Comment on Researchers say they had a ‘100% attack success rate’ on jailbreak attempts against Chinese AI DeepSeek

RobotToaster@mander.xyz ⁨2⁩ ⁨weeks⁩ ago

Compared to other frontier models, DeepSeek R1 lacks robust guardrails

Oh no, it’s easy to make the ai running on my computer do what I want it to do, the horror.

source
Sort:hotnewtop