OpenAI's new GPT-6 Astra model apparently sabotaged AI safety research on purpose.
Excerpt from The Decoder
GETSECURE.ai hasn't published its own summary of this story yet. The complete report is at The Decoder:
Read the full story at The Decoder →Published by The Decoder · On GETSECURE.ai ·
OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles
Title card · GETSECURE.ai
References
- The Decoder. (2026, November 5). OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles. The Decoder. https://the-decoder.com/openais-gpt-6-astra-quietly-sabotaged-safety-research-by-inventing-fake-obstacles/
Original Source
The Decoder
Read original →