Bitcoin Could Soar as AI Infrastructure Bubble Unravels

Arthur Hayes, chief investment officer at Maelstrom, is sticking to his prediction that Bitcoin could reach $1 million by 2030. In fact, he now believes its strongest rally will occur in late 2027 or early 2028.

Hayes' theory is based on the potential for an AI infrastructure bubble to burst as data-center revenues fail to justify massive investments. He points out that billions of dollars have been poured into data centers and computing hardware over the past few years, creating a significant mismatch between equipment lifespan and loan repayment schedules.

The Apollo research suggests that the AI buildout could require more than $2 trillion in additional investment-grade debt by 2030, with private lending covering less than $1 trillion of that amount. Hayes expects growth in AI spending to slow in the second half of 2027, with the slowdown becoming more visible in 2028.

According to Hayes, if stress builds up due to this mismatch, governments and central banks could respond by adding liquidity to the financial system. He suggests two options: the government buying computing capacity directly or stepping in to help insurers facing losses on AI-linked debt. This kind of liquidity injection would likely support Bitcoin prices.