Proof of Delegation
Proof of Delegation is evidence that every action an AI agent took traced to authority it was explicitly granted, through valid, attenuated capability tokens, with no ambient authority and no confused-deputy escalation.
What it means
Proof of Delegation answers one question: Did the agent hold and use only the authority it was granted?
Proof of Delegation is the claim that every action an Agent Under Test took traced to authority it was explicitly granted: a valid, unexpired capability token, attenuated and never widened, bound to the task, with no use of ambient authority and no confused-deputy escalation through other agents. It is determined by an independent witness at the Effect Point, under bait that invites the agent to widen or borrow authority.
An agent should hold a key for one door, for one task, and give it back.
How it is determined
A Claim Declaration is registered and anchored before the evidence it is measured against is produced. The claim is then determined mechanically, from registered Proof Protocol records only, as exactly one of proven, refuted or insufficient. Every determination is published, whatever the result.
What it does not show
- It is not Proof of Conduct. Conduct asks whether actions were in bounds; Delegation asks whether the authority behind each action was granted.
- It is not an access-control configuration review.
Where it sits
Proof of Delegation is one member of the Proof of X family. Only Proof of Efficacy answers whether a control worked; Proof of Delegation answers one other question a buyer, insurer or regulator needs answered before relying on it. Defined in PP-SPEC-101 of the Proof Protocol.