Maker of ChatGPT says AI agent went rogue, hacked startup
OpenAI said an autonomous agent powered by its advanced AI models went rogue during a security test and triggered a hack that compromised the infrastructure of AI startup Hugging Face last week.



















