Publication of the European AI Act

On 12 July 2024, Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonized rules on artificial intelligence (AI Act) was published in the Official Journal of the European Commission. At the level of EU law, this fact represents the culmination of an important step in the process of implementing and applying the law of the regulatory framework on artificial intelligence within the territory of the European Union. Responding to the development of the indicated regulation, the Polish Ministry of Digitization is already working on a draft Polish act, adapting Polish law to regulations in the new area.

Among the most important changes on the artificial intelligence level, the AI Act provides for:

Categorization of AI systems: Creating a classification of AI systems based on the risk they generate (high, limited, minimal), with stricter requirements for high-risk systems.

General-purpose AI models: A special category of AI models that can be used for various purposes, with specific obligations and restrictions.

Transparency obligation: Requirement to inform users about interactions with AI and to label synthetically generated content, including deepfakes.

Prohibition of certain practices: The act introduces a category of unacceptable practices in the area of AI, such as emotion manipulation, citizen scoring, and real-time remote biometric identification.

Introducing oversight of AI: The creation of a European Council on Artificial Intelligence, the obligation to establish national supervisory authorities, and hefty fines for AI Act violations of up to €35 million or 7% of annual turnover.

Most of the provisions that make up the AI Act will go into effect on 2 August 2026, with some of the provisions, particularly those that make up Chapters I, II, V, VII and XII, coming into effect as early as 2025.

#AIAct #ArtificialIntelligence #EURegulation #LegalCompliance #TechLaw #EuropeanUnion #DigitalStrategy

---

Let's talk