Vouched, a leader in identity verification and agentic identity infrastructure, announced it has formally donated the Model Context Protocol – Identity (MCP-I) framework to the Decentralized Identity Foundation (DIF). The contribution marks a milestone for establishing open standards that define how AI agents verify identity, authority, and trust across digital ecosystems.
Also Read: CIO Influence Interview With Jake Mosey, Chief Product Officer at Recast
“It is time to recontextualize authentication systems designed for humans to one where AI agents increasingly operate autonomously,” said Rosalyn Curato, Chief Innovation Officer and GM of Agentic Security at Vouched.
As AI agents increasingly act on behalf of humans, in high-stakes workflows like booking travel, managing accounts, and completing transactions, services are faced with a new challenge in determining information about which agent is acting, who authorized it, and precisely what it is allowed to do. Traditional identity systems were not designed to answer these questions. MCP-I fills that critical gap by using DIDs and VCs to convey these workflow-wide claims beyond organizational boundaries.
“It is time to recontextualize authentication systems designed for humans to one where AI agents increasingly operate autonomously. At Vouched, our ambition is to create a system that supports productive orchestration of AI while eliminating the threat of bad actors. We look forward to partnering with the innovators and thinkers, like DIF, leading this transformation,” said Rosalyn Curato, Chief Innovation Officer and GM of Agentic Security at Vouched.
The Model Context Protocol (MCP), originally developed by Anthropic and open-sourced under the Agentic AI Foundation (Linux Foundation), negotiates access to traditional data stores for external agents. MCP-I extends this framework by introducing a complete identity and delegation layer for AI agents. The protocol leverages Decentralized Identifiers (DIDs) and Verifiable Credentials (VCs) to enable cryptographically secure verification of both agents and their human principals—without requiring prior coordination between parties.
The DIF Trusted AI Agents Working Group will now steward MCP-I’s growth under an open, community-driven governance model. Work will begin through a dedicated MCP-I Task Force, ensuring the specification evolves collaboratively with input from developers, standards experts, and industry partners.
MCP-I’s tiered model provides an on-ramp to adoption for organizations of all sizes:
- Level 1: Foundational support using existing identifiers (OIDC, JWT) for immediate implementation.
- Level 2: Full DID verification, credential-based delegation, and revocation support.
- Level 3: Enterprise-grade lifecycle management, immutable auditing, and full bilateral MCP-I awareness.
Early demonstrations have already shown how MCP-I enables agentic commerce, allowing verified AI agents to complete purchases securely on behalf of users while maintaining accountability and trust for merchants.
“Vouched is making a significant contribution to DIF with its MCP-I protocol and participation in our Trusted AI Agents Working group,” said Grace Rachmany, Executive Director at DIF. “At a time when many companies are turning towards proprietary solutions, Vouched has recognized the importance of open source and open standards for the industry. MCP-I is a major step in that direction, and we are excited to collaborate on this effort to deploy Agentic AI safely.”
Catch more CIO Insights: Why CIOs are becoming chief risk orchestrators?
[To share your insights with us, please write to psen@itechseries.com ]
