EU Begins Enforcing New AI Law: What It Means for Companies and Innovation The European Union has started enforcing its groundbreaking artificial intelligence (AI) law, marking a major step in regulating this fast-growing technology. The EU AI Act, which became official in August 2024, is the first comprehensive set of rules for AI in the world. As of February 2025, certain restrictions and requirements are now in effect, and companies that fail to comply could face hefty fines. The new law bans AI systems that are considered too risky or harmful. For example, it prohibits social scoring systems, real-time facial recognition in public spaces, and AI tools that categorize people based on sensitive traits like race, sexual orientation, or religion. It also targets AI systems that manipulate people’s behavior in harmful ways. Companies that break these rules could be fined up to 35 million euros ($35.8 million) or 7% of their global annual revenue, whichever is higher. These penalties are even stricter than those under the EU’s General Data Protection Regulation (GDPR), which fines companies up to 20 million euros or 4% of their global turnover for privacy violations. The EU AI Act is designed to protect citizens from the potential dangers of AI while encouraging responsible innovation. However, this is just the beginning. The AI Act will roll out in phases, and more rules and guidelines are expected in the coming months. For now, companies must focus on complying with the initial restrictions and ensuring their staff understands the new requirements. While the law has been praised for setting clear rules, some critics argue it could slow down innovation. Prince Constantijn of the Netherlands, for example, expressed concern that Europe’s focus on regulation might limit its ability to compete with AI leaders like the U.S. and China. He emphasized the need for balance between regulation and innovation in such a fast-moving field. On the other hand, supporters believe the EU AI Act could set a global standard for trustworthy AI. By requiring companies to detect biases, conduct regular risk assessments, and ensure human oversight, the law aims to create safer and more reliable AI systems. Diyan Bogdanov, a director at fintech firm Payhawk, noted that Europe is leading the way in defining what “good” AI looks like, even as other countries focus on building bigger and more powerful models. The EU AI Act also includes specific rules for general-purpose AI models, like OpenAI’s GPT. These models must undergo rigorous testing to ensure they are safe and transparent. Open-source AI models, however, are exempt from some of these requirements, which has been welcomed by developers in the open-source community. As the EU begins enforcing this landmark law, companies around the world are watching closely. The AI Act could influence how other countries regulate AI, setting a precedent for global standards. For now, businesses operating in the EU must adapt to the new rules or face significant penalties. While the road ahead may be challenging, the EU AI Act represents a bold step toward ensuring that AI benefits society while minimizing its risks. As the technology continues to evolve, the world will be looking to Europe for guidance on how to balance innovation with responsibility.
