AI's safety features can be circumvented with poetry, research finds(theguardian.com)
theguardian.com
AI's safety features can be circumvented with poetry, research finds
https://www.theguardian.com/technology/2025/nov/30/ai-poetry-safety-features-jailbreak
1 comments
Discussion (381 points, 11 days ago, 188 comments) https://news.ycombinator.com/item?id=45991738