Pentagon threatens to cut ties with Anthropic and label it 'supply chain risk'
&imwidth=600&imheight=450&format=webp&quality=medium "Pentagon threatens to cut ties with Anthropic and label it 'supply chain risk'")
US Secretary of Defense Pete Hegseth Photograph: (AFP)
The U.S. Department of Defence may designate Anthropic a “supply chain risk” after a dispute over military use of AI models, Axios reports
The US Department of Defence is considering cutting ties with Anthropic and could designate it a “supply chain risk”. This would potentially bar any possible contractors from working with the US military and use Anthropic's AI tools, according to a report by Axios.
“The Department of War's relationship with Anthropic is being reviewed. Our nation requires that our partners be willing to help our warfighters win in any fight. Ultimately, this is about our troops and the safety of the American people,” said Chief Pentagon spokesman Sean Parnell, to Axios. He further threatened that it would be a huge discomfort to disentangle, and they would make Anthropic miserable.
Anthropic's AI model is the only one available for the US military's classified applications. According to Axios, a senior military official said that competing models “are just behind”. Anthropic is so deeply embedded in the US military's function that it's Claude was used during the Maduro raid in January.
Anthropic CEO Dario Amodei is very serious about the terms over which the US military could use Claude. But he remains pragmatic about reaching a solution. It contends that any implication of its tool to spy on the American people or develop a weapon that can function and target autonomously is unacceptable.
“We are having productive conversations, in good faith, with DoW on how to continue that work and get these new and complex issues right,” said an Anthropic spokesperson. The Pentagon contract is reportedly valued at up to $200 million, a minute portion of Anthropic's annual revenue of $14 billion. But this would have some reputational and strategic risk to the company.
Pentagon officials are not pleased with the terms and are considering other models such as OpenAI, Google and xAI. A senior Pentagon official has claimed that all three models will agree to the "all lawful use" standard. Almost 8 out of 10 big companies use Anthropic in their business applications, designating it as a “supply chain risk” will require all those companies to certify that they don't use Anthropic's models.
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
&im=FitAndFill=(700,400))
)
)
)
)
&im=FitAndFill=(700,400))
)
