AI Model Wiki

    Skip to main content
    22 min
    Lesson 1 of 4

    Risk Classification Framework

    From the course EU AI Act Compliance

    Master the EU AI Act's risk-based classification system.

    Four Risk Tiers

    Unacceptable Risk (BANNED)

    Prohibited AI systems:

    • Social scoring by governments
    • Real-time remote biometric ID in public (with limited exceptions)
    • Subliminal manipulation causing harm
    • Exploitation of vulnerable groups
    • Emotion inference in workplace/education (some uses)

    High Risk (STRICT REQUIREMENTS)

    Annex III High-Risk Categories:

    1. Biometric identification and categorization
    2. Critical infrastructure management
    3. Education and vocational training
    4. Employment and worker management
    5. Access to essential services
    6. Law enforcement
    7. Migration and border control
    8. Justice administration

    Limited Risk (TRANSPARENCY)

    • Chatbots and conversational AI
    • Emotion recognition systems
    • Biometric categorization
    • Deepfake generators

    Requirement: Users must be informed they're interacting with AI

    Minimal Risk (NO RESTRICTIONS)

    • Spam filters
    • AI-enhanced video games
    • Inventory management

    Classification Decision Tree

    Is it on the prohibited list?
    ├── YES → BANNED
    └── NO → Is it on Annex III?
             ├── YES → HIGH RISK
             └── NO → Does it generate/manipulate content?
                      ├── YES → LIMITED RISK
                      └── NO → MINIMAL RISK
    

    Key Takeaways

    • •EU AI Act uses four risk tiers: unacceptable, high, limited, and minimal
    • •High-risk AI includes HR systems, credit scoring, and law enforcement tools
    • •Limited-risk AI (chatbots, deepfakes) requires transparency disclosures
    • •Most AI systems fall into minimal risk with no specific requirements
    0 of 4 lessons marked complete

    Progress is kept in this browser only.