The company has disputed the Pentagon’s claims that it can somehow still control the Claude AI deployed in military networks.
AI developer Anthropic insists it has no backdoor or “kill the switch” for its Claude AI once deployed in classified Pentagon military networks, according to a new court filing.
The US military and the technology company found themselves in a policy dispute earlier this year, with the Pentagon insisting on using the system for “all legitimate military purposes,” while the company stood by its AI defenses regarding mass surveillance and the use of fully autonomous weapons.
The Pentagon eventually ended its partnership with Anthropic, appointing the technology company a “Supply chain risk,” a rare label usually reserved for entities associated with Washington’s foreign rivals. The designation prevents the company from working not only with the US government directly but also any other contractors from using its products as well.
In a new filing with a federal appeals court in Washington, DC, Anthropic contested the US federal government’s claim that the company still maintained some level of control over Claude AI once it was deployed to classified systems and properly committed. “Operational veto.” The company said “no back door or remote kill switch,” his time “Employees cannot log into the department’s system to modify or disable the running model.”
The AI system offered to the Pentagon comes as a “static” example, the company argued. Once it is sent, then “it does not degrade or change on its own, and Anthropic cannot push undisclosed or unapproved changes to the model after the department deploys it.”
Anthropic was officially designated a “Supply risk to national security” February 27, while US President Donald Trump accused it of being run by “quitting nut jobs.” The company challenged the label in court, with the case yielding mixed results.
Earlier this month, a DC court denied Anthropic’s request to stop the designation of supply chain risk. In a parallel case in California, however, the court sided with the company, temporarily blocking management’s decision. With the split decision, Anthropic remains barred from working with the Pentagon but can still continue its partnerships with other agencies while the legal battle continues.
You can share this story on social networks:
Follow RT on
