Introducing Privacy Center.
A unified platform where businesses can manage all their regulatory and transparency needs in one place — policies, DSRs, certifications, and representations in a single public-facing trust hub.
Your AI systems classified by risk, the Annex IV technical documentation held against each one, serious incidents reported under Article 73, and a named Article 22 authorised representative. For providers placing AI on the EU market from outside it.
Which of your systems are high risk under Annex III, which carry transparency duties under Article 50, and which carry nothing at all — written down, with the reasoning, so the answer survives being questioned.
Annex IV technical documentation per system: the design, the data, the testing, the oversight measures and the accuracy and robustness claims, kept against the system rather than in a folder.
A named Article 22 authorised representative holding the mandate and the documentation for the authorities. The part of this you cannot do yourself from outside the EU.
Including the systems that turn out to carry no obligation — because the question an authority asks is what you have, and "we checked and it is minimal risk" is only an answer if it was written down at the time.
Annex IV documentation for each high-risk system, built from what your engineers already know and kept with the system so a model change and a document change are the same event.
The assessment procedure that fits the system, the EU declaration of conformity, the CE marking where it applies, and the registration in the EU database before it goes on the market.
Serious incidents under Article 73 — to the market surveillance authority of the member state where it happened, inside the deadline the severity sets.
// System · Annex III 5(b) Classified High risk · creditworthiness Provider Acme Corp (non-EU) Rep World Presence j.d.o.o. Annex IV Pack v3.1 · 18 Apr 2026 Assessment Internal control (Annex VI) Declaration Signed · EU DoC Database Registered · EU-HR-00417 Oversight Art. 14 measures documented Review Annual · next 18 Apr 2027
The plan for watching how the system behaves once it is out, and the record of what that watching found — which is what turns a one-off assessment into something an authority will accept.
Telling people they are talking to an AI, marking synthetic content, and the disclosures a limited-risk system owes even when nothing else applies.
The oversight measures designed into the system, who exercises them, and the competence they need — documented, because Article 14 asks for effectiveness rather than the existence of a button.
For GPAI providers: the technical documentation, the training-data summary, the copyright policy, and the systemic-risk duties above the compute threshold.
A provider established outside the Union that places a high-risk AI system on the EU market must appoint an authorised representative inside it, by written mandate, before the system is placed. We become yours — and we keep the Annex IV documentation and the declaration of conformity available for ten years, which is what the mandate actually obliges.
The representative can terminate the mandate if it believes the provider is acting against its obligations, and must tell the authority when it does. That is the law, and it is worth knowing before you appoint anyone.
Learn more about the Article 22 mandate →AI Act authorised representative under Article 22. Catalogue every minimal-risk system free.
$0
at 1 high-risk AI systems · 1 included
$0
at 0 limited-risk AI systems · None included
Serious-incident report
Article 73, filings included.
$490
per event · 1 a year included on Pro
In the fee
This is on top of your platform plan, which starts at $0. See every price →
Training on or inferring from personal data is processing, with everything that follows.
If a model decides what users see on your platform, Article 13 and the DSA arrive with it.
The public side: your AI transparency disclosures, published and versioned.
The representative is part of it. AI Governance is the product: the register, the risk classification, the Annex IV files, conformity and the Article 22 mandate together. If all you need is the mandate itself, that is the piece described on the AI Act Representative page — but it is bought as AI Governance either way.
Then you are probably a deployer rather than a provider, and the obligations are lighter — but not nothing, and they change the moment you put your name on a system or substantially modify one. The register is where that line gets drawn and recorded.
Annex III lists the areas: biometrics, critical infrastructure, education, employment, essential services and creditworthiness, law enforcement, migration, and the administration of justice. Being in one of those areas is the test, not how advanced the model is.
One high-risk system is included; further high-risk systems are priced on the slider and limited-risk ones are a flat rate each. The minimal-risk inventory is free, however long it is, because cataloguing what carries no obligation should not cost you anything.
The mandate ends on 30 days' notice and the authorities are told. Until it does you stay covered: a high-risk system on the EU market without a current authorised representative is a system that should not be on the market.
Thirty minutes to classify what you have and find the ones that carry obligations. A named representative on record the moment you sign.
