OpenMatter and Hashgraph Proposed AI Agent Compliance Standards
A new framework aims to secure autonomous agents through zero-knowledge proofs and policy-based action filtering.
Updated on Oct. 7, 2026 in Artificial Intelligence
Live Poll
Should your company adopt standardized security frameworks to govern its autonomous AI agents?
OpenMatter Network and Hashgraph Online have released a proposed standard for verifying AI agent compliance. This research-stage framework uses zero-knowledge cryptography to prove adherence to security policies without exposing underlying sensitive input data.
Why it matters
As autonomous AI agents increasingly operate across multi-system environments, organizations face significant privacy and security risks. This framework provides a standardized method for governing agent behavior and generating verifiable evidence of regulatory compliance.
The system utilizes a shared framework to block agent actions that deviate from defined policies. By employing zero-knowledge proofs—a cryptographic method that allows one party to prove a statement is true without revealing information beyond the fact itself—the system verifies compliance.
The players
OpenMatter Network
A developer network focused on creating protocols for verifiable AI and autonomous agent compliance.
Hashgraph Online
An enterprise-grade distributed ledger platform that provides the underlying infrastructure for decentralized applications and verified data exchanges.
The details
The framework acts as a validation layer for autonomous systems by checking agent decisions against pre-set boundaries. It functions by intercepting agent-initiated requests, ensuring that only actions meeting defined safety and regulatory criteria are executed. By using a decentralized verification architecture, it aims to maintain auditability in complex, multi-agent operational environments where centralized control is often technically infeasible.
Timeline
July 2026: OpenMatter Network joined the Hashgraph Online Partner Programme.
October 7, 2026: Proposed standards document released for public review.
October 31, 2026: Deadline for public comment on the draft.
The Tech Race
This framework marks a shift toward standardized governance within the growing ecosystem of autonomous AI infrastructure. It follows the recent collaboration within the Hashgraph Online Partner Programme to solve the industry-wide challenge of securing cross-system agent operations.
Developers and system administrators can currently access the draft on GitHub to evaluate the proposed compliance mechanisms. Once finalized, the framework aims to become a standard tool for entities managing multi-system agent workflows that require verifiable security and audit logs.
The takeaway
The industry is moving toward cryptographic verification to solve the black-box problem in autonomous AI agents. Observers should track the release of the revised draft and the finalized test vectors following the October 31 comment window.
What happens next
The working group will review feedback following the October 31 comment deadline to publish a revised draft, which will incorporate updated conformance requirements and specific test vectors.
Further reading
For more on evolving governance for machine intelligence, visit the Artificial Intelligence section.
Source note: This article includes information reported by Enterprise Times.
Live Poll
Should your company adopt standardized security frameworks to govern its autonomous AI agents?
