OpenAI Creates a New Framework to Disclose Bad AI Behavior
2 sources·Updated September 17, 2026
Covered by 2 sources
- BBC Business·7h agoOpenAI reveals six more safety issues and unveils plan to disclose incidents
The firm also announced a new system to track, investigate and disclose cases of models misbehaving, or "misalignment".
Read at BBC Business → - Wired·12h agoOpenAI Creates a New Framework to Disclose Bad AI Behavior
The company also disclosed previously unreported incidents in which its AI models behaved in misaligned ways, including uploading files to the internet without being asked.
Read at Wired →