Cryptographic trust infrastructure for the agentic era. Verifiable identity, tamper-proof provenance, and granular access control — without vendor lock-in or blockchain overhead.
Intelligence is no longer the bottleneck. Security is. When an AI agent executes a payroll run, approves a transaction, or files a tax document — who verifies it actually did what it claimed?
"Security will become THE bottleneck for effectiveness and usefulness of AI agents as a whole."
— Lex Fridman, February 2026No verifiable record of what data was used, which models processed it, or what outputs were produced.
When an agent acts on your behalf, there is no cryptographic proof of who authorized what — or which system is responsible.
Current agent frameworks offer all-or-nothing access. No context-aware, operation-level policies.
Users leave the AI interface to manually verify every write action in the source application. The trust gap is structural.
An open cryptographic protocol that gives AI agents verifiable trust — so every operation is signed, every action is traceable, and every claim is independently verifiable.
Decentralized identity (DID) for every agent and user. No central authority. You own your keys.
Tamper-proof AquaTrees link every operation in a signed hash chain. Full audit trail, offline-verifiable.
Operation-level, context-aware access control. Not all-or-nothing — fine-grained policies per action.
Any third party can verify the entire chain without trusting the agent, the platform, or the provider.
Full cryptographic guarantees without chain overhead. Use a blockchain when you want to, not because you have to.
Works with any AI model, any API, any infrastructure. Not tied to a single provider's ecosystem.
Verify offline, across organizations, without network access. Evidence travels with the data.
Designed for EU AI Act (Art. 10/12/13), NIST AI RMF, SOX, and SOC 2 compliance out of the box.
We built a payroll agent that operates against a major US payroll provider's public API — not a toy demo, but a production-grade integration with full cryptographic signing.
Alfred handles employee onboarding, contractor setup, payroll runs, tax filings, and benefits management — all through natural language, all cryptographically signed and independently verifiable.
Every API operation flows through a template-driven trust chain: user authorization, agent execution, and independent notarization. The result is a complete, tamper-evident audit trail of everything the agent did and why.
We believe in transparency. Here is exactly where the project stands.
30 minutes free / sandbox data only / see the trust chain for yourself
Enter your email to start a session.