Bias and discrimination in artificial intelligence systems, causes and legal responses in the European Union

Resumen

Artificial Intelligence (AI) increasingly shapes daily life in areas like healthcare, education, and public services. The European Union’s Artificial Intelligence Act (Regulation (EU) 2021/0106, 2024) recognizes AI as a powerful tool but highlights risk that require regulation. A major concern is discriminatory bias in AI systems, often caused by poor data quality, misuse, underrepresentation of relevant information in algorithm design, or excessive irrelevant data leading to biased or discriminatory decisions. This paper explores the root causes of algorithmic bias and reviews the EU’s legal responses, focusing on the AI Act’s regulatory framework. Ensuring fairness and non-discrimination in AI is crucial to protect fundamental rights in an increasingly automated world.