Arthur Hayes, chief investment officer at Maelstrom and co-founder of BitMEX, has restated his long-term projection that Bitcoin could reach $1 million by 2030. The forecast centers on a sequence in which slowing capital spending in the artificial intelligence sector exposes underlying debt vulnerabilities, prompting governments and central banks to expand liquidity in ways that favor scarce assets.

Hayes has identified late 2027 or early 2028 as the period when the strongest phase of any advance may occur. He argues that the rapid buildout of data centers, advanced chips and related computing systems has relied heavily on borrowed funds rather than solely on near-term earnings from technology firms.

AI Infrastructure Boom Framed as Credit Risk

In Hayes’s assessment, the AI investment cycle resembles a credit-driven episode more than a pure technology valuation excess. He has described the financing of land, buildings, power connections, cooling systems and processors as akin to property development funded by debt. Hardware useful life is short, often around two years for frontier applications, while many loan repayment schedules extend longer, creating a potential mismatch as equipment ages and revenue assumptions face pressure.

Under this view, major technology companies could remain profitable even as weaker projects and their lenders encounter difficulties meeting interest, lease and principal obligations. Losses could then transmit through banks, private-credit funds, insurers and infrastructure investors. Hayes expects authorities to respond with measures that increase the money supply rather than allow widespread defaults to cascade through the financial system.

Projected Timeline and Spending Slowdown

Hayes anticipates that growth in announced AI capital expenditures will begin to decelerate in the second half of 2027, becoming more evident in 2028. At that stage, projects based on optimistic demand forecasts may become easier to identify, revealing strains in the credit structure supporting the buildout.

He has previously outlined scenarios in which Bitcoin could trade in a range of $60,000 to $70,000, with possible further downside toward $50,000, before any sustained advance toward the longer-term target. The $1 million level remains conditional on the full sequence of investment slowdown, credit losses and policy response materializing.

Scale of Financing and Institutional Attention

Market data point to the volume of debt involved. AI-related borrowing in the U.S. leveraged-finance market is projected to reach $88 billion in 2026, rising from about $20 billion in early 2025. The International Monetary Fund has referenced Morgan Stanley estimates placing data-center capital expenditures through 2028 at $2.9 trillion, an amount that exceeds the expected cash flows of large hyperscale operators and points to continued reliance on private loans, corporate debt and securitization.

Research from Apollo has estimated that the AI ecosystem could support more than $2 trillion of additional investment-grade debt. Public markets may absorb less than $1 trillion through 2030 due to concentration and rating constraints, leaving more than $1 trillion potentially dependent on private placements, infrastructure lending, equipment financing and project-specific structures. AI-related issuance has already accounted for a substantial share of longer-duration investment-grade corporate bond supply.

U.S. insurance regulators have also moved to improve transparency. The National Association of Insurance Commissioners has adopted changes effective at year-end 2026 that require enhanced reporting of private credit holdings, reflecting ongoing monitoring of valuations, lending standards and sector exposures.

Policy Response Paths and Bitcoin Implications

Hayes has outlined possible official reactions that include authorities acting as a buyer of last resort for computing capacity or providing support to institutions holding AI-linked debt. In either case, the resulting expansion of liquidity would, in his framework, increase demand for assets with fixed supply such as Bitcoin.

Bitcoin was trading near $83,000 to $85,000 in early October 2026 as these comments circulated. Hayes’s projection does not treat ongoing adoption alone as sufficient; it depends on the interplay between AI sector credit dynamics and the monetary response that may follow.

The outlook remains one scenario among several possible paths for both the technology investment cycle and digital asset markets. Hayes has noted that the precise trigger or timing of any credit event cannot be predicted with certainty.