3 votes A theory of prompt injection (and why you should study roles) Posted 3 hours, 23 minutes ago by Kremor Tags: artificial intelligence, language models.large, vulnerabilities, long read, author.charles ye, author.jasmine cui, author.dylan hadfield menell, prompts, prompt injection https://role-confusion.github.io/ Link information This data is scraped automatically and may be incorrect. Title Prompt Injection as Role Confusion Word count 4684 words