EU AI Act
Is your AI system high-risk? Check by use case
The AI Act classifies systems by intended purpose. These pages walk through the most common purposes, the Annex III or Article 50 entry that applies, what is banned, and the documents each one needs.
High-risk
Recruitment & CV screening
High-risk (Annex III, point 4(a))
Read the rules
High-risk
Employee monitoring & performance
High-risk (Annex III, point 4(b)); partly prohibited (Art. 5(1)(f))
Read the rules
High-risk
Credit scoring & lending
High-risk (Annex III, point 5(b))
Read the rules
Depends on use
Insurance risk assessment & pricing
High-risk for life and health insurance (Annex III, point 5(c)); other lines usually minimal
Read the rules
Limited risk
Customer service chatbots
Limited risk (Article 50 transparency), unless used for an Annex III purpose
Read the rules
High-risk
Medical diagnosis & clinical decision support
High-risk (Art. 6(1) via the MDR/IVDR, or Annex III(5)(d) and health-related points)
Read the rules
High-risk
Education: admissions, grading & proctoring
High-risk (Annex III, point 3); partly prohibited (Art. 5(1)(f))
Read the rules
Depends on use
Biometric identification & categorisation
High-risk (Annex III, point 1); prohibited in several configurations (Art. 5); verification-only is out of scope
Read the rules
Minimal risk
Recommendation engines & personalisation
Minimal risk in most configurations; Article 50 and Article 5 limits; Annex III when used for listed purposes
Read the rules
Depends on use
Fraud detection & AML
Minimal risk for financial fraud detection (Annex III(5)(b) exception); high-risk where used for law-enforcement purposes (Annex III(6))
Read the rules
Limited risk
Generative AI content & deepfakes
Limited risk (Article 50(2) and 50(4) transparency); GPAI model duties for model providers
Read the rules
High-risk
Public benefits & essential services eligibility
High-risk (Annex III, point 5(a))
Read the rules
High-risk
Critical infrastructure safety components
High-risk (Annex III, point 2)
Read the rules
Depends on use
Predictive policing & law enforcement AI
High-risk (Annex III, point 6); a narrow profiling-only configuration is prohibited (Art. 5(1)(d))
Read the rules
High-risk
Migration, asylum & border control AI
High-risk (Annex III, point 7)
Read the rules
High-risk
Legal tech & judicial assistance AI
High-risk (Annex III, point 8(a)); a narrow administrative carve-out exists
Read the rules