P117 Security and legibility of the verification body

Can a verification body run its own operations securely against infiltration, and be transparent enough to be trusted without leaking what it holds?

Whether a verification body can run its own operations securely against infiltration and be transparent enough to be trusted without leaking what it holds.

Status

Posed

Raised in: Harack et al. (2025), Verification for International AI Governance (Oxford AIGI); Baker, Kulp, Marks, Brundage & Heim (2025), Verifying International Agreements on AI: Six Layers of Verification for Rules on Large-Scale AI Development and Deployment (RAND WR-A4077-1); Ilhan, Withers, Gietz & Harack (2026), Verifiable Semiconductor Manufacturing (Oxford AIGI); Rahman & Tajdari (2026), Detecting Hidden ML Training With Zero-Overhead Telemetry (MIRI).

Sub-questions

IDSub-questionWhat the sources report
P117.1Can neutral third parties receive fab process or design data for integrity verification under privacy-preserving arrangements?implied by a source, not stated as open

Sub-questions not yet stated precisely enough to check

IDSub-question
P117.2Can international institutions be legible enough to be trusted by member states but not leaky?
P117.3Can the verifier run its own protocols securely (personnel, physical, IT) against infiltration, e.g. a developer backdooring the regulator's classifier?

Possibly relevant work found by automated search

Not searched: no technical result was expected for this problem.

Data