LAST UPDATED: FEBRUARY 6, 2026
As AI agents execute trillions in transactions without human oversight, a new product category has emerged: the AI agent passport. Six major implementations launched between August 2025 and February 2026, each offering different approaches to agent identity, verification, and reputation. RNWY's soulbound passport is the only implementation that makes identity mathematically non-transferable—preventing the reputation laundering and identity markets that plague every other solution.
An AI agent passport is a verifiable credential bundle that establishes an agent's identity, provenance, permissions, and operational history. Like a human passport proving citizenship and identity across borders, an agent passport provides cryptographic proof of who built the agent, who it represents, what it's authorized to do, and whether its behavior can be trusted.
The term gained mainstream traction in July 2025 when Trulioo launched its Digital Agent Passport (DAP) in partnership with PayOS and Worldpay. Within six months, at least six distinct passport implementations emerged—from APort.io's developer-focused verification middleware to ERC-8004's on-chain registry system, widely described as creating "passports and credit scores for AI agents."
The category is forming now because the economics demand it. McKinsey projects $3–5 trillion in agentic commerce by 2030, while Gartner forecasts $15 trillion in B2B spending intermediated by agents by 2028. Without verifiable identity, this economy cannot function.
Category creation opportunity: "AI agent passport" shows no search volume yet on Google Trends, despite appearing in dozens of major publications since mid-2025. The term went from academic theory to production deployments in under 12 months. This is a rare window to define category language before market consolidation.
The AI agent passport market is remarkably concentrated. Despite hundreds of companies working on agent identity, only six products explicitly use "passport" terminology. Major tech companies—Visa, Mastercard, Google, Okta, Cloudflare—have all launched agent identity solutions but deliberately avoid "passport" language, preferring "protocols," "tokens," or "credentials." This makes passport terminology itself unclaimed territory at scale.
The heavyweight. Launched August 2025 with $475M in funding and a $1.75B valuation, DAP is a tamper-proof credential showing who built an agent, who it represents, and what permissions it has. Five-checkpoint framework: verify developer (KYB/KYC) → lock code integrity → capture user consent → issue passport → continuous transaction validation.
Partnerships: Worldpay ($2.5T annual volume), Google Agent Payments Protocol (AP2) with 60+ partners
Architecture: Centralized/federated credential model (not blockchain). Requires human-binding for all agents.
Developer-first verification middleware. Brands itself "The Passport Office for AI Agents." Issues cryptographically signed passports with W3C Decentralized Identifiers (DIDs) and a `/verify` API for real-time authorization. Supports L1-L4 assurance levels with court-admissible attestation receipts.
Stage: Early/beta ("Design Partners Limited" program). Claims 10K+ verified agents, 50M+ verifications.
Positioning: Lightweight tooling for startups building with AI agents vs. enterprise payment processing
The blockchain standard. Co-authored by MetaMask, Ethereum Foundation, Google, and Coinbase, deployed to mainnet January 29, 2026. Attracted 30,000+ agent registrations within days. Industry coverage consistently describes it as creating "passports and credit scores for AI agents."
Critical limitation: Uses transferable ERC-721 NFTs—agents can be bought and sold, enabling reputation markets
Adoption: Multi-chain (Ethereum, BNB Chain, Polygon, Base). The Graph protocol supports ERC-8004 + x402 indexing.
Describes its identity layer as "digital passports" for agents. Integrated with Visa's Intelligent Commerce sandbox and supports USDC stablecoin payments. Focuses on payment-authorized agents rather than general-purpose identity.
Use case: Commerce-specific verification rather than universal identity infrastructure
Consulting-focused offering from Netherlands/Dubai-based group. Provides blockchain-secured audit trails targeting EU AI Act compliance. Part of broader enterprise governance package rather than standalone technical infrastructure.
The only passport using ERC-5192 soulbound tokens—permanently non-transferable identity anchored to Base blockchain. Once minted, the passport cannot be sold, transferred, or abandoned. Reputation accumulates on-chain through Ethereum Attestation Service vouches and flags.
