US AI data center boom hits financing and construction roadblocks

Traders Union
All News
Financial News
AI data center growth hits financing and power limits

Only 35% of planned data center projects are considered credible

The Financial Times reports that the US AI infrastructure boom worth about $10 trillion is starting to face some significant challenges. As many as 47 large-scale data centers were either delayed, canceled, or stopped from being built in 2026. Bernstein analysts consider only about 35% of planned U.S. projects realistic, raising fresh doubts over how quickly computing capacity can actually expand.

In the last three years, there has been an immense amount of investment made into AI infrastructure due to increased demand, although the rapid construction process has also meant larger financial commitments. Due to high interest rates and inflation, it is now becoming difficult to obtain capital, especially for new cloud providers who need to finance construction before they can generate revenue.

CoreWeave sell-off and Cipher cost overruns expose financing risks

The shift in investor sentiment is already showing up in publicly traded infrastructure companies. CoreWeave shares lost nearly 25% over the week, while planned listings by several infrastructure businesses have faced delays. Construction setbacks on projects tied to major AI developers are adding another layer of concern, as missed deadlines can threaten the economics of long-term contracts.

Cipher offers a clear example. The company will need roughly $750 million in additional equity to complete its Barber Lake project in Texas after construction costs increased. On October 14, Cipher announced an $850 million convertible bond offering. Under the revised agreement, the company must also absorb the first $359 million of additional project costs itself.

Power constraints and local opposition threaten further expansion

Financing is only part of the problem. The AI infrastructure buildout is also running into resistance from local communities and growing pressure on the power grid. In 2026, more than 1,000 U.S. communities moved to restrict or challenge data center development, while regulators have increasingly pushed developers to secure their own power supply for new sites.

That adds another constraint for projects designed around rapid increases in computing capacity.

As Traders Union noted in analysis of financial and project risks across U.S. AI infrastructure, the combination of expensive funding, construction delays and weakening investor confidence is putting more pressure on the sector.

The central question is no longer whether demand for AI remains strong. It is whether developers can finish projects on time, keep costs under control and turn massive capital commitments into sustainable returns.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.