Anthropic CEO Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies,’ report says

Anthropic co-founder and CEO Dario Amodei is not happy – perhaps predictably – with OpenAI chief Sam Altman. In a memo to staff said reported by The InformationAmodei called OpenAI’s dealings with the Defense Department “security theater.”

“The main reason [OpenAI] accepted [the DoD’s deal] and we didn’t, because they cared about satisfying employees and we cared about preventing abuse,” Amodei wrote.

Last week, Anthropic and the U.S. Department of Defense (DoD) failed to reach an agreement on the military’s request for unrestricted access to the AI ​​company’s technology. Anthropic, which already had a $200 million contract with the military, urged the DoD not to use the company’s AI to enable domestic mass surveillance or autonomous weapons.

Instead, the DoD – known as the Department of War under the Trump administration – struck a deal with OpenAI. Altman declared that his company’s new defense contract would include protection against the same red lines that Anthropic had claimed.

In a letter to employees, Amodei calls OpenAI’s messages “outright lies,” stating that Altman falsely “presents himself as a peacemaker and dealmaker.”

Amodei may not be speaking here exclusively from a position of bitterness. Anthropic specifically mentioned with the Department of Defense insisting that the company’s AI is available for “any lawful use.” OpenAI said in a blog post that its contract permits the use of its AI systems for “all lawful purposes.”

“During our interaction, it was clear that the DoW considers mass domestic surveillance to be illegal and had no intention of using it for this purpose,” OpenAI’s blog post said. “We made sure that the fact that it does not fall under lawful use was made explicit in our contract.”

Critics have pointed out that the law is subject to change, and that what is now considered illegal may be allowed in the future.

And the audience seems to be siding with Anthropic. ChatGPT takedowns increased 295% after OpenAI struck a deal with the DoD.

“I don’t think this attempt at spin/gaslighting plays very well with the general public or the media, where people tend to view OpenAI’s deal with the DoW as vague or suspect, and see us as the heroes (we’re now #2 in the App Store!),” Amodei wrote to his staff. “It works on some Twitter idiots, which doesn’t matter, but my biggest concern is how to make sure it doesn’t work on OpenAI workers.”