AI and the EU: What's New?

Artificial intelligence surrounds us. Given AI’s increased competencies and prevalence, supranational institutions like the EU created institutional and legal frameworks to regulate AI and build trust around AI in the EU.

The first artificial intelligence law was proposed by the European Commission in 2021. After three years of revisions and committee meetings, the European Council formally adopted the EU AI Act in May of 2024, and the Act was formally published in the Official Journal of the EU in July 2024. 

So what does this mean for the EU and artificial intelligence? 

The EU AI Act’s primary objective is to create a clear structure for AI developers and deployers regarding specific uses of AI. Analysis of artificial intelligence programs is structured around risk-based rules. The risk and its subsequent roles are divided into four levels: unacceptable, high, limited, and minimal (or no) risk. 

For some perspective of scale, unacceptable risks of AI include criminal profiling based on personality traits, high risk encapsulates migration, asylum and border control management, and limited and minimal risk comprise of chatbots and spam filters. 

The higher risk associated with artificial intelligence platforms constitutes more stringent rules for its use. These policies are in place to ensure that safety and human-centric AI are guaranteed across the EU. 

As the first major regulator of AI anywhere, the EU’s policy impacts more than EU-based companies establishing new AI platforms. All AI platforms regardless of origin must abide by its framework if used in an EU member country. American AI companies like OpenAI (ChatGPT) must ensure that their generative programming follows EU legislation and privacy acts with the understanding that if rights, safety, or ethical principles are violated, action by the European Commission is likely.

February 2025 was an action-packed month for AI. The European Commission announced a €200 billion investment in artificial intelligence and banned AI initiatives in the “unacceptable” category with distinct timelines for the future. 

The EU through its prioritization of artificial intelligence and role as the first large institutional framework to create actionable policies on AI set a precedent for other countries and institutions to follow. It will be fascinating to see how AI investments, privacy, and legislation progress in the coming years throughout the EU.