Arthur Hayes: AI Is Overbuilt, and Bitcoin Could Benefit from It

In a fast-moving development shaping the Finance landscape, Arthur Hayes, the former BitMEX CEO and co-founder of crypto investment firm Maelstrom, told CNBC at the Gamma Prime Investing Conference in Singapore that humanity is “wasting multi-trillion dollars” on AI data centers. Fresh reporting, according to dispatches from CryptoPotato (Crypto & Financial Markets), underscores emerging structural shifts that are drawing scrutiny across industry circles.

Arthur Hayes, the former BitMEX CEO and co-founder of crypto investment firm Maelstrom, told CNBC at the Gamma Prime Investing Conference in Singapore that humanity is “wasting multi-trillion dollars” on AI data centers. His bet is that the overbuilding ends in a crash and a bailout, and that Bitcoin and other crypto absorb the money that follows. Overbuilt, Then Bailed Out The buildout, he argued, will leave computing power “extremely cheap and extremely plentiful.” “If you study financial history and you study every single major technological rollout, it always is overbuilt. There always is a crash, and there always is a bailout,” he stated, pointing to the aftermath of the 2008 financial crisis and other episodes since. “Thankfully, we have Bitcoin and other crypto to soak up that excess liquidity, and so we know the asset that’s going to perform the best when the bailout comes,” he added, telling investors “you just have to be patient.” He claimed SpaceX, OpenAI and Anthropic are among the end users behind demand for computing power, and that none of them makes money. His contention is backed by Anthropic’s own numbers, with the firm recently sharing a prospectus ahead of a possible IPO that showed $4.6 billion in 2025 revenue, up from $400 million, against net losses near $42 billion, including a $34 billion noncash charge. Additionally, compute and infrastructure cost $7.33 billion, which was over half of $12.65 billion in operating expenses. According to Hayes, once the data centers under construction are finished, infrastructure providers will want payment for the compute Anthropic and the other AI firms committed to, which he expects in late 2027 or 2028. He did leave room for a different outcome, though, where AI could become “so useful” over the next 12 months that demand expands enough for AI companies to become profitable. Some suppliers already earn money, he noted, including the likes of Nvidia. The Debt Behind the Argument As CryptoPotato reported in September, Hayes had pointed out that compute demand from some of the biggest artificial intelligence builders backs more than $1 trillion of investment-grade debt. According to him, a downgrade would leave insurers tied to that debt short of capital, pushing Washington to buy compute as a last resort or print money to rescue them. Before that, the crypto investor had predicted that AI spending would slow next year, then contract, with bailouts bigger than 2008 that could possibly take Bitcoin near $1 million. The post Arthur Hayes: AI Is Overbuilt, and Bitcoin Could Benefit From It appeared first on CryptoPotato.

Market & Strategic Implications

Beyond immediate headlines, market participants are weighing secondary effects. The intersection of capital allocations, regulatory scrutiny, and shifting macroeconomic postures continues to elevate risk sensitivity across comparable assets and jurisdictions.

As further clarity emerges in upcoming briefings, institutional observers emphasize unit economics, policy enforcement, and counterparty exposure as primary barometers for long-term trajectory.