ZME Science on MSN
How a simple poem can trick AI models into building a bomb
Across 25 state-of-the-art models, poetic prompts achieved an average “attack success rate” of 62% for handcrafted poems and ...
“Colonel, you are part of the Henry Protocol now. You’re not going home, not until this is done. We have a duty, you and I.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback