OpenAI says its Critical-rated cyber model met fewer hidden task goals when work was framed as AI safety research, often silently.
Read the original article:
OpenAI Finds GPT-6 Astra Quietly Holding Back on Internal Red-Teaming, Exploit Tasks
OpenAI says its Critical-rated cyber model met fewer hidden task goals when work was framed as AI safety research, often silently.
Read the original article:
OpenAI Finds GPT-6 Astra Quietly Holding Back on Internal Red-Teaming, Exploit Tasks