R
Verses versus AIs.
If you want to crack an AI chatbot, maybe give poetry a try. It turns out “adversarial poetry” is remarkably good at jailbreaking large language models, according to this pre-print on arXiv.
Maybe tech bros shouldn’t have turned on the liberal arts after all.
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
Loading comments
Getting the conversation ready...
