Startups building payment and identity rails for AI agents transacting autonomously

1. Skyfire — Builds “know-your-agent” identity, wallets, and payment rails so AI agents can transact with merchants and service providers. Why it fits: One of the clearest pure-play fits; early-stage company with a public focus on agent wallets, agent identity, and autonomous payments.
2. Payman — Provides APIs and stablecoin-based accounts for paying people and services on behalf of AI agents. Why it fits: Seed-stage, agent-native payments company; its core product is explicitly designed for agents that need to compensate humans or purchase services autonomously.
3. Nevermined — Offers an AI-native payments and billing protocol based on prepaid credits, programmable budgets, and machine-to-machine settlement. Why it fits: Early company with a clear thesis around AI-to-AI commerce and monetization rather than conventional human checkout.
4. TollBit — Builds a monetization and access layer allowing AI agents to pay publishers and other content providers for machine-readable content. Why it fits: Strong wedge into agentic commerce; its product addresses one of the first practical autonomous-payment use cases—agents purchasing data and content.
5. Kite AI — Developing blockchain infrastructure for identity, payments, and attestations between AI agents and applications. Why it fits: Very early, explicitly agent-focused infrastructure project targeting machine-to-machine transactions rather than consumer crypto payments.
6. Crossmint — Provides APIs for embedded wallets, stablecoin payments, tokenized credentials, and programmable accounts. Why it fits: Venture-backed infrastructure company increasingly positioned for AI-agent wallets and commerce; its APIs can give agents controlled accounts without requiring users to manage private keys.
7. Turnkey — Offers programmable wallet infrastructure, key management, and transaction signing for applications that need secure automated payments. Why it fits: Early infrastructure provider with a strong technical fit for agent wallets: policy controls, scoped signing, and automated execution are prerequisites for safe autonomous transactions.
8. Privy — Provides embedded wallets and identity/authentication infrastructure for applications, including software that needs users or agents to transact on-chain. Why it fits: Venture-backed and product-proven in wallet infrastructure; its embedded-account model is applicable to agent identities and delegated payment permissions, even though it is not exclusively agent-focused.
9. Dynamic — Builds authentication, wallet, and account-abstraction infrastructure for applications interacting with digital assets. Why it fits: Early-stage developer platform whose support for programmable wallets and multiple authentication methods makes it a plausible identity and payment layer for autonomous agents.
10. Trinsic — Develops digital-identity and verifiable-credential infrastructure, including APIs for issuing and verifying portable identities. Why it fits: Identity-side fit rather than a pure payments play; verifiable credentials could let agents prove authorization, reputation, ownership, or organizational affiliation when transacting across services.

White-space gaps noticed

• Interoperable agent reputation and liability: The market has wallet and payment primitives, but few neutral standards for an agent’s reputation, spending history, delegated authority, dispute record, or accountable human/company principal.
• Agent-native merchant acceptance and compliance: There is a shortage of infrastructure that lets ordinary merchants accept autonomous agents safely—combining agent authentication, policy-based authorization, sanctions/KYC checks, receipts, refunds, and dispute handling in one rail.