Yesterday, the European Union (EU) reached a groundbreaking consensus on AI regulations, setting a precedent as the first substantial legal framework for artificial intelligence (AI) in the Western world. This development marks a pivotal moment in the EU’s ongoing efforts to establish a comprehensive regulatory landscape for AI technologies, a journey that began with the European Commission’s proposal for a unified legal and regulatory framework in 2021.
The agreement follows intense discussions among major EU institutions, focusing on how to manage generative AI models like ChatGPT and biometric identification tools. Germany, France and Italy, home to burgeoning AI startups such as DeepL and Mistral AI, advocated for self-regulation by companies, fearing that stringent regulations could hinder Europe’s competitive edge in the global AI race.
The newly established EU AI Act classifies AI technologies based on risk levels, ranging from “unacceptable” technologies that warrant prohibition to varying degrees of lower-risk AI. The emergence and capabilities of generative AI tools like ChatGPT, which can produce complex outputs from simple prompts, have significantly influenced this legislative development, addressing concerns about job displacement, discriminatory language, and privacy infringement.
