Digital Partners

A successful first Global Dialogue on AI Governance should move beyond high-level principles and deliver tangible foundations for implementation. First, it should establish a shared global baseline on AI governance, not as a rigid framework but as a set of interoperable principles that can be adapted across regions. This includes clarity on data governance, accountability mechanisms, and minimum standards for transparency and safety. Second, the Dialogue should bridge the persistent gap between policy and investment. AI governance cannot be effective without addressing the underlying infrastructure—compute capacity, data ecosystems, and digital public infrastructure. A key outcome would be the recognition that governance must be aligned with financing mechanisms and development strategies, particularly in emerging markets. Third, it should advance a more balanced global AI value chain. This means enabling countries, especially in the Global South, to move from being passive consumers of AI technologies to active participants in data value creation, innovation, and infrastructure development. Fourth, the Dialogue should propose an operational multi-level governance model, connecting global coordination (UN level), regional frameworks (e.g., African Union, European Union), and national implementation capacities. Without this articulation, fragmentation will persist. Finally, success would be measured by the creation of a follow-up mechanism, ensuring continuity beyond the Dialogue—through working groups, pilot initiatives, and partnerships involving public and private stakeholders. In essence, the Dialogue should mark a shift from principles to execution, from fragmentation to coordination, and from dependency to shared value creation.

• AI capacity-building
• Interoperability of governance approaches
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Open-source software, open data and open AI models

My selected priorities reflect the need to move from principles to implementation in AI governance. AI capacity building is foundational. Without local capabilities-skills, institutions, and infrastructure-countries cannot effectively govern or benefit from AI. Capacity must go beyond training to include data ecosystems and access to compute. Compatibility of governance approaches is critical to address fragmentation. As regions develop their own frameworks, ensuring interoperability between global, regional, and national levels is essential to enable collaboration, investment, and cross-border innovation. Transparency, accountability, and human oversight remain central to building trust in AI systems. However, these must be operationalized through clear mechanisms embedded in both public sector deployment and private sector practices, including measurable standards and governance processes. Finally, open source, open data, and open AI models are key enablers of more balanced participation in the global AI ecosystem. They can help reduce technological dependency, foster innovation, and support more inclusive access to AI capabilities, particularly in emerging markets. Taken together, these priorities emphasise a governance approach that is practical, interoperable, and aligned with the realities of infrastructure, investment, and global value chains.

Yes, several critical cross-cutting and emerging issues remain insufficiently addressed. First, the alignment between AI governance and investment frameworks is largely missing. Governance discussions often focus on principles and regulation, while the financing of AI infrastructure, compute, data centers, connectivity, and digital public infrastructure follows separate tracks. Without integrating governance into investment decisions, there is a risk of locking countries into dependent technological pathways. Second, the governance of AI infrastructure and compute capacity is an emerging priority. Access to compute is becoming a key determinant of participation in the AI value chain, yet it remains highly concentrated. Questions around ownership, access models, and strategic autonomy need to be explicitly addressed. Third, there is a growing need to consider data value creation ecosystems, not only data protection or access. Countries need frameworks to capture, share, and retain value from data, especially in sectors such as health, agriculture, and public services. Fourth, global AI value chain imbalances require attention. Many countries remain positioned as consumers rather than producers of AI solutions. Governance frameworks should explicitly support more equitable participation, including through local innovation ecosystems and partnerships. Finally, the link between AI governance and development outcomes including SDGs remains underdeveloped. AI should not only be governed as a risk domain, but also as a driver of inclusive and sustainable development. Addressing these cross-cutting issues would help shift AI governance from a primarily normative exercise to a more systemic and implementation-oriented framework.

Gaps in AI governance, regulation, and uneven progress across priority areas significantly affect both emerging markets and global sectors. In regions such as Africa, limited AI capacity and infrastructure constrain the ability not only to adopt AI, but also to design and enforce effective regulatory frameworks, increasing the risk of long-term technological dependency. At the same time, fragmented and sometimes misaligned regulatory and governance approaches across regions create uncertainty for investors and slow cross-border collaboration. Insufficient progress in transparency and accountability mechanisms also undermines trust, particularly in public sector deployments where regulatory frameworks remain incomplete or weakly implemented. However, these challenges also present opportunities. Strengthening capacity and regulatory capabilities can support more effective governance and attract investment. Promoting interoperable regulatory frameworks can facilitate regional integration and scale. Finally, advancing open models and data initiatives can lower barriers to entry and enable more balanced participation in the global AI ecosystem. Addressing these gaps is essential to ensure that AI development is both well-governed and aligned with sustainable and inclusive growth.

