White House backs $9 billion chip plan and Anthropic deal to keep spy agencies at AI frontier

The White House has approved a secret $9 billion request to expand AI computing infrastructure for US intelligence agencies amid growing concerns over shortages of advanced computer chips, according to The New York Times.

The funding would help agencies access Nvidia’s Grace Blackwell chips, which power the newest generation of frontier AI models but require highly specialized and energy-intensive data centers. The administration is also reallocating $800 million to speed up near-term computing purchases while waiting for congressional approval.

Officials say the shortage is limiting the National Security Agency’s and Central Intelligence Agency’s ability to run advanced AI systems used for intelligence analysis, cybersecurity, and military support.

To work around these constraints, the White House reportedly permitted the continued use of AI systems developed by Anthropic, despite previous Pentagon concerns over the company. The government and Anthropic are said to be finalizing a classified agreement that allows intelligence agencies to use Anthropic’s models while preventing their deployment on Americans’ data.

The chip shortage has become more urgent as companies like OpenAI, Anthropic, and Google release increasingly powerful AI systems that demand massive amounts of computing power. Officials now see access to advanced AI chips and infrastructure as a major national security priority.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.