Poster: Towards ProofWeave: A Privacy-Minimised, Integrity-Anchored Evidence Plane for Continuous Agentic Assurance

2026-09-28

AI Summary

This study addresses the absence of real-time, policy-bound, and privacy-preserving evidence chains in the continuous monitoring of AI agents by proposing the ProofWeave framework. This framework introduces the concept of an "evidence chain at recording time," wherein a bounded Weaver Agent translates policy intents into proof obligations at action boundaries. It generates privacy-minimizing transactions that bind control responses to policy snapshots, storing them in an immutable ledger to construct a proof graph. Experimental results demonstrate that this approach reduces the number of candidate bindings to one, decreases verification operations to approximately 26, and lowers storage overhead from 0.79 MiB to 0.15 MiB. Consequently, ProofWeave significantly enhances adjudication efficiency, privacy protection, and tamper resistance.

Abstract

Agentic AI systems increasingly act via tools, memory, delegation, and external services. Existing observability and provenance mechanisms can reconstruct events post hoc, but they rarely show, at the time of the record, whether each policy-relevant action was checked by the intended control before execution. This leaves a trust-observability gap for continuous monitoring, detection, and response: later assurance may rest on evidence that is incomplete, privacy-leaking, mutable, or detached from the policy context that governed the event. What's missing in the literature is contemporaneous, policy-bound evidence that the intended control was evaluated under the policy in force at the time. We introduce ProofWeave, a record-time chain-of-evidence concept for agentic AI assurance. At each policy-relevant action boundary, ProofWeave generates a privacy-minimised and integrity-anchored evidence transaction that binds (i) agent intent or action, (ii) control response, and (iii) a policy-at-time snapshot. Each transaction is committed to an append-only ledger and materialised into a derived proof graph. A bounded Weaver Agent translates policy intent into proof obligations, while deterministic validators check evidence completeness, privacy minimisation, policy binding, and integrity. In the minimal scenario, an agent attempts to transmit a secret to an unapproved external sink. The audit compares a logs-only correlation baseline with ProofWeave across verdict latency, join ambiguity, privacy exposure, tamper detection, and resistance to graph-only proof injection. ProofWeave reduces candidate bindings per verdict from up to `10,201` to one, validation operations from up to `10,201` to approximately `26`, and assurance evidence storage from `0.79`\~MiB to `0.15`\~MiB per project.

Problem

• Agentic AI
• trust-observability gap
• policy-bound evidence
• continuous assurance
• provenance

Innovation

• ProofWeave
• Agentic AI Assurance
• Chain-of-Evidence
• Privacy-Minimisation
• Policy-Bound Verification