Ken Griffin Says AI Is "Almost Certainly" in a Bubble, But Advises Investors Not to Panic

Ken Griffin, the billionaire founder and CEO of Wall Street giant Citadel, stated on Tuesday that investors are operating "almost certainly" inside an artificial intelligence bubble—though he is stopping short of telling market participants to dump their holdings. Speaking at Citadel’s Future of Global Markets Conference in New York on October 20, Griffin offered a nuanced perspective on the historic capital expenditure boom that has come to define modern equity markets.

While acknowledging the classic symptoms of market overexuberance, Griffin emphasized that the destructive phase of a bubble often leaves behind foundational innovations that fundamentally reshape human society.

"What’s left in the wake of these bubbles, if we want to use that word, is often game-changing for the world," Griffin said. "And the companies that are on the other side of that bubble are the market leaders for decades to come."

The debate over an AI bubble extends far beyond tech-focused venture capitalists and individual traders. The massive AI infrastructure buildout has heavily underpinned broader equity markets over the past year, meaning countless index funds, pension funds, and retirement accounts hold a direct slice of the sector. A market bubble inherently materializes when asset prices climb far faster than the underlying businesses can economically justify, creating a temporary disconnect until economic reality eventually catches up. Citadel operates as a premier hedge fund, a private investment firm that manages capital for wealthy individuals and massive institutional clients.

Despite the obvious parallels to past market manias, Griffin argues that the emergence of a bubble is a natural byproduct of technological revolutions rather than a reason for immediate panic. Drawing a parallel to the runaway internet boom of the late 1990s—noting with hindsight that he wishes he had kept his early Amazon stock—he explained that radical economic transformations are almost always intertwined with periods of speculative excess.

The companies that successfully navigate the turbulence and emerge on the other side tend to dominate global markets for generations, he argued, though he stopped short of forecasting an imminent market collapse. Paradoxically, Griffin’s own firm stands as a prime case study of the technology’s sweeping utility. He revealed that Citadel utilizes advanced AI agents—sophisticated software capable of carrying out complex, multistep tasks autonomously—to slash internal research timelines that once took weeks down to a mere couple of hours.

The conversation around AI valuations inevitably leads back to the underlying hardware powering the revolution. Griffin drew a direct economic comparison between Graphics Processing Units (GPUs)—the specialized semiconductor chips that companies rely on to train and execute advanced AI models—and airline seats. Both commodities share a very low marginal cost to run, meaning the incremental expense of serving one additional customer is practically negligible. Consequently, when overall market supply eventually outstrips real-world demand, pricing structures can face steep and rapid corrections.

To illustrate this market dynamic, Griffin pointed back to the telecommunications boom around the year 2000, an era when corporations laid far more fiber-optic cable than immediate demand required, causing communication prices to plummet. He also expressed skepticism that older generations of chips, which currently command rental rates he described as extraordinary, will maintain that same pricing power for another five or six years.

Corporate buyers are becoming increasingly discerning, Griffin observed. Businesses are sorting carefully through various options to determine which complex tasks genuinely justify the premium subscription models offered by labs like OpenAI and Anthropic, and which workloads can easily run on cheaper open-source alternatives—flexible models that any enterprise can download and operate on their own local computer infrastructure. Corporate America’s ability to extract tangible value from artificial intelligence has been uneven at best, he added, noting that executives are frequently studying steep software bills from AI labs and asking where the promised productivity gains actually are.

Market forecasts underscore the staggering scale of the financial commitments currently flowing through the global economy. Gartner has projected that worldwide AI spending will approach nearly $1.5 trillion for 2025, a massive tally that encompasses everything from high-end semiconductors and dedicated data center servers to consumer AI-enabled smartphones. Meanwhile, Nvidia CEO Jensen Huang asserted during an appearance at Davos in January that the industry is still only "a few hundred billion dollars" into what he characterized as the largest infrastructure buildout in human history, with trillions more in investment required. If corporate customers ultimately balk at premium pricing models, the chipmakers and cloud providers relying on that continuous capital expenditure will feel the impact first.

The delicate interplay between massive leverage, market corrections, and opportunistic buying was vividly illustrated earlier this year. On July 30, Situational Awareness, a specialized hedge fund founded by former OpenAI researcher Leopold Aschenbrenner, was forced to liquidate the vast majority of its public stock portfolio. The fund sold its holdings to Citadel at roughly a 10% discount after the portfolio’s value dropped by approximately 67% over the course of July.

The boutique fund had aggressively borrowed as much as four times its core capital to snap up AI-related stocks—a classic trading tactic known as leverage. As asset prices tumbled, the fund’s prime lenders demanded additional collateral on a daily basis. Selling assets because creditors demand it rather than by strategic choice is known as forced selling, a self-reinforcing phenomenon where each forced liquidation drives market prices lower, which in turn triggers fresh collateral calls elsewhere.

Citadel subsequently shed more than 80% of the aggregate risk associated with that specific portfolio, as Griffin noted in an August letter to clients. In essence, the prominent Wall Street figure publicly warning of a market bubble was simultaneously positioned to scoop up heavily discounted assets during a moment of acute market distress.

Griffin has also kept a watchful eye on broader macroeconomic risks that could intersect with the tech sector. Earlier this year, he warned that if the Strait of Hormuz—a critically important oil shipping lane—remained closed for an extended period of six to 12 months, global energy prices would rise materially and potentially push the broader world economy into a recession.

Meanwhile, Wall Street analysts anticipate that capital expenditure growth among the largest AI infrastructure builders will experience a notable deceleration. Research published by Goldman Sachs in December indicated that spending growth among major tech firms pouring capital into chips and data centers is expected to slow from 75% in the third quarter of 2025 down to 25% by the end of 2026.