The European Parliament has approved the most comprehensive artificial intelligence regulation in the world, setting new global standards for AI transparency, safety, and accountability.
The AI Act 2.0, passed with a decisive 487-153 vote, expands upon the original 2024 framework with sweeping new provisions targeting generative AI, autonomous systems, and AI-powered decision-making in critical sectors.
Key provisions include mandatory watermarking of all AI-generated content, real-time monitoring requirements for high-risk AI systems, and the establishment of a European AI Safety Institute with enforcement powers.
"This legislation ensures that Europe leads in responsible AI development," said European Commissioner for Digital Transition, Marie-Louise Björk. "Innovation must go hand in hand with fundamental rights."
Tech companies face significant compliance costs, with major firms like OpenAI, Google, and Meta required to submit their models for safety testing before deployment in the EU market. Non-compliance penalties can reach up to 7% of global revenue.
The regulation has drawn mixed reactions. While civil rights organizations praised the transparency requirements, tech industry groups warned that overly strict rules could drive AI innovation out of Europe.
The act is expected to take full effect within 18 months, with a phased implementation schedule for different risk categories.