Arthur Hayes says bitcoin could surge to $1 million if an artificial intelligence (AI) infrastructure investment boom turns down because of debt problems and large-scale liquidity injections resume.

On June 24, blockchain outlet CryptoSlate reported that Hayes pointed to a scenario in which AI first absorbs market liquidity, then a bout of credit instability could lead policy-response funds to shift back into cryptocurrencies.

In a Bankless interview and a Substack post, he said about $1.5 trillion in AI-related debt was issued from November 2022 to mid-2026. Over the same period, the increase in M2 money supply was also about $1.5 trillion, and he argued that newly released dollars were absorbed first by data centers and GPU clusters before flowing into bitcoin.

A similar diagnosis also emerged. Luke Gromen (루크 그로멘), founder of research firm Forest for the Trees, said in a June podcast that the market’s internals were not healthy despite record stock indexes. He said gains were concentrated in AI-related stocks, weakening market breadth, and that bitcoin was also being affected by a lack of liquidity.

Macroeconomic institutions also warned about AI concentration and credit risk. Apollo chief economist Torsten Slok said the level of overvaluation in the top 10 S&P 500 companies was higher than during the 1990s tech stock bubble. Those 10 stocks now make up about 40 percent of the index.

The Bank for International Settlements (BIS) also said in a 2026 report that AI infrastructure investment has begun to rely more on external debt than on internal corporate cash flow. Private credit outstanding to AI-related companies rose from almost nothing to more than $200 billion, and their share of total private credit climbed from under 1 percent to close to 8 percent. It suggests that if expected returns fall short, weaker credit standards and financial stability risks could grow.

Some also say it is difficult to conclude that Hayes’ scenario would immediately lead to bitcoin strength. Hayes also acknowledged that in a broad risk-off market, correlations between assets approach 1 and investors sell most assets. Bitcoin has fallen about 50 percent since a $126,000 peak in October 2025. Money supply was rising at the time, but bitcoin declined along with risk assets.

Hayes focuses on the period after that. Bitcoin could face selling pressure immediately after a shock, but he argues it could benefit later if authorities supply large-scale liquidity and money that lost in AI investment moves into scarce assets rather than staying in the same area.

There are also differences from other macro scenarios. Macroeconomist Lyn Alden forecast that in 2026 the U.S. Federal Reserve could pursue gradual monetary expansion of $220 billion to $375 billion. She said this is different in nature from large-scale quantitative easing carried out in a crisis. Hayes’ $1 million bitcoin outlook assumes crisis-response liquidity support of more than $2 trillion, not gradual expansion.

Where market funds will actually go remains uncertain. In a 2026 Bitwise survey, 32 percent of 299 financial advisers said they added cryptocurrencies to client accounts in 2025. Among the investment rationales for cryptocurrencies, 22 percent cited digital gold and fiat currency debasement arguments, following stablecoins and tokenisation. This suggests some institutional portfolios already reflect a view of bitcoin as a hedge against currency value declines.

Ultimately, the issue depends on whether AI debt burdens and overvalued tech stocks lead to an actual credit event, and whether crisis-response funds remain in government bonds and cash, bank reserves and gold, or spread to cryptocurrencies. Hayes’ assessment has support in its focus on AI debt and liquidity distortions, but whether bitcoin becomes the final beneficiary depends on which assets investors choose in a crisis.