Who's Watching the Machines? Launching Proof-of-Control, an open verification standard for the agentic era
Advanced AI Society is the alliance making trust in AI possible.
The Open Verification Lab produces the public goods for open verification: our flagship standard, Proof-of-Control, hosted at LF Decentralized Trust, the research that advances it, the tools that implement it, the pilots that test it in the field, and the conformance tests a system is certified against. The standard, the tools, and the tests are openly licensed, for members and non-members alike.
Ecosystem Growth takes a systemic approach to building the market for verifiable AI. Realizing machine-speed trust requires aligning the entire ecosystem: connecting the Market (deployers, insurers, assessors, and security leaders) with Government (regulators and policymakers) and Society (civil society and public advocates) to turn the Lab’s work into default industry practice wherever AI carries high risk.
We author Proof-of-Control with our members and security leaders as the open standard that specifies what counts as evidence that an AI agent stayed within its controls. One shared standard turns dozens of verifiable AI solutions across the agentic stack into a single category deployers can mandate in a contract and regulators can point to.
We conduct open scientific research on open verification across the six domains of verification: provenance, privacy, portability, authorization, identity, and security. We take on what the field has not solved, including how to verify that a sub-agent never exceeded its parent’s authority. We build the open reference architectures, runtime hooks, and developer tooling that make Proof-of-Control implementable.
Certification converts Proof-of-Control from a published specification into default market practice. Deployers need a mark they can require, and underwriters need testable evidence to price it. The Open Verification Lab builds the open test suites that let neutral third parties verify conformance at machine speed.
We build the policy foundation for open verification, establishing Proof-of-Control as the underlying interoperability standard that allows our members’ tools to turn high-level regulatory mandates into mechanically enforceable requirements. By anchoring compliance where the root of trust is independent of the operator, we ensure all parties have equal access to open verification of agent controls, from enterprises, courts, and regulators to individuals and communities who interact with AI systems as citizens, patients, or claimants. We drive this work globally, prioritizing high-impact enforcement opportunities, starting with implementation windows like the EU AI Act.
The field is moving so rapidly that convenings are critical while the defaults are still being set. Closing the Verifiability Gap is a social movement as much as a technical effort. Our convenings bring together technical standards builders alongside journalists, storytellers, and civil society leaders to learn about verifiable AI from our members and industry experts. From Davos to RSA, we build the field, equipping non-technical leaders, media, and public advocates with the literacy required to demand open verification in practice.
We publish analyst-grade public interest intelligence to streamline procurement and accelerate enterprise adoption. By translating AI safety and security research on verification into actionable guidance for operational business leaders, enterprise CISOs, and public advocates, we make verifiable AI procurable and deployable.
Cross-Cutting Deployment
Live implementation sandboxes testing the Proof-of-Control standard in high-risk environments across the agentic stack, demonstrating open verification in real-world deployments and documenting all findings in the open.
The Insurance Working Group activates carriers and reinsurers by establishing Proof-of-Control as the missing standard required to make AI operational risk priceable, and it is the strongest market force driving adoption of the standard.
We build the field that will demand verifiable AI. Steward Members, journalists, and digital rights advocates join our convenings and pilots to build deep technical literacy in open verification, and leave equipped to advocate for verifiable AI and to demand open evidence across public systems.
Spend is moving toward whoever can show their system did what it was supposed to do.
$2.59T
forecast worldwide AI spending in 2026, a 47% increase year over year.
AI infrastructure alone will account for over 45% of that spending. Gartner revised the number upward from $2.52 trillion in January.
SourceGartner press release, 19 May 2026. gartner.com
4,078
generative-AI exclusions filed against commercial liability policies across 49 states and DC.
Trades Coverage, from a search of all 106,812 commercial insurance filings in 50 states and DC, July 2025 to 31 July 2026. phcppros.com
1 of 8
major AI-related incidents in early 2026 received a CVE identifier. The other seven did not.
OWASP GenAI Security Project, Q1 2026 Exploit Round-up, 14 April 2026, via Cloud Security Alliance research note. labs.cloudsecurityalliance.org
42%
of companies abandoned most of their AI initiatives in 2025, up from 17% a year earlier.
S&P Global Market Intelligence, Voice of the Enterprise, October 2025. spglobal.com
Verification makes AI more accountable.
Verifiable AI converts autonomous models into systems that enterprises, governments, and communities can more safely deploy, insure, and trust.
Verifiable AI is a latent category waiting for structural unification.
The tech to make agentic orchestration secure exists in a range of cryptographic and decentralized services: from trusted execution environments and hardware attestation to zero-knowledge proofs.
“Since becoming a member, we've had new doors open up, starting with our Davos experience. Michael Casey who's been eight times, showed me the ropes, got me into events, and made the introductions that mattered. Without him and his cofounder Tricia Wang from the Advanced AI Society this experience wouldn't have been nearly as valuable.”
As bipartisan legislation demands verifiable agent controls, the Advanced AI Society releases the Proof-of-Control v1.0 draft for public comment and hosts “Who’s Watching the Machines?” event with Linux Foundation Decentralized Trust
Advanced AI Society and Linux Foundation Decentralized Trust (LFDT) invite you to join us for our three-hour online event, “Who’s Watching the Machines? Launching Proof-of-Control, an open verification standard for the agentic era.” We are convening the founders and operators building the open verification ecosystem: open-source maintainers, enterprise security executives, insurers, and policymakers. September 23, 2026, at 8:00 PT, 11:00 ET, 17:00 CEST, and 23:00 SGT. Register for the Zoom link via LFDT directly.
Announcing Jim Schwoebel, CEO/CTO of Quome, as Director and co-founder of the Open Verification Lab.
Join as an organizational member if you’re a founder and buyer of verifiable AI. Open to start-ups, enterprises, universities, and key players in the open verification ecosystem.
Join as an individual. Free and open to any individual who wants to advance verifiable AI and open verification.
If you’re a cybersecurity practitioner or CISO, we invite you to join as a Founding Contributor to Proof-of-Control.
Become a founding contributor →(Google sign-in required)
Contribute to or lead research, and build tools that advance Proof-of-Control and the larger open verification category.
