EU AI Act: Complete Enforcement Timeline and Deadlines
The EU AI Act implements the world's first comprehensive AI regulation through a phased enforcement timeline stretching from 2025 through 2027. Prohibited practices including social scoring and most real-time biometric identification took effect in February 2025. General purpose AI model obligations started in August 2025. High-risk system requirements become fully applicable in August 2026. Maximum penalties reach 35 million euros or 7 percent of global annual turnover.
Phase 1: Prohibited Practices (February 2025)
The first enforcement phase bans AI practices deemed unacceptable: social scoring by governments, real-time remote biometric identification for law enforcement except in narrowly defined emergencies, emotion recognition in workplaces and educational institutions, AI systems exploiting vulnerabilities of specific groups, and subliminal manipulation techniques. Violations carry maximum fines of 35 million euros or 7 percent of global annual turnover, whichever is higher.
Phase 2: General Purpose AI (August 2025)
Providers of general purpose AI models including GPT-4, Claude, Gemini, and Llama must comply with transparency requirements: publishing training data summaries, complying with EU copyright law, and providing technical documentation. Models with systemic risk, determined by compute thresholds exceeding 10^25 FLOPS, face additional obligations including adversarial testing, incident monitoring and reporting, and cybersecurity measures.
Phase 3: High-Risk Systems (August 2026)
AI systems classified as high-risk must comply with comprehensive requirements for risk management, data governance, technical documentation, record-keeping, transparency, human oversight, accuracy, robustness, and cybersecurity. This applies to AI in employment recruitment, creditworthiness assessment, law enforcement, migration and asylum, and democratic processes. Conformity assessment procedures must be completed before market placement.
Key Findings
- Maximum penalties reach 35 million euros or 7 percent of global annual turnover for prohibited practice violations
- General purpose AI models exceeding 10^25 FLOPS training compute face systemic risk obligations
- High-risk system requirements become fully applicable in August 2026 with mandatory conformity assessment
Timeline
AI Act enters into force
Prohibited practices ban
GPAI obligations
High-risk requirements
Annex III high-risk extensions