EU AI Act Enforcement Begins With First Wave of Compliance Deadlines

The European Union’s landmark AI Act has officially entered enforcement, marking the world’s first comprehensive regulatory framework for artificial intelligence. Companies operating in the EU must now comply with the first wave of requirements covering prohibited AI practices and high-risk system classifications.

The regulations ban AI systems that manipulate human behavior, exploit vulnerabilities, or enable social scoring by governments. High-risk applications in healthcare, education, employment, and law enforcement face mandatory conformity assessments.

What Companies Must Do Now

Organizations deploying AI systems classified as high-risk must complete technical documentation, implement risk management systems, and ensure human oversight mechanisms are in place. Data governance requirements mandate that training datasets be documented and tested for bias.

“This is the moment AI governance moves from theory to practice,” said Thierry Breton, European Commissioner for Internal Market. “Companies have had two years to prepare. The time for compliance is now.”

Industry Pushback

Tech industry groups have raised concerns about implementation costs and regulatory uncertainty. The European AI Business Alliance estimates compliance costs for large enterprises at between $5 million and $30 million, depending on the number of high-risk systems deployed.

Several US-based companies have restricted AI feature availability in Europe while they work through compliance requirements, drawing criticism from EU officials who see the moves as unnecessarily cautious.

Global Ripple Effects

The EU’s approach is influencing AI regulation worldwide. Canada, Brazil, and Japan have introduced legislation modeled on key provisions of the AI Act, while the United States continues to rely primarily on sector-specific guidelines and executive orders.

]]>