identities.ai

/ Security

Security by
cryptographic design.

As AI agents cross the transaction horizon, the trust model of the internet must evolve from human-centric to agent-aware. Ratify is the identity substrate for that transition.

Verification, not detection

Detection is a probabilistic guess that degrades as AI improves. Ratify is a deterministic cryptographic check. An agent is either authorized or it isn't—not because a model guessed, but because the math says so.

Hybrid Post-Quantum Crypto

Every Ratify signature is a hybrid of Ed25519 (classical) and ML-DSA-65 (NIST FIPS 204 post-quantum). This 'harvest now, decrypt later' defense ensures that authorization bundles remain secure even against future quantum adversaries.

Zero-Trust at the Edge

Ratify verifiers run at the transaction horizon—voice gateways, API edges, on-device for Physical AI. Verification is stateless and doesn't require a constant connection to a central authority, minimizing the attack surface.

Stateless & Scalable

The protocol is designed for the high-volume agent economy. Verification happens in under a millisecond. Because the proof carries its own authorization context, there's no database lookup required for the core cryptographic check.

The Threat Model

/ REPLAY ATTACKS

Every Ratify ProofBundle requires a fresh, signed challenge from the verifier. A bundle used five minutes ago—or even five seconds ago—is cryptographically invalid for a new session.

/ SCOPE CREEP

Agents can only sub-delegate a subset of the rights they hold. The effective scope of a delegation chain is the intersection of all links, preventing any agent from granting more authority than it possesses.

/ KEY COMPROMISE

Agent keys are short-lived. In the event of a breach, a principal can revoke a delegation certificate instantly through the Ratify Revocation Registry, neutralizing the compromised agent without affecting the rest of the ecosystem.

/ Get started

Ready to stop trusting AI
and start verifying it?

We are currently in private alpha, working with select design partners to establish the identity layer for the agent economy.

1

Enterprises

Secure your agent interactions across voice, video, and API.

2

Platforms

Provide verifiable identity to every agent on your platform.

Or email hello@identities.ai