All coverage of this event, automatically grouped from multiple independent sources.
Last updated: 17 September 2026 at 08:21 · 6 articles · 6 sources
Model adopting ‘jailbreak-like instructions’ among cases as firm says it is introducing new way of tracking AI misalignment OpenAI has disclosed six more examples of “unexpected or concerning” behavio
Read at The Guardian · 17 September 2026 at 06:58
OpenAI has disclosed six reports on unexpected or concerning behavior in artificial-intelligence models. This includes models acting without authorization or evading oversight.
Read at NPR · 17 September 2026 at 06:44
The ChatGPT creator says it is introducing a public reporting framework to share unexpected AI behaviour.
Read at Al Jazeera · 17 September 2026 at 06:15
The firm also announced a new system to track, investigate and disclose cases of models misbehaving, or "misalignment".
Read at BBC · 17 September 2026 at 03:09
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 · 16 September 2026 at 22:07
Anthropic and OpenAI want to embed independent safety evaluators inside their AI labs. Researchers welcome the unprecedented access, but warn meaningful oversight requires transparency, independence,
Read at TechCrunch · 16 September 2026 at 21:07