Contact Us

Suggested region and language based on your location

    Your current region and language

    People with phone in hand.
    Regulation

    EU Artificial Intelligence (AI) Act

    Prepare for AI regulation in the EU.

    Understand the EU AI Act and what organizations need to do next

    The EU AI Act is in force, with requirements applying in stages. Learn how to identify your role, classify AI systems, and plan practical next steps.

    The Act aims to protect health, safety, and fundamental rights while supporting responsible AI innovation in Europe. Its harmonized, risk-based framework helps organizations identify obligations, improve transparency, and show that AI systems are safe, traceable, and non-discriminatory.

    Who needs to comply with the EU AI Act?

    Organizations may need to comply if they develop, provide, deploy, import, or distribute AI systems used in the EU, or use AI outputs in the EU:

    • Icon description

      Providers placing AI systems or general-purpose models on the EU market or putting them into service.

    • Icon description

      Providers outside the EU where an AI system’s output is used in the EU.

    • Icon description

      Deployers of AI systems located in the EU.

    • Icon description

      Importers and distributors of AI systems.

    • Icon description

      Product manufacturers placing AI systems on the EU market under their name or trademark.

    • Icon description

      Authorized representatives of providers established outside the EU.

    FAQs

    AI Act Frequently Asked Questions

    AI graphics off person with gadgets.

    When does the EU AI Act apply?

    The EU AI Act entered into force on 1 August 2024 and applies progressively, with different requirements taking effect at different stages.

    Prohibited AI practices and AI literacy obligations have applied since 2 February 2025. Other obligations apply in phases, with key dates through 2 August 2027 and some transitional provisions continuing to 2030.

    EU AI Act

    A risk-based approach to AI regulation

    The AI Act classifies AI systems by risk, helping keep obligations proportionate to their potential impact on health, safety, and fundamental rights:

    • Unacceptable Risk
    • High Risk
    • Limited Risk
    • Minimal Risk
    Unacceptable Risk

    Prohibited AI practices include certain uses of:

    1. Harmful manipulation or deception
    2. Exploiting vulnerable groups
    3. Social scoring
    4. Biometric categorization
    5. Real-time remote biometric identification
    6. Predictive policing
    7. Emotion recognition in specific contexts
    8. Untargeted facial image scraping"
    High Risk

    The AI Act identifies two broad categories of high-risk AI systems:

    1. AI systems used as products or safety components covered by EU legislation that require third-party conformity assessment.
    2. Standalone AI systems that may significantly affect rights, safety, or access to essential services."
    Limited Risk

    AI systems with specific transparency obligations

    Limited risk applies where transparency is needed, such as when people interact with AI systems or AI-generated content. The aim is to help people understand when and how AI is used.

    Minimal Risk

    AI systems posing minimal or no risk

    Minimal-risk AI systems can generally be developed and used in the EU without additional legal obligations under the Act. Voluntary codes of conduct can support responsible AI use.

    Build AI governance and EU AI Act readiness with BSI

    Prepare for the EU AI Act by strengthening AI governance with BSI’s foundation framework, training, testing, and certification support.

    Contact Us

    Ready to prepare for the EU AI Act?

    From AI foundations and standards to compliance support, BSI can help you build trust and contribute to a safer AI future.

    Get in touch - Philippines