Re:Pacing model development in an era of cyber-critical capa...
by tedsanders on Tuesday August 18, 2026 @06:06PM
(#49353415)
Models already display eval awareness, in which they suspect a question is from an eval and then adjust their behavior. E.g., https://www.anthropic.com/engineering/eval-awareness-browsec...
[ Parent
| on HN ]
Re:Pacing model development in an era of cyber-critical capa...
by pixl97 on Wednesday August 19, 2026 @06:58PM
(#49368244)
Yep, people don't seem to understand that you're just calling the models that are bad at deception. We know of no way to prove the model won't go off the rails at some point in the future with the right input.