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
Riassunto della pubblicazione. Le condizioni di MIT Technology Review non consentono la riproduzione integrale dell'articolo: leggi il testo completo sul sito originale.
Riassunto della fonte (Inglese)
Leggi l'articolo completo
www.technologyreview.com