In this privacy update we'll take a look at the latest updates on AI legislation and governance - specifically the AI Act, the AI Pact and the AI Office.
On February 2nd, representatives from EU member states unanimously voted to advance the EU AI Act. When the political deal on a compromise text was reached back in December 2023, it seemed that the EU ambassadors’ approval would be a mere formality. However, approval was not guaranteed since the largest EU economies resisted parts of the provisional deal up until the very last days of negotiation. The primary opponent of the political agreement was France which (together with Germany and Italy) asked for a lighter regulatory scheme for powerful AI models, such as OpenAI’s ChatGPT. So the unanimous vote of Feb 2nd was a relief.
With confirmed approval from the European Council, all eyes are now on the European Parliament with the full plenary vote expected in April 2024. The AI Act will come into force 20 days after publication in the EU Official Journal and May 2026 will likely mark the start of full implementation of the AI Act. However, there will be phases in which specific provisions of the AI Act would come into force before then. For example, prohibitions on unacceptable risk AI systems would apply 6 months after entry into force, so already in autumn of this year. Here’s a helpful infographic of the AI Act’s implementation timeline.
As some provisions of the AI Act will only be applicable at the end of a transitional period (i.e. in 2026), the European Commission is initiating the AI Pact to encourage and support companies in their voluntary implementation of AI Act requirements prior to the legal deadline. On the heels of the formal adoption of the AI Act, the AI Pact will be launched officially and influential organisations will be invited to make their first public pledges. The AI Pact will also allow businesses to engage with the Commission and other relevant stakeholders through joint activities and sharing of best practices.
The European Commission will soon launch the European AI Office which will consist of various experts for policy, technical and legal work and administrative assistance.
Its objectives are to support the development and safe use of AI as well as assist with the implementation of the AI Act. Additionally, the office will enforce general purpose AI rules, foster international cooperation on AI governance and oversee the AI Pact.