Anthropic will likely continue supplying Claude to the NSA despite Pentagon labeling it a supply chain risk, with White House approval

The US government classified Anthropic as a 'supply chain risk' because the company refused to release its technology for 'any lawful use.' A contract now being finalized includes a clause barring the model from processing Americans' data, and the controversial 'any lawful use' language is reportedly not part of the deal.

Anthropic's new 'Mythos' model runs on older chips, making it the only short-term fix for the NSA's classified networks because intelligence agencies lack enough of Nvidia's latest Grace Blackwell chips to run current models from OpenAI and others.

White House Chief of Staff Susie Wiles personally approved the arrangement. The White House intends to use the contract as a template for future agreements with other companies.

Separately, the White House approved $9 billion for new AI chips to address the shortage long-term, pending Congressional approval.