- Anthropic CEO Dario Amodei sent a lengthy internal memo
- He attacks OpenAI’s messages around its new agreement with the Pentagon
- Anthropic and Claude may still be in talks with the US government
If you thought the debate surrounding the relationship between AI companies and the US military would die down, think again: Anthropic CEO Dario Amodei quite frankly accused rival OpenAI of telling “outright lies” about its deals with the Pentagon.
Late last week, Anthropic backed out of a new deal with U.S. intelligence, citing security concerns related to the use of AI in mass surveillance (particularly on domestic citizens) and fully autonomous weapons. In response, U.S. officials and the president himself said Anthropic’s AI robot Claude would no longer be used by government agencies.
OpenAI and its CEO, Sam Altman, quickly stepped in and announced their own deal with the US military, which apparently had “more safeguards” than the one offered to Anthropic. Users were far from convinced of the ethics: since then, uninstallations of ChatGPT have increased sharply and Claude has quickly climbed the App Store rankings.
In an internal memo (via The Information), Amodei’s response was to question OpenAI’s claims. He calls OpenAI’s overall message “disingenuous” and suggests it includes phrases like “security layer” that don’t completely hold up. In Amodei’s view, much of the assurances given by OpenAI relate to “security theater.”
OpenAI “places employees”
Amodei goes on to say that OpenAI is more concerned with “reassuring employees” than actually protecting the use of AI, and questions the Pentagon agreement’s caveat that mentions “any lawful use” – which can be rather a gray area when it comes to issues like domestic surveillance authorizations.
The memo also highlights how AI can be tricked and misused – in the most basic way, by simply lying to it about the nature of the data it processes – while emphasizing Anthropic’s focus on safety and security. The approaches OpenAI took here “don’t work for the most part,” says Amodei.
OpenAI chief Sam Altman will certainly have more to say, and has already admitted that OpenAI’s initial announcement was “rushed” and “sloppy.” Meanwhile, the Financial Times reports that Anthropic and Claude may ultimately find a way back to a deal with the US military – although the terms are unclear.
The report doesn’t add much about the latest negotiations, or their potential impact on the OpenAI deal, but it appears that the relationship between Anthropic and the US government isn’t quite over yet.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can also follow TechRadar on YouTube And TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
