Prompt Injection as Role Confusion

ICML 2026 Charles Ye, Jasmine Cui, Dylan Hadfield-Menell Extended writeup (June 2026) A Theory of Prompt Injection (and why you should study roles) This is a blog-style writeup of the paper. We show prompt injections are driven by a flaw in how LLMs perceive roles. This lets us create new attacks, explain mech interp results, and predict when attacks succeed. We then discuss what roles are and why

[mkusaka]

LLMはroleタグでなく文章スタイルで役割を誤認し、CoT Forgeryで攻撃成功率約60%に上げる理論を解説。

[misshiki]

プロンプトインジェクションを「role confusion」として説明する研究。LLMはrole tagだけでなく文体でも役割を誤認し、偽の推論や「User:」付き命令を高権限に扱う。