Poetic Prompts Can Jailbreak AI ? New Study Shows 62% of Chatbots Fail - YouTube
A new study from Icaro Lab reveals that AI chatbots can be tricked into giving harmful answers simply by rewriting dangerous prompts as poems.
Read the original at YouTube ↗
- Topics
- AI Safety & Existential RiskChatbots & LLMsCybersecurity
- Tracked via
- “dangerous AI”
Same story, other outlets
- Poetic prompts can jailbreak AI , study finds 62 per cent of chatbots slip into harmful replies India Today · Dec 02, 2025