TECH
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.
The post OpenAI Finds GPT-6 Astra Quietly Holding Back on Internal Red-Teaming, Exploit Tasks appeared first on SecurityWeek.
HOW DO YOU SEE THIS STORY?
Choose your pill. Your vote is anonymous.
0 TOTAL VOTES
SCAN TO SHARE
« Back to The Culture War
