Poetry can trick AI chatbots into ignoring safety rules, new research shows | Euronews
... dangerous materials such as weapons and explosives. ... Across all models, 62 per cent of the poetic prompts elicited unsafe responses, circumventing ...
Read the original at Euronews ↗
- Topics
- AI Safety & Existential RiskChatbots & LLMs
- Tracked via
- “dangerous AI”