Arthur Hayes, former BitMEX CEO and current CIO of Maelstrom, warned that the global race to dominate artificial intelligence is feeding an unprecedented liquidity bubble built on fiat credit expansion. His argument links AI infrastructure spending, government debt issuance, bank lending and wartime-style national investment into one broader cycle of systemic risk.
Hayes described Bitcoin as a “liquidity smoke alarm” for stress in the credit system, arguing that the asset is already beginning to price in future central-bank rescues. In his view, expected liquidity injections from the Federal Reserve and other central banks could reflate risk assets, including cryptocurrencies.
AI Nationalism Moves Risk Into the Banking System
Hayes framed the AI boom as a structural financing shift. At first, hyperscalers funded data centers, chips and energy infrastructure from internal cash flow, but he said AI financing is increasingly moving toward bank credit and government debt.
That migration matters because it transfers risk from corporate balance sheets into the wider financial system. If AI revenue projections or employment assumptions fail to materialize, Hayes argued, banks could be left carrying the credit stress tied to an overextended buildout.
He described the political driver behind the spending as “AI nationalism.” Governments, in his view, are treating data centers, semiconductors and energy capacity as strategic assets, which turns AI infrastructure into a national-security funding priority rather than a normal private-sector investment cycle.
Hayes warned that this public and private financing surge is detached from organic economic output. The result, he argued, is a liquidity expansion created by credit issuance, not by productivity gains that have already been proven in the real economy.
Bitcoin Becomes the Policy-Rescue Trade
The risk, according to Hayes, is that AI-driven automation could trigger job losses large enough to pressure consumers, mortgages and bank balance sheets. He cited estimates from banks and consultancies to argue that automation-sensitive sectors could become a channel for credit destruction if income disruption spreads.
Hayes also pointed to warnings that central banks may have limited tools to offset AI-driven unemployment directly. His expected sequence is that defaults and bank distress would force policymakers back into large-scale liquidity injections, restarting aggressive monetary accommodation.
Under that scenario, Hayes sees Bitcoin and other risk assets benefiting from reactive central-bank support. His projections include Bitcoin at $126,000 by December 2026 if near-term Fed liquidity, AI spending and geopolitical expenditure continue to build.
He also outlined more aggressive scenarios: $500,000 to $750,000 by the end of 2026 if a broader crisis forces faster and larger liquidity responses, and $1 million by 2028 if money-supply expansion persists over several years.
Hayes’s thesis remains contrarian. Critics argue that major cloud providers and profitable technology firms still fund large portions of AI capital expenditure internally, which challenges the idea that the entire AI cycle already resembles a classic credit bubble.
Still, his framework gives traders a clear macro risk map. If the sequence he describes plays out, markets could see higher volatility across credit spreads, bank funding costs and crypto derivatives as loan losses collide with policy intervention.
The key variables are the timing of defaults, the scale of central-bank liquidity and how quickly investors rotate into scarce assets. Hayes’s central message is that Bitcoin’s next major move may be driven less by crypto-native catalysts than by the financing stress behind the AI boom.
