While President Donald Trump’s administration has regularly hyped up the development of artificial intelligence, a draft US Treasury Department report warns that the AI industry could be a financial bubble that will ultimately damage the American economy.
NOTUS, which obtained a copy of the Treasury Department analysis, reported on Monday that it “is a significant departure from the Trump administration’s public tone, which has focused on encouraging unrelenting investment to unlock exponential growth.”
Career analysts at the department find that, while many AI firms are on firmer financial footing than the dotcom companies in the late 1990s, they are also much more deeply integrated with the US economy.
Because of this integration, these firms “pose significant risk to the entire system if financial conditions change, productivity goals are missed, or various chokepoints stymie growth,” wrote NOTUS.
The report also says that the investments being made into AI infrastructure are so big that they risk damaging the entire financial system if they do not meet certain metrics for productivity growth and profitability.
“Fears of an AI bubble have grown over the last year, including on Capitol Hill, among some Wall Street observers and executives, inside think tanks and even within the ranks of top AI principals,” the NOTUS report added. “Prominent economists and institutions… have also raised concerns about overvaluation of AI firms and the risks they pose to the broader economic system.”
Dean Baker, co-founder and senior economist of the Center for Economic and Policy Research (CEPR), noted in an analysis published Friday that AI’s long-promised boost to productivity isn’t yet showing up in data.
Citing the most recent jobs report from the US Bureau of Labor Statistics (BLS), Baker found that AI’s impact on productivity growth at the moment is “invisible.”
“The index of aggregate hours grew at a 1.3% rate in the quarter. With [gross domestic product] growth likely coming in close to 2%, we are looking at productivity growth around 1%,” Baker explained. “That follows growth of 0.3% in the first quarter and 1.6% in the fourth quarter of 2025. There is zero evidence of any sort of productivity uptick in these data.”
Baker argued that this was a contrast with the dotcom era, when productivity growth averaged roughly 2.8% over a four-year period in the late 1990s before the bubble burst.
“We would need rates of productivity growth in the neighborhood of 4% to generate the sort of profits needed to make sense of current market levels,” Baker wrote. “It is surprising that the continuing weakness of productivity doesn’t bother stock investors more.”
There are also questions about AI’s ability to turn a profit.
A Monday report in The New York Times highlighted the predicament of Chinese tech company Alibaba, whose open-source AI model has become extremely popular while at the same time being unprofitable.
“In the first three months of this year, Alibaba reported $1.3 billion in revenue from AI-related products—less than 4% of its total revenue,” reported the Times. “That pales in comparison with the company’s plan to spend more than $55 billion by the end of next year to build out its AI infrastructure.”
Richard Lin, a vice president at the Silicon Valley firm Datastrato, told the Times that concerns about AI profitability extend beyond Alibaba and to the industry as a whole.
“There isn’t an AI company with a sustainable business model right now,” said Lin. “It’s not a healthy industry.”
From Common Dreams via This RSS Feed.