Anthropic co-founder and CEO Dario Amodei just isn’t comfortable — maybe predictably so — with OpenAI chief Sam Altman. In a memo to workers, reported by The Info, Amodei referred to OpenAI’s dealings with the Division of Protection as “security theater.”
“The primary purpose [OpenAI] accepted [the DoD’s deal] and we didn’t is that they cared about placating staff, and we really cared about stopping abuses,” Amodei wrote.
Final week, Anthropic and the U.S. Division of Protection (DoD) failed to return to an settlement over the navy’s request for unrestricted entry to the AI firm’s know-how. Anthropic, which already had a $200 million contract with the navy, insisted the DoD affirm that it will not use the corporate’s AI to allow home mass surveillance or autonomous weaponry.
As a substitute, the DoD — recognized beneath the Trump administration because the Division of Warfare — struck a deal with OpenAI. Altman said that his firm’s new protection contract would come with protections towards the identical crimson traces that Anthropic had asserted.
In a letter to workers, Amodei refers to OpenAI’s messaging as “straight up lies,” stating that Altman is falsely “presenting himself as a peacemaker and dealmaker.”
Amodei may not be talking solely from a place of bitterness, right here. Anthropic particularly took problem with the DoD’s insistence on the corporate’s AI being obtainable for “any lawful use.” OpenAI stated in a weblog put up that its contract permits use of its AI programs for “all lawful functions.”
“It was clear in our interplay that the DoW considers mass home surveillance unlawful and was not planning to make use of it for this goal,” OpenAI’s weblog put up said. “We ensured that the truth that it’s not lined beneath lawful use was made express in our contract.”
Techcrunch occasion
San Francisco, CA
|
October 13-15, 2026
Critics have identified that the regulation is topic to alter, and what’s thought-about unlawful now may find yourself being allowed sooner or later.
And the general public appears to be siding with Anthropic. ChatGPT uninstalls jumped 295% after OpenAI made its take care of the DoD.
“I believe this tried spin/gaslighting just isn’t working very properly on most of the people or the media, the place individuals principally see OpenAI’s take care of the DoW as sketchy or suspicious, and see us because the heroes (we’re #2 within the App Retailer now!),” Amodei wrote to his workers. “It’s engaged on some Twitter morons, which doesn’t matter, however my predominant fear is how to verify it doesn’t work on OpenAI staff.”
