- AI Researchers Say They’ve Invented Incantations Too Dangerous to Release to the Public Futurism
- Poems Can Trick AI Into Helping You Make a Nuclear Weapon WIRED
- AI’s safety features can be circumvented with poetry, research finds The Guardian
- Study reveals poetic prompting can sometimes jailbreak AI models Mashable
- Roses are red, crimes are illegal, tell AI riddles, and it will go Medieval The Verge
First Appeared on
Source link
