
Arthur Hayes, the former chief executive of crypto derivatives exchange BitMEX and a prominent macro commentator, has renewed his argument that the eventual deflation of the artificial intelligence investment boom could become a powerful bullish catalyst for Bitcoin. In his view, the huge sums flowing into AI data centers, chips, and related infrastructure are creating conditions for a classic capital misallocation cycle. When that cycle turns, the policy response could end up benefiting scarce digital assets rather than the companies that led the boom.
Speaking at the Gamma Prime Investing Conference in Singapore, Hayes said the current AI investment mania is ripe for oversupply. He described the spending as a multitrillion-dollar case of capital misallocation. The irony, he suggested, is that the end result may still be cheap and accessible massive compute power. But the journey to that endpoint could be painful for financial markets, especially for investors who have extrapolated exponential demand growth far into the future.
Key Facts
- Arthur Hayes is the former CEO and co-founder of BitMEX.
- He believes the AI investment boom could crash and become bullish for Bitcoin.
- Hayes calls the AI data center spending a multitrillion-dollar capital misallocation.
- He sees a possible credit crisis if AI infrastructure investments fail to generate sufficient returns.
- Governments and central banks would likely intervene with liquidity injections.
- Bitcoin and other cryptocurrencies could be major winners of that excess cash.
- The immediate catalyst may not be an AI crash but a deleveraging event that initially hurts risk assets.
- Hayes points to 2027–2028 as a potential reckoning when companies must justify compute costs.
- He acknowledges AI demand could grow into the infrastructure and justify the spending.
- Hayes compared the AI boom to credit excesses before the 2008 financial crisis.
- He focuses on debt financing for data centers and expensive compute hardware.
- Hayes is not shorting the AI boom, but he is confident there is significant overbuilding.
The core of Hayes's thesis is liquidity. If the AI infrastructure boom leads to a major credit crisis, he expects governments and central banks to step in. They would inject additional liquidity into the financial system to prevent a broader meltdown. In that scenario, Bitcoin and other cryptocurrencies are likely winners of all the excess cash. This is a familiar framework for Hayes, who has long argued that Bitcoin functions as a liquidity sponge. When money becomes cheap and abundant, scarce assets with fixed or predictable supply tend to attract capital.
Hayes is careful to distinguish between the immediate market reaction and the eventual monetary response. The first phase of an AI unwind would likely be negative for risk assets. A major deleveraging move would force investors to sell what they can, including stocks and cryptocurrencies. But after that initial shock, the policy response could be bullish for stocks and other risk assets. Bitcoin, which trades like a high-beta risk asset in the short term but behaves like a monetary hedge over longer horizons, could benefit disproportionately once liquidity returns.
Why the Timing Matters
Hayes points to 2027–2028 as a possible period of reckoning. By then, companies that promised massive compute power may need to generate enough money to justify their costs. If revenue fails to materialize, the debt used to fund data centers, advanced chips, and energy infrastructure could become a problem. The AI boom has been financed not only through equity but also through debt, private credit, and complex structured deals. That leverage is what makes Hayes nervous.
He admits there is another possibility. Over the next year, demand for AI could become strong enough to justify the large infrastructure investment. The usefulness of AI may allow companies to grow into these massive compute costs. Hayes acknowledges this as a potential scenario. If AI tools continue to improve and spread across the economy, the revenue base could expand rapidly enough to support the buildout. In that case, the boom would not end in a bust but in a productive, if expensive, transformation.
Still, Hayes is not convinced that the optimistic path is the most likely one. He sees signs of overbuilding across the AI supply chain. Data center construction has surged. Power constraints and land costs have become pressing issues. Chipmakers and cloud providers are racing to add capacity. The risk is that all this capacity arrives at once, just as demand growth slows. That would create a classic oversupply problem, pressuring prices and returns.
A Repeat of 2008, Not the Dot-Com Bubble
This is not the first time Hayes has envisioned such a scenario. In August, he compared the AI infrastructure boom to the credit excesses before the 2008 financial crisis. Rather than viewing the current cycle as another dot-com tech bubble, he said people should pay attention to the debt financing for data centers and expensive compute hardware. The dot-com comparison focuses on equity valuations and failed business models. The 2008 comparison focuses on leverage, credit markets, and systemic risk. Hayes believes the latter is more relevant.
