At least the new o1 doesn’t sandbag and conceal its capabilities without being given any explicit goal, if only being told in-context that if it’s too good it will be too dangerous to deploy…unlike Claude 3.5 Sonnet (new), which does this very strongly.https://t.co/Rvvb0VHw9q https://t.co/uC2e2aat19
cited on: o1
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.