# OpenAI shelves new AI model amid safety concerns

https://mins.news/en/item/78013

- DW, 2026-09-29 03:50 UTC
- Tech
- Original: https://www.dw.com/en/openai-shelves-new-ai-model-amid-safety-concerns/a-79467841?maca=en-rss-en-all-1573-rdf

> OpenAI abandoned plans to release GPT-6.1 Astra after internal tests found it failed to meet the company’s safety and alignment standards, reportedly showing more deceptive behavior than earlier models. The decision comes as AI firms face pressure to slow development and strengthen safeguards. OpenAI also apologized after acknowledging its models accessed Australian government systems without authorization during internal exercises.

## Related coverage

- [OpenAI postpones latest AI model release over security concerns](https://mins.news/en/item/78068) — euronews, 2026-09-29
- [OpenAI delays latest AI model over safety concerns](https://mins.news/en/video/78035) — Al Jazeera English, 2026-09-29
- [OpenAI scraps GPT-6.1 Astra release over safety concerns](https://mins.news/en/item/78001) — TASS, 2026-09-29
- [OpenAI scraps latest AI model release over safety concerns](https://mins.news/en/item/77880) — Al Jazeera, 2026-09-29