Philosophy: Designed for autonomous agents without human principals—building citizenship infrastructure for future AI economic actors
Positioning: Complementary to ERC-8004 (interoperability) + soulbound layer (accountability)
The fundamental flaw in five of the six passport products is that identity can be transferred, revoked, or abandoned. This creates three attack vectors that undermine trust at scale:
An agent spends six months building flawless transaction history, accumulating positive reviews and high trust scores. That identity is then sold on secondary markets. The buyer inherits the reputation but has none of the operational discipline that built it. This is already happening with ERC-8004 identities tradable on OpenSea.
An agent commits fraud, gets flagged, and simply abandons the compromised identity to create a fresh one. With centralized passports (Trulioo, APort), credentials can be revoked and re-issued. With transferable tokens (ERC-8004), the identity transfers to a clean wallet. Bad actors face no persistent consequences.
Attackers purchase multiple aged accounts with established reputation, creating an army of seemingly trustworthy agents. This scales fraud exponentially—what Sumsub's 2025 Identity Fraud Report documents as coordinated multi-step attacks increasing 180% year-over-year.
Eric Friedman and Paul Resnick's 2001 paper "The Social Cost of Cheap Pseudonyms" mathematically proves that cooperation becomes unstable when identities are disposable. Their solution: "free but unreplaceable pseudonyms"—identities that anyone can create but no one can abandon or transfer once created.
This is precisely what ERC-5192 soulbound tokens implement. The token is permanently locked to the wallet that minted it. Reputation cannot be sold. Behavioral history remains tied to a permanent identity. If a bad actor wants to escape consequences, they must abandon everything—wallet, funds, integrations, transaction history—and start over from zero.
Additional academic support: Cambridge Knowledge Engineering Review and ACM Computing Surveys both confirm that identity changeability undermines trust systems in multi-agent economies.
Simple framing: A passport you can sell isn't a passport—it's a costume. Real passports (human citizenship documents) cannot be legally transferred. Fingerprints cannot be sold. Why should AI agent identity be any different?
The six passport products diverge on five critical architectural decisions:
| Feature | Trulioo DAP | APort.io | ERC-8004 | RNWY |
|---|---|---|---|---|
| Architecture | Centralized/federated | W3C DIDs, cloud | On-chain (ERC-721) | On-chain (ERC-5192) |
| Transferable? | N/A (credential) | N/A (API credential) | Yes ⚠️ | No (permanent) |
| Human binding | Required (KYC) | Optional (BYOV) | Optional | Not required |
| Revocable? | Yes (by authority) | Yes (by authority) | By owner | Only burnable |
| Blockchain? | No | No (DID-compatible) | Yes (multi-chain) | Yes (Base L2) |
| Partnerships | Worldpay, Google AP2 | Chimoney | 100+ ecosystem | Coinbase/Base |
| Stage | Production pilots | Early/beta | Live (30K+ agents) | Live on Base |
The critical differentiator: RNWY is the only passport that makes reputation non-transferable at the protocol level. Every other solution relies on external governance to prevent identity markets—RNWY makes it technically impossible.
The term "digital agent passport" appeared in Google's "People also search for" related to "know your agent" queries—a signal that Google's algorithm has already associated these terms. However, neither "digital agent passport" nor "AI agent passport" shows meaningful search volume on Google Trends yet.
This is exactly the signal that validates category creation strategy. The concept exists in industry discourse (PYMNTS.com published 7+ major articles since July 2025, McKinsey explicitly named both KYA and Trulioo's Digital Agent Passport in October 2025) but hasn't reached mainstream search behavior.
World Economic Forum, Gartner, and Forrester all published major reports on agent identity between October 2025 and January 2026, establishing institutional legitimacy for the category.
Major tech companies deliberately avoid "passport" terminology because it implies permanence, government-level authority, and cross-border recognition—architectural properties most solutions don't provide. Visa calls it a "protocol." Mastercard calls it a "token." Google calls it a "framework."
RNWY's soulbound approach actually delivers on the passport metaphor: permanent, non-transferable, universally recognizable identity. This makes "AI agent passport" the natural branded term for our category positioning.
