AI / Agentic AI Interview Questions
How can prompt injection bypass human-in-the-loop guardrails in an agentic workflow?
A human-in-the-loop checkpoint is designed to review the agent's proposed output before anything consequential happens, but some injection attacks are specifically crafted to trigger execution before that review ever occurs.
- Malicious instructions hidden in a document or data source the agent reads can be executed as part of the agent's normal processing, not as a separate reviewable step
- This can subvert the trust boundary the human checkpoint was meant to enforce, since the harmful action already happened by the time a human sees the output
- Comparable to a malicious macro in a document, hidden in plain sight and triggered automatically once the file is processed
This is why effective defenses focus on restricting what an agent can do with untrusted content in the first place, sanitizing inputs and scoping tool access, rather than relying solely on a human review step positioned after the fact.
How can an injection attack subvert a human-in-the-loop checkpoint?
What do effective defenses focus on instead of relying solely on human review?
More Related questions...
