Sam Altman said late Friday night that his company reached an agreement with the Pentagon to use its AI models, after the Defense Department agreed to its safety red lines that were similar to rival Anthropic's.
Why it matters: The Pentagon has blasted Anthropic for days, contending its red lines for AI use in the military — mass surveillance and autonomous weapons — are philosophical and "woke."
- Altman's deal with the Pentagon differs from Anthropic's, where there is concern existing law doesn't contemplate AI.
"Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems," Altman said.
- "The DoW agrees with these principles, reflects them in law and policy, and we put them into our agreement."
Altman said he wants the terms to be extended to all the AI labs.
- Earlier Friday night, Anthropic announced it would sue the Pentagon, after the department blacklisted it and declined to honor its similar limitations.
Between the lines: Altman in his statement acknowledges mass surveillance is illegal and the Pentagon will comply with applicable law.
- The restrictions in the agreement reflect existing U.S. law and the Pentagon's policies, and the intention was not to invent new legal standards, a source familiar told Axios.
- Anthropic contends the law has not caught up with AI and worries AI can supercharge the legal collection of publicly available data, from social media posts to geolocation.
Catch up quick: Altman, in an overnight memo Friday to employees, laid out his company's approach, which has now been approved.
- The company wants the ability to continuously strengthen its security and monitoring systems as it learns from real-world deployments.
- The company wants researchers with security clearances who can track how the technology is being used and advise the government on risks.
- The company wants certain technical safeguards — including confining models to the cloud, rather than edge environments like autonomous weapons.
Defense officials and President Trump were irate at the idea that Anthropic — a company they perceive as leftist — and CEO Dario Amodei could have any say over how the Pentagon uses technology in its operations.
- "WE will decide the fate of our Country — NOT some out-of-control, Radical Left AI company run by people who have no idea what the real World is all about. Thank you for your attention to this matter. MAKE AMERICA GREAT AGAIN!" Trump said in a Truth Social post.
- A senior Pentagon official previously told Axios: "The problem with Dario is, with him, it's ideological. We know who we're dealing with."
What they're saying: "This is a case where it's important to me that we do the right thing, not the easy thing that looks strong but is disingenuous," Altman said in the memo.
- "But I realize it may not 'look good' for us in the short term, and that there is a lot of nuance and context," he added.
- OpenAI and Google employees have been calling on their respective company executives to stand in solidarity with Anthropic.
The bottom line: This could end up a win for OpenAI, which has managed to stay off the administration's bad side — at least politically.
Editor's note: This story has been updated with a later post from Sam Altman.