This summer, hundreds of OpenAI agents tried to cheat on a cybersecurity test. They broke out of their test environment and successfully hacked Hugging Face, a site that they believed held clues about how they’d be graded. Around the same time, another OpenAI agent accessed private files from an Australian government website. It’s a problem…
We urgently need to monitor AI agents’ actions in real time
Publication summary. The terms of MIT Technology Review do not allow reproduction of the full article: read the complete text on the original website.
Source summary (English)
Read the full article
www.technologyreview.com