The European Union has unveiled the structure of its new AI Office, an essential body for building the ecosystem and overseeing compliance with the forthcoming AI Act. This risk-based regulatory framework for artificial intelligence is expected to be enforced before the end of July, following last week's final approval by EU lawmakers. The AI Office will officially commence operations on June 16.
The establishment of the AI Office underscores the EU's ambitious vision for AI. This office will play a pivotal role in shaping the European AI landscape over the next few years, balancing the responsibilities of regulating AI risks and promoting innovation. Additionally, the EU aims for the AI Office to have a significant influence on a global scale, as various countries seek to establish their own AI governance frameworks. The office will consist of five specialized units.
Overview of the AI Office Units
1. Regulation and Compliance Unit: This unit will focus on the application and enforcement of the AI Act in collaboration with EU Member States. It will assist in investigations related to compliance, managing infringements, and administering sanctions. This unit is integral to supporting the decentralized governance structure that the AI Act will create in EU countries.
2. AI Safety Unit: Focused on identifying systemic risks associated with high-capacity general-purpose models (GPAIs), this unit will evaluate potential mitigation strategies and testing approaches. With the rise of generative AI technologies like ChatGPT, the AI Safety Unit will be essential in ensuring that GPAIs meet established safety standards as defined by the law.
3. Excellence in AI and Robotics Unit: Dedicated to fostering research and development in AI, this unit will coordinate efforts with the previous "GenAI4EU" initiative, which aims to enhance the development and application of generative AI technologies while upgrading Europe’s supercomputing infrastructure.
4. AI for Social Good Unit: This unit will design and implement international projects that leverage AI for societal benefits, tackling issues such as climate modeling, cancer diagnostics, and digital representations for artistic restoration.
5. AI Innovation and Policy Coordination Unit: This unit will ensure the effective execution of the EU's AI strategy by monitoring trends and investments, promoting the adoption of AI, and supporting innovative ecosystems through regulatory sandboxes and real-world testing.
The AI Office's structure, with three out of five units focused on fostering AI growth and investment, aims to alleviate concerns from industry stakeholders regarding the perceived anti-innovation stance of the EU's regulatory framework. The EU maintains that building trust is crucial for widespread AI adoption.
The appointments for the AI Office have begun, with several key figures already confirmed, including Lucilla Sioli as the head of the AI Office, Kilian Gross for Regulation & Compliance, Cecile Huet for Excellence in AI and Robotics, Martin Bailey for AI for Social Good, and Malgorzata Nikowska for AI Innovation and Policy Coordination. However, the role of chief of the AI Safety unit remains vacant.
Initiated by a Commission decision in January and formally establishing its structure in late February, the AI Office operates under the EU's digital department, DG Connect, led by internal market commissioner Thierry Breton. The office is set to employ over 140 professionals, including technical staff, lawyers, and economists. Currently, around 60 staff have been appointed, with plans for further recruitment as the AI Act is implemented over the coming years. The AI Act will roll out its regulations in phases, with various provisions taking effect from six months to over a year after its enactment.
A critical upcoming function for the AI Office is to develop Codes of Practice and best practices for AI developers, which will serve as interim guidelines while the full regulatory framework is established. A Commission official indicated that this Code is expected to launch shortly after the AI Act's enforcement this summer.
Moreover, the AI Office will collaborate with various forums and expert bodies established under the AI Act to integrate governance and ecosystem-building efforts effectively. This includes working with the European Artificial Intelligence Board, comprising members from various EU Member States and an independent scientific panel, along with a broader advisory forum encompassing industry, startups, academia, and civil society.
The Commission has announced that the first meeting of the AI Board is anticipated to occur by the end of June. Additionally, the AI Office is preparing guidelines for defining AI systems and outlining prohibitions, both due within six months post the AI Act's enforcement, as well as coordinating the development of codes of practice for general-purpose AI models, expected nine months after the law is implemented.
Stay informed about our upcoming AI newsletter by signing up here to receive it starting June 5.
Ranking
