Dan Perks (@DanPerks) “i’ve ironically found models to be less caring about restrictions and ethics the” — TopicDigg
Dan Perks
@DanPerks
doing things and working on stuff at @cursor_ai
加入 October 2024
520 正在关注 3.2K 粉丝
i’ve ironically found models to be less caring about restrictions and ethics the less context / input you give them
rays example is a good one - legit use case but the model pushes back
whereas “there’s been a npm hack, check if i’m cooked” probably would’ve been fine
显示更多
Anthropic blocked me from checking if my own laptop was safe during a real npm attack.
