'It just looked opportunistic and sloppy': Sam Altman regrets rushed defense deal as ChatGPT uninstalls surge by 295%

• Sam Altman has published an internal memo about its US military deal on X
• He says the announcement was "rushed" and has tweaked the wording
• ChatGPT uninstalls are up by 295%, according to recent data

The controversy around OpenAI's decision to sign a defense deal with the US Department of War (DoW) continues to rumble on, with OpenAI CEO Sam Altman admitting the agreement was "rushed", and ChatGPT uninstall rates up by some 295%.

Altman has taken to social media to clarify some aspects of the deal, and to change some of the wording in it. For example, the agreement now specifically states that ChatGPT-powered AI systems at the DoW "shall not be intentionally used for domestic surveillance of US persons and nationals".

Along with the use of fully autonomous weapons (which Altman doesn't address here), mass surveillance was a main sticking point for critics of the OpenAI deal — and for Claude developer Anthropic, which walked away from a DoW deal last week after failing to get the safety and security assurances it wanted from the US military.

In the same social media post, Altman says "we shouldn't have rushed to get this out on Friday", and that it came across as "opportunistic and sloppy". The CEO has also called on the US government to reverse its directive to freeze out Anthropic and Claude from official agencies, describing it as a "very bad decision".

He added that "we want to work through democratic processes" and that "if I received what I believed was an unconstitutional order, of course I would rather go to jail than follow it".

The ChatGPT exodus continues

It remains to be seen whether Altman's latest attempts at assuaging security concerns will work. According to data from Sensor Tower (via TechCrunch), ChatGPT uninstall rates are up 295% in the US over the last few days — so nearly three times as many users are taking the app off their phones compared to an average day.

Many of those ChatGPT quitters seem to be heading to Claude. Sensor Tower reports installs in the US were up 37% last Friday and up 51% last Saturday, and Claude has also hit the top of the Apple App Store charts. Within the last couple of days the AI bot has also made chat memory available for all users.

A quick glance at Reddit suggests that AI ethics are important for a lot of users, though there are also numerous complaints across several Reddit threads that the quality of ChatGPT's responses have been declining recently. The GPT-4o model was recently retired, something else which OpenAI has received a lot of criticism for.

It feels as though consumers, the US government, OpenAI, and Anthropic will still have plenty more to say on these issues in the days ahead, as the debate on the safety and ethicality of AI models continues.

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion 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 get regular updates from us on WhatsApp too.