Free EU AI Act Risk Assessment Tool
Our free EU AI Act risk assessment tool tells you which risk tier an AI system falls into (prohibited, high-risk, limited, or minimal) in about 3 minutes, with the reasoning written out and the obligation list that follows from it. No account needed to run the check. We built it because every compliance conversation starts with the same question: what tier are we?
What the tool actually checks
The assessment walks the same ladder Article 6 and Annex III describe:
- Use-case screen. What the system does and who it affects, matched against the Article 5 prohibited practices and the eight Annex III categories (including employment, credit, education, and essential services).
- Role screen. Whether you act as provider or deployer for this system, because the two roles carry different obligation sets.
- Interaction screen. Whether the system talks to people or generates content, which triggers Article 50 transparency duties even outside high-risk.
A worked example
Feed it "an internal chatbot that screens incoming job applications and ranks candidates" and you get: high-risk (Annex III employment category), role mapping, and the obligation list with article numbers, the same 12 rows we publish in the high-risk checklist. Feed it "a chatbot that answers product questions" and you get limited risk with the Article 50 disclosure duties and example wording.
Why we give the wedge away
Classification is the decision everything else hangs on, and it should not cost money to find out where you stand. If your tier turns out to carry real obligations, that is where AI Comply HQ earns its keep: a guided interview that captures your system facts once, then generates the documents your tier requires, with versioning and an audit trail.
FAQ
Is the result legal advice?
No. It is a structured application of the published classification rules, with the reasoning shown so your counsel can verify it quickly.
What happens to the answers we enter?
We ask for an email to send the snapshot, with explicit consent, and you can opt out anytime. We do not auto-dial anyone and we do not resell lists. Ever.
What if the rules change?
The classification logic tracks the Regulation as enacted. When Brussels amends dates or categories, the tool updates and your snapshot says which version of the rules it used.
Start the free risk assessment
Sources
This page is informational content, not legal advice. Talk to a qualified lawyer about your specific situation.