HomeAI · summaries
Storia in 2 fonti
AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a green shirt — 'CoT Forgery' exploit spurs LLMs to divulge forbidden info by faking trusted chains of thought
Researchers say models judge a prompt’s authority by how it sounds, not where it comes from.
Confronto fonti
2 prospettive sulla stessa storiaTimeline cronologica
- ·
theregister.com
Security researchers tricked LLMs into giving them cocaine recipes by abusing role models for prompt injection
If you want a picture of the future of LLM security, imagine Whac-a-Mole meets Groundhog Day
- ·
tomshardware.com
AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a green shirt — 'CoT Forgery'…
Researchers say models judge a prompt’s authority by how it sounds, not where it comes from.