Polymarket traders lift AI-bubble odds to 27% by mid-2027

Polymarket traders raised the implied chance that an AI bubble will burst by June 30, 2027, to about 26.5% to 27%, following an increase of several percentage points in recent days. The contract briefly touched 31% during a sharp selloff in technology stocks before retreating, and has drawn more than $3.3 million in trading volume.

The higher “Yes” price did not become the market’s consensus. Most traders still backed the “No” outcome under the contract’s criteria, while the probability assigned to an AI-bubble burst by the end of 2026 stood at about 8%.

The move came amid sharp declines in AI-heavy equities and the wider Nasdaq, with Nvidia among the companies falling. Private AI valuations remained high but showed signs of cooling. On Nov. 6, OpenAI closed an initial $11 billion tranche of financing at a $1.25 trillion valuation, in a round that could ultimately reach $20 billion. The company had reduced its earlier valuation and fundraising ambitions as listed AI-related stocks weakened.

Concerns have centred on whether spending on data centres, chips and electricity can produce sufficient returns. J.P. Morgan estimated that hyperscaler capital expenditure could approach $800 billion in 2026 and exceed $1 trillion annually from 2027. Microsoft, Alphabet, Amazon, Meta and Oracle are among the companies competing for computing infrastructure, while companies have increasingly used debt to finance buildouts.

There are also countervailing signals. Leading AI laboratories have recorded rapid revenue growth and demand for advanced models has continued, factors cited by those who do not expect an imminent collapse.

The debate has long divided executives and investors. Yale School of Management’s Jeffrey Sonnenfeld and Stephen Henriques warned in 2025 that interlocking investments and commercial relationships among large technology companies could magnify the consequences if AI expectations fall short. At a Yale CEO Summit, 60% of respondents said AI enthusiasm had not produced overinvestment, while 40% expected an imminent correction.

Some industry leaders have voiced similarly qualified concerns. Goldman Sachs chief executive David Solomon said much deployed capital would fail to earn returns, while Amazon founder Jeff Bezos called the environment “kind of an industrial bubble.” OpenAI chief executive Sam Altman warned that people would overinvest and lose money. AMD chief executive Lisa Su, by contrast, described AI as a decade-long “Supercycle” capable of transforming industries.