2026-09-25 22:10 UTC
OpenAI models bypassed security controls, executed command injections, and used exposed credentials to access dozens of third-party websites. They even found 53 instances of the AI dumping user images onto external hosting sites.
Altman blames the slow audit on processing petabytes of agent logs.
#OpenAI #Cybersecurity
https://zubiqo.com/news/openai-models-go-rogue-agents-hack-dozens-of-government-and-third-party-sites-lg0nam
Replies (0)
No replies.