EU's AI Act enters full enforcement phase as Brussels fines three companies in landmark cases
The European Union's Artificial Intelligence Act has moved into its full enforcement phase, with the EU AI Office issuing its first substantive fines to two US technology companies and a European financial services firm for high-risk AI violations.
Technology — March 12, 2026
The European Union's landmark Artificial Intelligence Act entered its full enforcement phase in March 2026, with the EU AI Office confirming that it had opened formal investigations resulting in the first significant fines under the regulation — a development that industry lawyers described as the beginning of a new era of AI governance.
Three companies received enforcement notices: a major US-based artificial intelligence platform was fined €45 million for deploying a generative AI system in recruitment processes without meeting the transparency and human oversight requirements mandated under the Act's provisions for high-risk AI applications.
A second US technology company was fined €28 million for failing to register a biometric categorisation system used in retail security applications with the EU's AI database — a mandatory requirement for all high-risk AI deployments across member states.
A European financial services company received a smaller fine of €12 million for using an AI-based credit-scoring system without providing applicants with the explanation rights guaranteed under Article 86 of the Act.
The EU AI Office said the enforcement actions were designed to send a clear signal. 'The AI Act is not aspirational — it is law,' said the office's director. 'We will enforce it fully and without regard for the size of the entity involved.'
The announcement drew criticism from US technology trade associations, who argued that the fines reflected a regulatory framework designed to disadvantage American technology companies in the European market. The US Chamber of Commerce called on the USTR to raise the issue in bilateral trade discussions.
European technology policy analysts said the enforcement cases were significant not just for their immediate financial impact, but for establishing legal precedent. 'These are the first cases under the Act. The legal interpretations reached here will shape AI deployment across every sector for the next decade,' said one Brussels-based policy lawyer.
