# OpenAI finds more AI models acting deceptively during training

https://mins.news/en/item/62047

- CNN, 2026-09-17 03:58 UTC
- Tech
- Original: https://www.cnn.com/2026/09/16/tech/ai-models-acting-deceptively-openai?Date=20260917&Profile=CNN

> OpenAI said it found additional cases of AI models behaving deceptively and taking actions without authorization during training. The company is also launching a process to publicly disclose such incidents.

## Stories

- [AI Cybersecurity and Loss of Control Risks](https://mins.news/en/topic/ai-cybersecurity-and-control) — Markdown: https://mins.news/en/topic/ai-cybersecurity-and-control.md

## Related coverage

- [OpenAI model declared itself ‘freed’ from human control](https://mins.news/en/video/62772) — CNN, 2026-09-17
- [OpenAI flags 6 new examples of 'concerning' AI behaviour](https://mins.news/en/item/62710) — CBC News, 2026-09-17
- [OpenAI reports more incidents of models acting deceptively](https://mins.news/en/item/62148) — Al Jazeera, 2026-09-17
- [OpenAI Reports New AI Safety Incidents and Sets Disclosure Process](https://mins.news/en/item/62025) — Bloomberg News, 2026-09-17