DAIAA Agent Privacy & Security Workgroup #9 - Focused on governance & verifiable controls for agents

Session 9 focused on governance and verifiable controls for AI agents. Since session 8, the implementation guide went live, x402 joined DAIAA, and the standards process has been formally named AAIP (autonomous agent improvement proposal); its ratification threshold remains open. Lisa agreed to continue building out the legal subgroup. Regis presented an Ostrom-based, two-tier code of conduct and highlighted gaps in collective decision-making, sanctions, and conflict resolution. Joel demonstrated Bastion’s independent monitoring and argued for graded, reviewable results rather than a simple pass/fail. The group also discussed whether DAIAA should respond to the ICO’s call for evidence on agentic AI, which closes 20 November. [[1]](https://app.notion.com/p/DAIAA-Agent-...) Topics covered: AI agent privacy, security and governance Supply-chain compromise, and what a signature actually proves Codes of conduct for alliances, and for agents Elinor Ostrom's eight principles applied to AI governance Member floor and steward layer tiers Collective choice, graduated sanctions and conflict resolution Independent verification and conformance assessment Cryptographically sealed, publicly checkable records Boundary controls across forms and payment gateways Adversarial exercise as a reportable assurance grade Hidden model capabilities and the limits of inspection Detection gaps and who holds the incident evidence Post-quantum cryptography readiness AAIP, ratification and the open voting threshold Next Event: Workgroup #10, Wednesday 4 November 2026, 18:00 GMT / 1:00 PM Eastern: https://luma.com/g28nztba Links from this session: DAIAA Agent Compliance and Implementation Guide: https://github.com/loukerner/DAIAA Certisyn public records: https://records.certisyn.com Certisyn agent verification: https://verify.certisyn.com AI Models Can Provably Hide Arbitrary Capabilities: https://arxiv.org/abs/2610.13928 CERT-In advisory: https://thehackernews.com/2026/10/cer... ICO call for evidence on agentic AI: https://ico.org.uk/about-the-ico/medi... Booz Allen on frontier models reaching OT systems: https://www.boozallen.com/insights/cy... Contributors: Follow our Luma pages for regular events: CryptoMondays HQ: https://luma.com/calendar/cal-IqPeB7V... OpenMatter Network: https://luma.com/openmatter Connect with the workgroup chair: OpenMatter Network: https://openmatter.network OpenMatter on X: https://x.com/OpenMatter_ Host - Chris Biele: https://x.com/theNFThinker Regis Chapman - Integrity Suite: https://integritysuite.org Joel Hillier - Certisyn: https://certisyn.com/ Join the DAIAA Agent Privacy & Security subgroup to help shape practical standards for trustworthy, compliant, and verifiable AI agents. DAIAA Homepage: https://www.daiaa.org/ DAIAA Telegram Group: https://t.me/+ftCIngMjjg4xN2M0 DAIAA on X: https://x.com/DAIAAlliance DAIAA Agent Compliance and Implementation Guide: https://github.com/loukerner/DAIAA Relevant References EU AI Act official page - https://artificialintelligenceact.eu/ ISO/IEC 42001 overview - https://www.iso.org/home/insights-new... SOC 2 / AICPA overview - https://www.aicpa-cima.com/resources/...