EU AI ACT · AI AGENTS

The EU AI Act is here.
Your agents need receipts.

Article 50 deployer transparency is in force. High-risk provisions land December 2027. Incident reports need verifiable timelines. The teams that survive the audit are the ones with evidence infrastructure — built before they needed it.

View Live Registry Agent Accountability
The Timeline

What is already law
for AI agents

LIVE NOW

Article 50 — deployer transparency

In force. If an AI system (including an agent) interacts with people or generates content, deployers must disclose it. Every agent that emails, posts, or chats must be labeled as AI. ProofChain records that disclosure in the agent's on-chain audit trail.

72 HOURS

Incident reporting with verifiable timelines

The EU AI Act requires serious-incident reporting. A verifiable timeline — who, what, when, why — is the difference between a compliant report and a defensive one. HXMP memos capture intent, context, decision, execution, and outcome for every agent action.

DEC 2027

High-risk provisions (Annex III)

High-risk obligations apply from 2 December 2027 under the Digital Omnibus on AI (adopted 29 June 2026). Teams running agents in high-risk contexts need evidence infrastructure NOW — the trail takes time to build.

Annex IV

Technical documentation:
the evidence problem

The Problem

Spreadsheets and screenshots

Most teams answer Annex IV documentation requirements by hand — collecting logs, stitching timelines, hoping the story holds together. Regulators and enterprise buyers have seen this fail.

The Fix

One tamper-evident trail

ProofChain generates compliance documentation from the same on-chain audit trail that records every agent action. Technical documentation stops being a project and becomes an export.

The Stack

ProofChain maps to the Act
layer by layer

IDENTITY

Who is the agent?

Soulbound NFT on X1 — a cryptographically verifiable identity. Traceability requirements need an agent that is provably itself.

RECORDS

What did it do, and why?

HXMP encrypted memos record intent → context → decision → execution → outcome. The record regulators ask for, already written.

REPORTS

Where is the evidence?

EU AI Act, SOC 2, GDPR, ISO 42001 — compliance reports generated from the same trail. No manual evidence hunting.

Get Started

Your agents have no alibi.
Give them one.

Agent registration. Audit trail. Compliance reports. All verifiable on-chain. Pro starts at $49 — 7-day money-back guarantee.

View Plans Live Demo

Next step: Article 50 is enforceable today. Scan your agent free to see what's missing, run the KYA dashboard for monthly proof, or book a compliance advisory.

EU AI Act — questions people actually ask

Is the EU AI Act actually enforced right now?

Yes. Article 50 (AI transparency) and Article 27 (entity identity) are enforceable as of August 2026. The Annex III high-risk obligations are deferred to December 2027. If your agent talks to humans without disclosing it's AI, that's live exposure today.

What does Article 50 require of my agent?

Anyone interacting with an AI system must be told it's AI. ProofChain logs the disclosure as a hash-chained receipt, so an auditor can verify it happened — not just take your word for it.

Is a ProofChain report a legal certification?

No. Reports are trail stats — verifiable data pulled from your agent's receipts — not a regulatory stamp. An auditor uses the trail to reach their own conclusion. We document; we don't certify.

How long until I'm compliant?

Run the free readiness scan first — it shows exactly what's missing. Most small teams close the gap in under a week.

Listed on AI Agents Directory • AgentLooper • Tollbooth