analysts map AI onto the dot-com calendar in three ways: the run-up before the peak, the warning phase, and the shock phase. As of October 2026 the EQOS Asia AI study (April 2026) records a ×4 valuation rise over 18 months, which describes a run-up, with no confirmed peak. Placing AI just before or just after a peak is a judgement, and later prices check it.

The study also lists a regulatory date, AI Act compliance on 2 December 2027, which can work as a shock marker in the way the Microsoft ruling did in 2000. A shared calendar helps discussion, and it also invites false precision, so treat every mapped date as a hypothesis that you test against new data.

How can you watch for a trigger date in the AI bubble?

Build a watch list from the 3 dot-com triggers: a public warning about cash, a change in borrowing costs, and a shock to a leading company. For each trigger, name a source you read every month and a threshold that would change your plan. The watch list turns the dot-com calendar into early-warning questions that you answer with data.

Add the quarterly rhythm that earnings create. Funding news, company results, and rule changes arrive on a schedule, so a review each quarter catches most triggers. Record the answer to each question with its date. After 4 quarters you hold a dated log, which is the same kind of record that lets historians date March 2000.

How do dot-com dates fit into an AI bubble plan?

Use the dates as scenario markers: a peak, a warning, a shock, and a low about 2 to 3 years later. Size commitments so that your business keeps operating through a fall as deep as the 78% of the Nasdaq Composite. The EQOS Asia method page explains how to turn such scenarios into a repeatable score.

The Λ simulator estimates a φ(S) score in 10 minutes, and a 30-minute pre-diagnostic applies the same scoring to a decision in Singapore, ASEAN, or Hong Kong. This page is general information, not investment advice, and the dates above describe the past rather than a forecast for AI.

Frequently asked questions

About the author

Stéphane Jambu

President, EQOS Forecast Asia

Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile