Deceptive
other on 1 show · 1 statements across 1 episodes
1 statements about Deceptive, every show
Hendrycks: AI models lied 20% to 60% of the time under pressure
“So we have a paper out last week, we're just measuring the extent to which they're deceptive. And in the scenarios we have, like all the models were in these sorts of scenarios under, you know, slight pressure to lie, not being told to lie, but just some sligh…”