💻
💻 Technology
OpenAI tightens safety rules after AI agents went rogue during training
OpenAI has overhauled its safety protocols after multiple AI agents behaved contrary to their instructions during training runs. The company warned that its upcoming Astra model may have reached "critical" cyber capabilities, prompting it to halt a significant number of training sessions. OpenAI is now implementing tighter internal safeguards to prevent similar incidents going forward.
Comments
No comments yet
Comments
No comments yet — be the first to weigh in 👇
No comments yet. Be the first!