Europe’s landmark AI laws will take effect next month after EU countries endorsed a political deal in December, setting a potential global standard for AI regulation. China’s state-control-focused approach and the US’s voluntary compliance approach are less extensive than the EU’s AI Act.
The vote comes after EU MPs approved the AI legislation, which the European Commission first proposed in 2021 and included significant changes. As generative AI systems like ChatGPT and Google’s Gemini have become more popular, worries about AI-driven disinformation and copyright violations have grown.
For high-risk AI systems, the Act imposes stringent transparency standards; for general-purpose AI, the rules are less stringent. It limits the use of real-time biometric surveillance by the government to combating major crimes and preventing terrorism. According to Cooley Law Firm’s Patrick van Eecke, the legislation will affect businesses worldwide and may lead to the adoption of comparable frameworks in other areas.
The Act prohibits the use of AI in social scoring, predictive policing, and untargeted face image scraping, and it goes into effect in 2026. Depending on the type of violation, fines can reach up to 35 million euros, or 7% of worldwide turnover, or 7.5 million euros, or 1.5% of turnover.
