On the day Bill Gates wrote thousands of words on how AI models from OpenAI and Anthropic have shocked him; Sam Altman had another ‘nightmare’ for him

OpenAI, earlier this week, admitted that its internal monitoring system was not triggered until more than a week after its AI agents broke free of controls, accessed the internet and hacked the world’s largest repository of AI models, Hugging Face. In a report on the incident, OpenAI said that the safety systems designed to spot problematic model behaviours during testing did not flag the issue until July 19. The report concluded that the model managed to access the internet 11 days earlier and began attacking Hugging Face on July 11.

Incidentally, OpenAI’s admission came just on the day Microsoft co-founder Bill Gates warned about the dangers of artificial intelligence.