The Centre for the Governance of AI
Budget
Not declared
EP Access
0
accredited persons
Staff
3
1.1 FTE
EU Grants
None
Mission & Goals
The Centre for the Governance of AI's mission is to positively shape the lasting impact of artificial intelligence on the world, by supporting informed and effective decision-making by private and public institutions regarding the development and deployment of artificial intelligence. This is accomplished by conducting and producing research, developing talent, advising relevant stakeholders, and organising public and non-public events. The central focus of our research is threats that general-purpose AI systems may pose to security. We seek to understand the risks they pose today, while also looking ahead to the more extreme risks they could pose in the future. Although our central focus is on threats to security, we also support research on a broader array of risks from AI including the risks of unemployment, inequality, lack of privacy, and unaccountable decision-making by companies. Our researchers have advised decision makers in government, industry, and civil society.
EU Legislative Interests
The Centre for the Governance of AI aims to support informed and effective decision-making regarding the development and deployment of artificial intelligence. In the EU, this has primary revolved around the Artificial Intelligence Act's draft Code of Practice, as well as requests for technical expertise from the EU AI Office. The following examples illustrate areas of possible future monitoring and engagement: - Digital Markets Act - Digital Services Act - Chips Act - Comprehensive European industrial policy on artificial intelligence and robotics
Communication Activities
Much of the research produced by the Centre for the Governance of AI is published publicly in journals, online publications, preprint servers, and on the governance.ai website. Some of these research papers have discussed EU policies and their wider impact on the governance of AI systems. Recent examples include: Training Compute Thresholds: Features and Functions in AI Regulation: https://www.governance.ai/research-paper/training-compute-thresholds-features-and-functions-in-ai-regulation; Risk management in the Artificial Intelligence Act: https://www.governance.ai/research-paper/risk-management-in-the-artificial-intelligence-act; and The Brussel Effect and Artificial Intelligence: https://www.governance.ai/research-paper/brussels-effect-ai. The Centre for the Governance of AI also regularly conducts surveys. Relevant work includes a survey of economists on the economic effects of AI, which contained a question on whether there will be a Brussels Effect for AI, and a public opinion survey with questions related to AI policy, part of which was conducted in the EU. Researchers from the Centre for the Governance of AI have also provided technical expertise and research support to the European Commission's Joint Research Centre and General-Purpose AI Code of Practice drafting process. Our Director of Policy and Research currently serves as a vice-chair for the Code of Practice for general-purpose AI systems with systemic risk.
Interests Represented
Does not represent commercial interests
Member Of
The Centre for the Governance of AI is a member of the US AI Safety Institute Consortium: https://www.nist.gov/artificial-intelligence-safety-institute/aisic-members
Organisation Members
Individual affiliates of the Centre for the Governance of AI are listed on our website: https://www.governance.ai/people
Additional Information
The Centre for the Governance of AI became an independent nonprofit entity in H2 2024 and has not closed a financial year under the new structure.
Commissioner Meetings
No recorded meetings with EU commissioners.