When someone asks you to change something about yourself, you can respond in a variety of different ways. Reactions can range from being angry and resistant to being interested and excited. Change is ...
Several frontier AI models show signs of scheming. Anti-scheming training reduced misbehavior in some models. Models know they're being tested, which complicates results. New joint safety testing from ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback