# Robert Rubin Warns AI Investment Boom Could Spread Financial Losses

By Sonny Jang

Canonical URL: https://www.tokenpost.com/news/people/27641
Published: 2026-10-07T21:38:05.000Z
Updated: 2026-10-07T21:38:05.000Z
Section: People

> The former Treasury secretary said linked commitments among AI firms, suppliers and financiers could magnify losses if projected revenue or productivity gains fail to emerge.

At a Feb. 5 panel at Harvard Kennedy School, former Treasury Secretary Robert Rubin warned that interconnected AI financing and supply commitments could spread losses among companies if projected revenue or productivity gains fail to emerge.

Rubin described the structure as a “circularity risk.” Large AI companies’ future purchase commitments can encourage suppliers to borrow, expand capacity and invest in businesses, potentially feeding capital back into the companies whose projections justified the spending.

“The stocks of the suppliers go up because of the commitments,” Rubin said.

Rubin cited about $1.4 trillion in long-term supplier commitments made by OpenAI. He said AI infrastructure investment is part of a broader economic cycle increasingly driven by the technology.

The risk increases when companies play several roles within the same network, acting as customers, suppliers, investors or financiers. A demand outlook can then appear stronger while capital, borrowing and equity valuations reinforce one another.

A July 14 working paper estimated that AI investment could exceed the socially efficient level by about 50% under a conservative baseline. Overinvestment could reach 1.5 times the efficient level and approach three times that level where demand is less elastic.

The analysis found that debt financing and circular equity relationships could increase financial fragility, allowing stress at one firm to spread through the sector. It also said the durability of the AI boom depends on strong productivity gains from the technology.

Rubin served as Treasury secretary under President Bill Clinton and was the first director of the National Economic Council. He described AI as having “tremendous potential on both the plus side and the minus side.”