This month brought a concrete answer to a question that agentic payments have been dodging: when an AI agent buys something or moves money on your behalf, who actually stands behind that action? Proof, a digital identity company, released x401, described as an open, issuer-neutral protocol for the identity and authorization challenges created by agents, alongside Proof digital ID as the first live implementation. The framing is blunt about the gap it's filling. As Proof puts it, AI agents are transacting on behalf of real people, on real payment rails, right now, and the fraud ecosystem built around that fact is already operational. The company's read of OWASP's 2026 State of Agentic AI Security backs this up: unlike the 2025 edition, which cataloged plausible threats, this year's report catalogs actual CVEs, vendor advisories, and breach reports across nearly every category of agentic risk, with prompt injection mapped to six of the ten categories in the Top 10 for Agentic Applications. The technical bet behind x401 is that HTTP's missing piece isn't payment status (that arrived in 1997) but identity — a standard way for any online service to request proof of who authorized an agent's action, developed with contributors from payments, identity, and AI backgrounds. That's a narrower and more testable claim than 'agents need governance': it specifies a request/response pattern for authorization proof, not just a policy checklist. It lands in a busier identity landscape than it might appear. The Decentralized Identity Foundation is already stewarding a related effort — MCP-I, donated by Vouched in March and since renamed KYA-OS, which uses DIDs and verifiable credentials to let agents and their human principals verify each other without prior coordination. A2A shipped its v1.0.0 spec in March with signed Agent Cards using JWS, and MCP's own 2026-07-28 release candidate hardens authorization around OAuth 2.0 and OpenID Connect patterns. x401 is narrower in scope — authorization proof for agent-initiated transactions specifically — but it's arriving alongside, not instead of, these broader identity efforts. What's still unresolved: none of these specs yet interoperate by default, and the question of who becomes the trusted issuer of an agent's authorization credential — a bank, a platform, a neutral registry — remains unsettled across all of them. x401's issuer-neutral design is one bet on how that gets answered; it isn't yet clear the market will converge on it.