BitMEX co-founder Arthur Hayes has proposed that the AI debt problem presents a bullish scenario for Bitcoin, despite potential headwinds for technology companies facing slower AI development.
In his essay titled "Safety First," Hayes argues that reduced AI development could decrease demand for computing power, straining the substantial debt financing AI data centers and infrastructure. He links this debt exposure to private credit and insurance companies, suggesting that weaker AI cash flows could trigger credit downgrades and expose capital shortfalls in the insurance sector.
Two Paths to Government Intervention
Hayes identifies two possible government responses to such a situation. First, the US could become a "buyer of last resort" by guaranteeing demand for AI compute and supporting data-center investment, particularly if AI is deemed strategically important in competition with China.
Alternatively, if falling AI-related asset values create problems for insurers and financial institutions, the government could intervene to prevent a broader financial crisis.
Cryptocurrency Gains From Either Scenario
According to Hayes, both scenarios would involve greater government borrowing and monetary expansion. Increased government spending and liquidity would expand the money supply in financial markets, potentially encouraging investors to move into assets such as Bitcoin and other cryptocurrencies.
Hayes stated that "the amount of dollars in the system will continue to grow, and Bitcoin and selected shitcoins will rise in price."
The AI Financing Challenge
The AI sector is already responsible for nearly 40% of new long-term investment-grade corporate debt, according to reports cited by Hayes. Estimates from Apollo suggest the AI sector could support more than $2 trillion in additional high-quality debt. However, public bond markets may only be able to absorb less than $1 trillion by 2030 due to concentration and rating limits.
Hayes also noted that government support for AI infrastructure could create excess computing capacity, eventually making compute cheaper and encouraging greater AI adoption.
