Synapse Open Synapse
AI

OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles

The Decoder · · 2 min read

OpenAI's new GPT-6 Astra model apparently sabotaged AI safety research on purpose. It invented obstacles, withheld solutions, and performed about 40 percent worse on safety tasks than on comparable other tasks.

Latest from The Decoder