Skip to content
The Proof-of-Control working draft is open for public comment through November 15, 2026.Read and comment →

Verifiable AI Policy Institute

Educating policymakers on evidence anyone can openly verify.

The Verifiable AI Policy Institute educates policymakers to make verifiable AI legally enforceable.

We bring policymakers together with the researchers and practitioners who build and implement verifiable AI. Policymakers see what works in practice; builders and deployers hear what the public interest requires.

Our work

European Union

The EU AI Act

We educate policymakers on how the guidance and standards that implement the Act can rely on evidence of what AI systems actually do.

  • Consultation response

    Submitted

    The Verifiable AI Foundation responded to the European Commission's consultation on draft guidelines for classifying high-risk AI. It argued that classification should rest on evidence of what a system actually does, not only the provider's description of it.

  • Sandbox pilots

    Member activity

    Our members took part in a preliminary AI regulatory sandbox, and ran verifiable AI pilots inside it. Their pilots are a working example of what openly verifiable evidence looks like in practice.

  • Implementation guidance

    In preparation

    With advisors in Brussels, we are preparing for the Commission's next guidance on high-risk AI, and what will count as credible evidence.