EU’s AI Regulation Agreement Makes History
This post is also available in: 
עברית (Hebrew)
The EU has reached a historic agreement on the world’s first comprehensive AI regulation, legislation dubbed the “AI Act” that paves the way for ethical and responsible AI development while minimizing the potential risks.
According to Interesting Engineering, the deal establishes a clear framework for regulating AI and categorizes AI systems into three tiers based on their risk profile: minimal, high, and unacceptable. AI systems posing minimal risk will be subject to minimal regulations, while those deemed high-risk (like medical devices and certain law enforcement technologies) will face strict requirements. It also bans certain AI practices deemed “unacceptable”, including manipulative social scoring systems and emotion recognition technologies used in workplaces.
President of the European Commission Ursula von der Leyen said: “Artificial intelligence is already changing our everyday lives. And this is just the beginning. Used wisely and widely, AI promises huge benefits to our economy and society. Therefore, I very much welcome today’s political agreement by the European Parliament and the Council on the Artificial Intelligence Act. The EU’s AI Act is the first-ever comprehensive legal framework on Artificial Intelligence worldwide. So, this is a historic moment. The AI Act transposes European values to a new era. By focusing regulation on identifiable risks, today’s agreement will foster responsible innovation in Europe.”
This revolutionary legislation represents a significant step forward in the global race to regulate AI, setting a precedent for other nations to follow and will likely shape the future of AI development worldwide.
Market surveillance authorities will oversee new regulations for AI models nationally, while the European Commission will establish an AI Office to ensure coordination across the EU. The AI Office will enforce the rules for general-purpose AI models and set a global standard assisted by an independent scientific panel of experts.
Following are several key takeaways from the new AI Act, according to Interesting Engineering:
- Risk-based approach: AI systems are categorized based on their potential risks, with corresponding levels of regulation.
- Transparency and documentation: Companies developing high-risk AI systems must provide detailed technical documentation and follow transparency requirements.
- Ban on unacceptable practices: Certain AI practices deemed harmful to individuals or society are explicitly banned.
- Focus on generative AI: Specific regulations are established for these powerful models, ensuring transparency and mitigating risks.
- International leadership: The EU’s AI Act sets a global standard for AI regulation, influencing other countries to follow suit.
The agreement still needs formal approval by the European Parliament and Council before taking effect, and once adopted there will be a transitional period before the act becomes applicable. The EU will continue to work with international organizations to promote trustworthy AI rules globally.
