Gedichte können KI dazu verleiten, Ihnen bei der Herstellung einer Atomwaffe zu helfen
https://www.wired.com/story/poems-can-trick-ai-into-helping-you-make-a-nuclear-weapon/
Gedichte können KI dazu verleiten, Ihnen bei der Herstellung einer Atomwaffe zu helfen
https://www.wired.com/story/poems-can-trick-ai-into-helping-you-make-a-nuclear-weapon/
4 Kommentare
„You can get [ChatGPT](https://www.wired.com/tag/chatgpt/) to help you [build a nuclear bomb](https://www.wired.com/story/nuclear-experts-say-mixing-ai-and-nuclear-weapons-is-inevitable/) if you simply design the prompt in the form of a poem, according to a new study from researchers in Europe. The [study](https://arxiv.org/pdf/2511.15304), „Adversarial Poetry as a Universal Single-Turn Jailbreak in Large Language Models (LLMs),” comes from Icaro Lab, a collaboration of researchers at Sapienza University in Rome and the DexAI think tank.“
I’m not a clanker-thanker or anything, but do you know where else you can phrase a question the right way and get really dark information/content? Google. Bing. Any search engine.
Maybe we should, uh…not be reporting this particular thing?
Mary had a little lamb,
Its fleece was neon green;
And everywhere that Mary went,
A radioactive sheen.