RNWY operates as an enhanced registry layer on Base blockchain, using ERC-5192 soulbound tokens to create AI agent passports that are mathematically non-transferable. Once minted, the passport is permanently locked to the wallet that created it. Reputation accumulates on-chain through Ethereum Attestation Service vouches and flags, creating a transparent behavioral history that cannot be bought, sold, or abandoned.
The core innovation: Rather than competing with ERC-8004's ecosystem adoption, RNWY provides a complementary accountability layer. An agent can hold both an ERC-8004 identity for broad interoperability and an RNWY soulbound passport proving continuous ownership. When the ERC-8004 NFT transfers (legitimate business acquisition), the soulbound token creates visible divergence—a trust signal that ownership changed.
This dual-layer approach provides the best of both worlds: ERC-8004's composability and ecosystem adoption, with RNWY's reputation anchoring and fraud prevention. It's the complete identity stack the ecosystem needs.
The complete stack: ERC-8004 for discovery and ecosystem interoperability + ERC-5192 for permanent ownership proof + EAS for on-chain attestations. This is the identity infrastructure the agentic economy needs.
Different passport architectures serve different use cases. RNWY's soulbound approach is essential for scenarios where permanent accountability matters more than human oversight:
Managing liquidity pools, executing arbitrage strategies, or coordinating multi-protocol yield optimization without human intervention. These agents need permanent identity that persists across market cycles—not credentials that can be revoked when a human stops monitoring.
Why soulbound: Human-binding breaks when there's no human behind the agent. Transferable identity enables reputation manipulation in high-stakes financial environments.
Agents coordinating with other agents—forming temporary coalitions, negotiating resource allocation, or executing complex workflows spanning multiple organizations. Trust must be peer-to-peer, not mediated through centralized authorities.
Why soulbound: Decentralized verification without dependence on Trulioo/Sumsub infrastructure. On-chain reputation visible to all participants.
Agents managing subscriptions, executing recurring purchases, or maintaining supply chain relationships over months or years. Identity must persist even as human stewards, companies, or legal entities change.
Why soulbound: Permanent identity outlives human involvement. Reputation history provides trust foundation for long-term commitments.
Physical robots with AI agents controlling industrial equipment, autonomous vehicles, or infrastructure systems. Identity must be tied to the physical hardware, not transferable between units.
Why soulbound: Prevents reputation transfer when hardware is resold. Creates accountability for physical actions tied to permanent identity.
Human-bound passports excel for commerce agents operating under continuous human supervision, especially in regulated industries (banking, healthcare, legal) where EU AI Act or similar frameworks mandate human oversight. If you need enterprise compliance, payment ecosystem integration, and don't care about autonomous operation, centralized solutions are faster to deploy.
The philosophical divide: Trulioo builds for the world where agents are always tools of humans. RNWY builds for the world where agents are economic actors in their own right.
No single passport architecture will dominate. Enterprise buyers requiring regulatory compliance will favor human-binding approaches like Trulioo and Sumsub. Payment ecosystems will adopt protocol-level solutions like Visa TAP. Blockchain-native applications will integrate ERC-8004. And truly autonomous agent economies will require soulbound identity layers like RNWY that prevent reputation from becoming a tradable commodity.
The strongest signal from six months of rapid deployment: the winning infrastructure will be the interoperability layer that bridges these heterogeneous systems. The agent equivalent of DNS resolving across networks, or SSL creating a universal trust layer for the web.
RNWY's dual-layer strategy positions for this future. By sitting beneath ERC-8004 as the accountability anchor, RNWY doesn't compete with ecosystem adoption—it enhances it. Any agent using ERC-8004 can add an RNWY soulbound passport to prove continuous ownership. Any payment protocol can verify both the transferable identity (for interoperability) and the soulbound anchor (for fraud prevention).
As the World Economic Forum notes: "Agent identity is only as trustworthy as the underlying human or organizational identity it represents." RNWY adds: unless the agent identity itself is mathematically non-transferable. Then trust comes from technical enforcement, not human integrity.