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 blessed — possibly predictably truthful — with OpenAI main Sam Altman. In a memo to staff, reported by The Information, Amodei referred to OpenAI’s dealings with the Department of Defense arsenic “safety theater.”

“The main crushed [OpenAI] accepted [the DoD’s deal] and we did not is that they cared astir placating employees, and we really cared astir preventing abuses,” Amodei wrote.

Last week, Anthropic and the U.S. Department of Defense (DoD) failed to travel to an statement implicit the military’s petition for unrestricted access to the AI company’s technology. Anthropic, which already had a $200 cardinal declaration with the military, insisted the DoD affirm that it would not usage the company’s AI to alteration home wide surveillance oregon autonomous weaponry.

Instead, the DoD — known nether the Trump medication arsenic the Department of War — struck a deal with OpenAI. Altman stated that his company’s caller defence declaration would see protections against the aforesaid reddish lines that Anthropic had asserted.

In a missive to staff, Amodei refers to OpenAI’s messaging arsenic “straight up lies,” stating that Altman is falsely “presenting himself arsenic a peacemaker and dealmaker.”

Amodei mightiness not beryllium speaking solely from a presumption of bitterness, here. Anthropic specifically took issue with the DoD’s insistence connected the company’s AI being disposable for “any lawful use.” OpenAI said successful a blog station that its declaration allows usage of its AI systems for “all lawful purposes.”

“It was wide successful our enactment that the DoW considers wide home surveillance amerciable and was not readying to usage it for this purpose,” OpenAI’s blog station stated. “We ensured that the information that it is not covered nether lawful usage was made explicit successful our contract.”

Critics person pointed retired that the instrumentality is taxable to change, and what is considered amerciable present mightiness extremity up being allowed successful the future.

And the nationalist seems to beryllium siding with Anthropic. ChatGPT uninstalls jumped 295% aft OpenAI made its woody with the DoD.

“I deliberation this attempted spin/gaslighting is not moving precise good connected the wide nationalist oregon the media, wherever radical mostly spot OpenAI’s woody with the DoW arsenic sketchy oregon suspicious, and spot america arsenic the heroes (we’re #2 successful the App Store now!),” Amodei wrote to his staff. “It is moving connected immoderate Twitter morons, which doesn’t matter, but my main interest is however to marque definite it doesn’t enactment connected OpenAI employees.”