ENNL
Free consultation
Home / Which EU AI Act risk tier is your AI in?

Which EU AI Act risk tier is your AI in?

The EU AI Act sorts every AI system into four risk tiers — and each tier carries its own obligations and deadlines. Answer 4 short questions and instantly see where your system lands, what you must put in place and by when. Free, no email required, not legal advice.

  1. 1Prohibited practicesDoes your system fall under Art. 5?
  2. 2High-risk usesAnnex III or regulated products.
  3. 3TransparencyChatbot, generative AI or emotion recognition.
  4. 4Your tier + deadlineObligations and the compliance date.
Free · no email · ~1 min

EU AI Act risk checker

Answer 4 short questions and see which EU AI Act risk tier your AI system falls into — with your obligations and the compliance deadline.

4 questions~1 minuteNo email needed

This tool gives an indication based on the EU AI Act (Regulation (EU) 2024/1689) and is not legal advice. Final classification depends on your specific system and context. Always have a judgement verified — for example with our AI audit. · Source: EU AI Act (Regulation 2024/1689)

EU AI Act — frequently asked questions

What are the EU AI Act risk tiers?

Four: unacceptable risk (banned), high risk (heaviest obligations), limited risk (transparency duty) and minimal risk (no obligations). Separate rules also apply to general-purpose AI models (GPAI).

When does the EU AI Act take effect?

The regulation has been in force since 1 August 2024. Prohibited practices apply from 2 February 2025, GPAI rules from 2 August 2025, most high-risk obligations from 2 August 2026, and product-related rules (Annex I) from 2 August 2027.

Does the EU AI Act apply to SMEs?

Yes. It applies to any provider or user of AI in the EU regardless of company size. Small companies get lighter burdens and support (such as regulatory sandboxes), and most SME use cases fall into the minimal-risk tier.

Is my chatbot a high-risk system?

Usually not. A customer-service chatbot typically falls under limited risk: you must tell users they are talking to AI. High risk only arises in uses such as recruitment, credit scoring or critical infrastructure.

What are the penalties for non-compliance?

Up to €35 million or 7% of global annual turnover for prohibited practices, and up to €15 million or 3% for other breaches — whichever is higher. SMEs face proportionally lower caps.

What is a general-purpose AI model (GPAI)?

A broadly usable model (such as a foundation or language model) that you train and provide yourself. Providers have their own obligations: technical documentation, a copyright policy and a training-data summary — with extra requirements at systemic risk.