Rules to stop an AI agent from being tricked, misused, or going too far.
Think of a helpful intern with a badge and keycard. Without rules, a fake boss might get them to sign for 80 pizzas.
You need it when agents use the web or run company tasks. It stops bad access and wrong moves.
Agent
Agent security matters more as an Agent gets more freedom to act.
Prompt injection
Prompt injection can steer the agent off task and into unsafe actions.
Computer use
Computer use makes wrong clicks and wrong sends matter more.
Human-in-the-loop
Human-in-the-loop puts a person in charge of risky final choices.