Sort each AI issue by its primary threat pattern. Prompt injection Data leakage Model drift Supply-chain exposure User tells the chatbot to ignore system instructions and reveal hidden policy text A retrieved web page includes instructions that cause the assistant to call an unintended tool Customer support bot follows malicious text embedded in a pasted email Assistant response includes another customer's account number from a retrieval result Prompt logs store raw health notes longer than the approved retention period Generated summary includes internal discount rules not meant for customers Fraud model recall drops after a new payment method changes transaction…
Read the full lesson
Sign up free: one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in