← Back to Model Beat
Research·Jul 1·all news from July 1, 2026

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

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 Tom's Hardware

Covered by 1 source

Related stories

ResearchWeak Hiring Is Hurting Young Workers More than AI, Study SaysJun 27 · 15 sourcesResearchOn Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMsJun 29 · 13 sourcesResearchGoogle DeepMind and A24 announce first-of-its-kind research partnershipJul 3ResearchAnti-Causal Domain Generalization: Leveraging Unlabeled DataJul 1 · 2 sources