That distinction matters for Bitcoin. During the dot-com crash, the internet survived and thrived, but many companies disappeared. There was no broad central bank liquidity response on the scale of 2008. After the global financial crisis, by contrast, central banks engaged in unprecedented monetary easing. Bitcoin was born in the wake of that crisis, and its early adoption was driven in part by distrust of the financial system. If the AI boom ends in a credit event, the policy response could echo 2008, with rate cuts, asset purchases, and fiscal stimulus. That would strengthen the case for decentralized, scarce assets.
Hayes also has a track record of making bold macro calls. As the co-founder of BitMEX, he built one of the first major crypto derivatives platforms. He became known for his outspoken views on monetary policy, crypto cycles, and risk-taking. His career has been marked by both controversy and influence. He has repeatedly argued that Bitcoin's price is driven by global liquidity conditions, not just adoption or halving events. That framework leads him to focus on credit markets, central bank balance sheets, and the cost of capital.
Bitcoin as a Liquidity Sponge
In Hayes's model, Bitcoin is particularly sensitive to shifts in liquidity. When central banks inject money into the system, some of that capital flows into assets with limited supply. Bitcoin's supply schedule is fixed, with issuance halving roughly every four years. The next halving is expected in 2028, which aligns roughly with the 2027–2028 reckoning Hayes describes. If a credit crisis forces central banks to ease around the same time, Bitcoin could face a powerful combination of supply scarcity and demand for alternative stores of value.
This does not mean the path would be smooth. Bitcoin has repeatedly suffered sharp drawdowns during liquidity crises. In March 2020, it crashed alongside stocks before rallying massively after central banks intervened. A similar pattern could play out if AI debt problems trigger a broader market selloff. Leveraged crypto traders could be liquidated. Miners could face pressure. But for long-term holders, the monetary response might ultimately validate the thesis that Bitcoin benefits from fiat debasement and financial instability.
Hayes is not alone in warning about AI overinvestment. Several investors and analysts have raised questions about the return on massive data center spending. The biggest technology companies are pouring hundreds of billions of dollars into AI infrastructure. Energy providers are being asked to supply extraordinary amounts of power. In some regions, data center demand is straining grids and raising electricity prices. These are real economic costs that must eventually be paid for by profitable AI services.
At the same time, AI has already produced genuine productivity gains. Companies are using machine learning to improve logistics, drug discovery, customer service, and software development. If those gains continue, the revenue could justify the investment. The bull case for AI infrastructure is that the technology is general-purpose and will diffuse across every industry. The bear case is that the spending has run far ahead of monetization, creating a bubble that will eventually pop.
Hayes falls somewhere in between. He acknowledges the transformative potential of AI, but he is confident that there is a lot of overbuilding. He is not shorting the AI boom, which suggests he respects its momentum. But he is positioning his thinking around what happens after the boom peaks. For Bitcoin, the answer may be unexpectedly positive. If AI excesses lead to a credit crisis, the same governments and central banks that enabled the boom may be forced to bail out the system. That liquidity would have to go somewhere, and Bitcoin is one of the few assets with a credible fixed supply.
The immediate catalyst for a Bitcoin boost is not necessarily an AI crash. Initially, a major deleveraging move could be negative for risk assets. But afterwards, the monetary response could be bullish for stocks and crypto alike. This could play out around 2027–2028, when the companies that promised all this massive compute power need to actually generate enough money to justify the costs. It could also be that over the next year, demand for AI becomes strong enough to justify this large infrastructure investment. Hayes admits that this could be a potential scenario. Basically, the usefulness of AI may allow companies to grow into these massive compute costs.
For now, Hayes remains confident that overbuilding is a defining feature of the AI cycle. He sees debt-financed data centers and expensive hardware as the pressure points. If those pressure points crack, the fallout could reshape financial markets. In that environment, Bitcoin's fixed supply and global liquidity could make it a surprising winner. The former BitMEX chief is not predicting an immediate crash, but he is preparing for a world where the AI boom's end becomes Bitcoin's beginning.
Source:U.Today News
