On February 2, the EU enforced a ban on AI systems deemed to pose an unacceptable risk, including tools designed for manipulation, deception, or social scoring.

In March 2024, the European Parliament approved a comprehensive AI regulation law, which came into effect on August 1.

The legislation classifies AI systems into risk categories:

Low risk: AI-powered spam filters or video games remain unregulated.
Limited risk: Chatbots and AI-generated content tools must meet transparency requirements.
High risk: AI used by law enforcement, government agencies, or for biometric identification and emotion recognition.
Unacceptable risk: Systems that deceive, manipulate, or assess individuals based on their behavior or personal traits.

As of February 2, companies must discontinue AI classified under the last category or face fines of up to €35 million or 7% of their annual revenue.

Further restrictions will be phased in:

August 2, 2025: Rules for general-purpose AI systems take effect.
August 2, 2026: Stricter regulations for “high-risk” AI, covering biometric data, critical infrastructure, and education.

In October 2023, Ukraine’s Ministry of Digital Transformation introduced an AI regulation roadmap to help businesses prepare for EU compliance.

Previously, Coinbase CEO Brian Armstrong spoke out against AI regulation, advocating for decentralization of the industry.