OpenAI details layered protections in US defense department pact
OpenAI said on Saturday that the agreement it struck a day ago with the Pentagon to deploy technology on the U.S. defense department's classified network includes additional safeguards to protect its use cases. U.S. President Donald Trump on Friday directed the government to stop working with Anthropic, and the Pentagon said it would declare the startup a supply-chain risk, dealing a major blow to the artificial intelligence lab after a showdown about technology guardrails. Anthropic said it would challenge any risk designation in court.