EU Statement -- Informal Consultations on Artificial Intelligence
Excellencies, Co-facilitators, dear colleagues,
I have the honour to deliver this statement on behalf of the European Union and its Member States.
The Candidate Countries Türkiye, North Macedonia, Montenegro, Albania, the Republic of Moldova, Bosnia and Herzegovina, Georgia and Ukraine align themselves with this statement.
I thank the co-facilitators for providing guiding questions to enable us to go more in-depth into the subject. As we gather for the second time, we look forward to hearing more ideas from delegations about how best to shape the Scientific Panel and Global Dialogue on AI.
Building on the statement we delivered last month, the EU would like to share the following additional elements. You will find our detailed proposal in the written submission uploaded to e-delegate.
As already stated, the multidisciplinary Independent International Scientific Panel on AI should provide reliable, credible, and representative scientific knowledge on the opportunities, risks, and capabilities of AI. The panel should synthesize and consolidate available research, and provide evidence-based impact, opportunity, and risk assessments. It should not prescribe policy but support knowledge- and science-based discussions and decisions.
The panel should consist of 35 to 50 members, selected by an independent committee, ensuring a balanced geographical representation as well as gender balance. It should be composed exclusively of scientists and researchers from a broad range of disciplines. It should be chaired by two members, one from a developed country and one from a developing country, assisted by vice-chairs, all originating from a different region, with the co-chairship rotating among regions every year. The panel should have a flexible governance structure to adapt to the fast-evolving AI landscape. It must be fully independent, supported by a lean secretariat, and endowed with adequate funding from the UN regular budget, with additional voluntary contributions documented transparently. The panel should aim to produce annual reports, which could be complemented by concise briefs on relevant developments on an ongoing basis. It should build on existing national, regional and international initiatives and research networks, without duplicating them.
The Global Dialogue on AI Governance should aim to strengthen stakeholders' understanding of AI. It should provide a space for knowledge-sharing and capacity building, focusing on international cooperation, security, trustworthiness, human rights, and sustainable development. The Dialogue should focus on practical strategies for harnessing global AI opportunities for sustainable development. It should also promote collaboration with existing and emerging initiatives and organizations, without duplicating other AI-related international and regional activities.
The Global Dialogue should be open to all stakeholders, including all UN Member States, and take place annually on the margins of existing multi-stakeholder initiatives such as the AI4Good summit, the Internet Governance Forum, or UNESCO’s Global Forum on the Ethics of AI, as they provide relatively low barriers to participation of stakeholders. It should provide shared spaces for interaction and engagement, with outcomes reflected in non-negotiated messages and/or chair summaries. The dialogue should neither lead to negotiated outcomes, nor seek to impose prescriptive policies or set new norms. It should support in-house dialogues on AI governance within the UN system and create synergies with relevant work streams.
Both the Panel and the Dialogue should maintain institutional and functional separation to preserve the independence of the scientific panel and avoid undue political pressure. They should draw on existing expertise within the UN system and consult regularly with relevant parts of the UN and other entities outside the UN system to promote integrated approaches and resource efficiency. It should draw in particular on the UN Group on the Information Society (UNGIS) and the UN Inter-Agency Working Group on AI (UN-IAWG on AI), as well as the new UN Working Group on GDC implementation.
The promotion and protection of human rights and the achievement of the Sustainable Development Goals should be at the core of both mechanisms.
The EU remains committed to engaging constructively with all stakeholders throughout this process.
Thank you.