Stonepeak CEO flags capital markets as biggest risk to AI infrastructure investment boom

The rapid expansion of AI infrastructure continues to attract strong backing from banks, private equity firms and capital markets, but Stonepeak Infrastructure Partners CEO Michael Dorrell has warned that sustained access to financing remains the sector’s greatest long-term vulnerability, according to a report by Bloomberg.

Speaking in an interview with Bloomberg Television, Dorrell said there is currently little evidence that investors are pulling back from funding AI-related infrastructure projects, despite the unprecedented levels of capital flowing into the sector.

He cautioned, however, that the pace of investment is unlikely to continue indefinitely, arguing that many businesses seeking funding have yet to establish sustainable cash flows.

According to Dorrell, the AI infrastructure boom remains heavily dependent on open capital markets. If financing conditions were to tighten significantly, companies reliant on external funding could face considerable challenges in completing projects or executing their growth strategies.

His comments come as investment in data centres accelerates globally. Consultancy McKinsey estimates that cumulative spending on data centre infrastructure could reach $5.2tn by 2030, driven by demand for AI computing capacity.

Dorrell argued that today’s investment cycle differs fundamentally from the technology bubble of the late 1990s, noting that many modern data centre developments are supported by long-term customer contracts rather than speculative demand. In his view, this provides a stronger commercial foundation than the fibre network investments that characterised the dot-com era.

He also highlighted increasing political and regulatory scrutiny of new data centre developments in the US. Restrictions on hyperscale data centre construction in states such as New York could become more significant if adopted more widely, he said, potentially creating tensions between local planning concerns and the US government’s strategic objective of maintaining leadership in artificial intelligence relative to China.