United Kingdom of Great Britain and Northern Ireland

The Dialogue should aim to enable adoption of AI globally for the benefit of all, through measures that increases trust in AI. To do this the Dialogue should focus on these outcomes: - Building an evidence base to inform action. Sharing the latest science around AI capabilities and risks to increase understandings globally and bridge knowledge divides. - Facilitating diverse connections. Enabling engagement within and between different communities (government, technical, academia, civil society, private sector) from all regions creating relationships and catalysing coalitions around common interests that continue beyond the Dialogue. - Knowledge exchange. Increasing awareness globally of national perspectives and approaches to AI governance, encouraging exchange on how policy, regulation, standards and technical tools can be used together to enable AI adoption. - Mobilising momentum though practical capacity building. Sharing of practical approaches, toolkits and case studies that could be adapted to national contexts to support development and implementation of governance that enables adoption.

• AI capacity-building
• Safe, secure and trustworthy AI
• Interoperability of governance approaches
• Protection and promotion of human rights

The UK has highlighted four areas that we believe the Dialogue could address as a priority. However, we assess that all themes are interconnected and play an important role in enabling widespread adoption of AI for the benefit of all. Safe, secure and trustworthy AI and the protection and promotion of human rights are interlinked and essential for individuals, businesses and governments to adopt AI at scale. Interoperability of governance approaches will help to facilitate cooperation while respecting different national contexts, approaches and levels of development. The UN can play a role in promoting shared understandings of governance approaches, encouraging awareness of 'crosswalks' between approaches, rather than seeking standardisation. AI capacity building is necessary for all countries to maximise the potential of AI for all. The UN can play a role in directing political momentum towards capacity building and acting as a platform to promote capacity building activities and expertise through the UN AI Resource Hub.

The UK assesses that the above themes are sufficiently broad to capture the vast majority of cross-cutting or emerging issues. However, we believe that it is important to consider how emergent risks (misuse, malfunction, systemic risks) could intersect with each of the themes and how this in turn informs governance approaches. In line with others, we also believe it is important to consider energy and environmental considerations, such as energy, water, and infrastructure demands of AI systems, and implications for sustainability, and energy security This could be considered as part of the social theme.

The Dialogue can add value by establishing itself as a trusted, practical and inclusive convening space focused on evidence and information sharing, learning and cooperation. - Scientific-evidence informed: ensuring that international cooperation is informed by robust scientific evidence. strengthening the quality, credibility and impact of global AI governance discussions. - Trusted and practical convening space: Establishes itself as a place to bring together diverse groups to support peer learning, identify common challenges, and help spread effective approaches across different contexts. Encouraging engagement between policymakers and experts, - Interoperability: facilitating practical exchange on how different governance approaches are being implemented.

The Dialogue should spotlight the following initiatives, fostering a shared evidence base and enabling countries to take practical steps on AI governance: - The International AI Safety Report offers a comprehensive review of science on the capabilities and risks of general-purpose AI systems. This provides institutions with a robust foundation to understand risks and respond proportionately to their likelihood and severity. - Featuring the Report in Dialogue sessions could help ground discussions in robust analysis and ensure alignment with the work of the UN's Independent International Scientific Panel. Showcasing these efforts would efforts would support a consistent evidence base, while enabling insights from broader initiatives to inform ongoing analysis. - AI Safety and Security Institutes, including the UK's AISI, provide a further example of practical implementation. These government-backed research organisations, dedicated to building the world's leading understanding of advanced AI risks and solutions, inform governments so they can keep the public safe. By spotlighting AISI's ongoing work, the Dialogue can support interoperability across AI security approaches. - Beyond UN-led discussions, there is substantial international activity focused on practical implementation through standards development organisations (SDOs) as the UN-backed AI Standards Summit in Seoul and the UK-led AI Standards Hub Summit. - Regional efforts such as the Council of Europe's HUDERIA framework are developing practical tools to operationalise responsible AI governance. The opportunity for the UN Global Dialogue is therefore not to duplicate this work, but to anchor shared values and principles, while actively connecting into these established SDO processes where detailed implementation can be developed..

Interactive formats: The Dialogue should prioritise smaller, focused roundtables and moderated exchanges that enable genuine interaction, peer learning and practical discussion. This would sit alongside plenary discussions.

Women and girls: continue to be underrepresented in both the design and governance of AI systems. Their inclusion can be strengthened through targeted outreach, support for participation, and ensuring gender balance among speakers, experts and contributors. Civil society organisations: particularly those working on human rights, equality and social impacts, are not always consistently or meaningfully engaged. Practitioners from non-traditional AI sectors: including public service delivery, development, science, health and education, are often overlooked in favour of purely technical or regulatory voices, particularly pertinent given AI is a general purpose technology. Communities from the Global South: Barriers include capacity constraints, resource limitations and access to technical forums.

Formats could include: - Smaller, themed roundtables: discuss specific implementation challenges, enabling more candid exchange than plenary sessions and helping participants share concrete experience and lessons learned. - "Show and tell" sessions: Short, practical demonstrations of concrete tools, frameworks or case studies followed by discussion on applicability in different contexts. - Case-based sessions: Structured examination of real‑world examples to explore what worked and how approaches can be adapted across regions or sectors.

Policy: The UK's AI Opportunities Action Plan directs actions that support foundational capabilities (e.g. infrastructure, data, skills, building an AI assurance ecosystem), adoption (e.g. public sector, removing adoption barriers) and investment (e.g. attracting and retaining companies, investing in sovereign capabilities). Practices: The UK's Algorithmic Transparency Reporting Standard (ATRS) promotes accountability by requiring public sector bodies to document how algorithms are used, supporting scrutiny and trust - we understand Estonia followed suit. Similarly, the GDS Data and AI Ethics Framework and AI Playbook embeds principles such as fairness, accountability, and transparency into decision-making processes. Platform: The OECD AI Policy Toolkit - an online platform which allows emerging economies to tailor their policy toolkits in line with responsible AI principles - this has been tested in countries who are not OECD members in regional engagements across Latin America, Africa and SE Asia. Platform: AI for Development and AI for Science: UK-supported efforts to apply AI to accelerate sustainable development outcomes and scientific progress. The Dialogue could help share lessons from UK initiatives, such as the AI for Development Programme and the AI Research Resource, and connect partners.