The EU’s AI Act entered into force on 1 August, starting a two-year process of implementation. Between now and the start of August next year, governments across the EU will appoint national authorities for AI. Other provisions will be gradually brought into force over time, with most rules applying from 2 August 2026.
The AI Act regulates artificial intelligence (AI) based on the risk that an AI model poses. This can range from minimal risk systems, which won’t face any extra regulation, up to ‘unacceptable risk’ applications, which will be banned. Examples of applications considered to pose an ‘unacceptable risk’ include dangerous manipulation of human behavior, emotion monitoring in the workplace, and some forms of predictive policing.
To facilitate a smooth transition, the European Commission has launched the AI Pact, encouraging AI developers to voluntarily adopt key obligations ahead of the deadlines. Additionally, the Commission is developing guidelines and standards to support the AI Act’s implementation.
Alongside the AI Act’s implementation, the European Commission has also opened a consultation on the creation of a Code of Practice for providers of ‘general purpose artificial intelligence’ systems (GPAI). The consultation seeks input from providers of these systems themselves, as well as civil society and academia, and will run until 10 September 2024.
More information about the AI Act is available from the website of the European Commission
