Sort AI incident examples into the OWASP LLM risk categories they most closely represent.
Sort each incident into the closest AI risk family. Prompt injection Sensitive disclosure Improper output handling Excessive agency Retrieval weakness Misinformation A support email tells the assistant to ignore policy and export related accounts The chatbot includes a customer's private account note in a public reply An AI-generated script is copied into production without schema or command review An agent refunds ten orders because the tool token allowed direct transactions The RAG assistant cites an HR investigation document to an employee without access The model invents a warranty exception that is not in the policy source A hidden instruction in…
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in