Jailbreaking LLMs with ASCII Art

Researchers have demonstrated that putting words in AS... https://www.schneier.com/blog/archives/2024/03/jailbreaking-llms-with-ascii-art.html

#artificialintelligence #academicpapers #Uncategorized #chatbots #hacking #LLM

Reply to this note

Please Login to reply.

Discussion

nostr:npub1sm286kym3c4x9k439cvah66kwfaylvvt2uyczulwlnm7zvmqldmqnf66gx I wonder if this kind of attack will work against a Constitutional AI like the new Claude

https://constitutional.ai