The New EU AI Legislation Has Been Passed – Here's What Applies
Europe takes the lead with its groundbreaking AI legislation, passed on Tuesday, setting a global standard for handling artificial intelligence in business and everyday life. What does this mean in practice, and when does it actually take effect?
Next month, Europe's groundbreaking rules on artificial intelligence take effect, after EU countries approved a political agreement reached in December on Tuesday, May 21. This legislation could become a global standard for a technology used both in business and everyday life.
The EU's Comprehensive AI Legislation
The EU's AI Act is more comprehensive than the US's softer voluntary compliance approach, while China's strategy aims to maintain social stability and state control. The decision by EU member states came two months after the European Parliament backed the AI legislation drafted by the European Commission in 2021, after making several key changes.
Steep Fines for AI Violations
The use of AI technology can carry strict consequences for those who violate the rules and regulations. Depending on the type of violation, companies could face fines ranging from €7.5 million (about 87 million SEK) or 1.5% of turnover, up to 406 million SEK or 7% of global turnover.
Growing Concern Over AI
Concerns over AI's contribution to disinformation, fake news, and protected material have grown globally in recent months, alongside the increasing popularity of generative AI systems like Microsoft-backed OpenAI's ChatGPT and Google's chatbot Gemini.
An Emphasis on Trust and Transparency
"This groundbreaking law, the first of its kind in the world, addresses a global technological challenge that also creates opportunities for our societies and economies," Belgium's digitalization minister Mathieu Michel said in a statement. "With the AI Act, Europe emphasizes the importance of trust, transparency, and accountability in handling new technology while ensuring that this rapidly evolving technology can thrive and foster European innovation."
Strict Requirements for AI Systems
The AI Act introduces strict transparency requirements for high-risk AI systems, while such requirements for general-purpose AI models will be lighter. It restricts authorities' use of real-time biometric surveillance in public spaces to cases involving certain crimes, preventing terrorist attacks, and searching for individuals suspected of the most serious crimes.
Global Impact
The new legislation will have an impact beyond the 27-country union, according to Patrick van Eecke of law firm Cooley. "The law will have global reach. Companies outside the EU that use EU customer data in their AI platforms will need to comply with the rules. Other countries and regions will likely use the AI Act as a template, just as they did with GDPR," he said, referring to the EU's privacy rules.
Implementation Timeline
The new legislation takes effect in 2026, but the bans on using artificial intelligence for social scoring, predictive policing, and untargeted scraping of facial images from the internet or CCTV footage take effect six months after the new rules come into force. Requirements for general-purpose AI models take effect after 12 months, and rules for AI systems integrated into regulated products after 36 months.