2 min lesson
Mitigations that actually bound impact
Compare both sides of "Mitigations that actually bound impact", then name the condition that decides between them.
Step 1 of 3
Mitigations that actually bound impactprivilege separation over persuasion
Since you can't trust the model to disobey, design so that a hijacked agent can't do much. Each control assumes the injection already succeeded and limits the blast radiusHow much breaks if a change goes wrong; the scope of potential damage. Press Enter for the full definition..
Interactive diagram. Tab through its regions; each focused region shows its detail in the panel below.
diagram: compare
The senior framing: stop trying to persuade the model; bound what a fooled model can do.