The Global AI Dialogue can play a key role as a platform for alignment and coordination of international AI governance. It can promote interoperability between regulatory frameworks, enabling cross-border collaboration and reducing fragmentation. It can also help bridge the gap between policy and implementation by aligning governance principles with infrastructure, investment, and real-world deployment challenges. Importantly, the Dialogue can ensure more inclusive participation, allowing emerging markets to actively shape AI governance rather than remain passive adopters. Finally, it can catalyse practical cooperation, through joint initiatives, knowledge sharing, and follow-up mechanisms that translate discussions into concrete actions. Overall, the Dialogue can help move from fragmented approaches toward a more coordinated and implementation-oriented global governance model.

The Global AI Dialogue should build on and connect with existing international, regional, and multi-stakeholder initiatives to avoid duplication and accelerate impact. Key frameworks include the United Nations system initiatives, the OECD AI Principles, and regional efforts such as the European Union AI Act and the African Union AI Strategy. Platforms like the Global Partnership on AI (GPAI) also contribute to international coordination. A critical dimension is the role of development finance institutions and agencies, including the World Bank, African Development Bank, European Investment Bank, European Bank for Reconstruction and Development, Agence Française de Développement, Enabel, GIZ, Proparco, and Finnfund. These actors are essential to translate governance frameworks into concrete investments in AI infrastructure, data ecosystems, and digital public infrastructure. The Dialogue's added value lies in its ability to connect policy, financing, and implementation. It can promote interoperability between governance frameworks, align regulatory approaches with investment strategies, and ensure stronger inclusion of emerging markets. Ultimately, it can move from fragmented initiatives to coordinated, implementation-oriented action at scale.

Different stakeholders can contribute to the Global AI Dialogue through complementary roles, provided the format enables both strategic alignment and practical outcomes. Governments and international organisations should focus on policy coordination and interoperability, ensuring alignment across regulatory frameworks. Development finance institutions and agencies should contribute by linking governance discussions to investment priorities and implementation mechanisms. The private sector and academia should bring technical expertise, innovation perspectives, and operational insights, while civil society should ensure inclusion, ethics, and accountability. To be effective, the Dialogue should adopt a multi-layered structure: 1- High-level plenary sessions to define strategic priorities and political alignment. 2- Thematic working groups focused on key areas such as AI infrastructure, data governance, capacity building, and regulatory interoperability. 3- Regional tracks to reflect specific contexts and ensure meaningful participation from emerging markets. 4- Implementation oriented labs or pilots, connecting policy discussions with real-world use cases and investment projects. In addition, the Dialogue should establish clear follow-up mechanisms, including timelines, deliverables, and coordination platforms, to ensure continuity beyond the event. Overall, the format should prioritise interaction, co-creation, and execution, rather than purely declarative discussions, enabling stakeholders to move from dialogue to concrete action.

everal voices remain under-represented in global AI governance discussions. Actors from emerging markets, particularly in Africa, are often consulted but rarely involved as co-designers, which limits ownership and relevance. Implementation actors such as regulators in practice, public sector operators, and development agencies are also not sufficiently represented, even though they are essential to translating policy into action. Local innovation ecosystems, including startups and research institutions, as well as investment and infrastructure actors, are often disconnected from governance discussions, despite their role in shaping access to data, compute, and AI capabilities. Inclusion should therefore move beyond consultation toward genuine co-creation. This requires stronger participation in decision-making processes, dedicated regional engagement, and closer integration of implementation and financing actors. Capacity building is also critical to enable meaningful and sustained participation. Better inclusion of these perspectives will lead to more practical, balanced, and globally relevant AI governance.

The Dialogue should prioritize interactive and outcome-oriented formats over traditional panels. Small, curated working groups can enable practical exchanges around concrete challenges. Implementation labs linked to real use cases or investment projects can help bridge policy and execution. Regional dialogues should be integrated to ensure contextual relevance, while structured matchmaking sessions can connect public, private, and development actors. Finally, digital platforms can support continuous collaboration beyond the event. Overall, the focus should be on co-creation, practical outcomes, and sustained engagement.

Several concrete initiatives demonstrate how AI governance can be effectively operationalised. The Data Governance Value Creation in Africa approach, developed under the EU-AU partnership, provides a practical framework linking data governance to socio-economic value creation, investment, and sectorial use cases, moving beyond compliance toward impact-driven policies. At the continental level, the African Union AI Strategy illustrates how governance can integrate policy, capacity building, and public value objectives, aligned with development priorities. The World Bank report Global Trends in AI Governance highlights how countries combine regulatory tools-soft law, hard law, and regulatory sandboxes to balance innovation and risk, emphasising the need for context-specific and adaptive frameworks. In Europe, the EU AI Act provides a concrete risk-based regulatory model, complemented by AI Watch, which supports evidence-based policymaking through monitoring and benchmarking of AI use in public services. The PRIDA (Policy and Regulation Initiative for Digital Africa) programme demonstrates how regulation, capacity building, and sectorial strategies can be combined to support implementation across multiple countries. Finally, initiatives such as AI for Good illustrate the value of multi-stakeholder platforms in translating governance discussions into practical solutions, particularly around standards, trust, and societal impact. These examples show that effective AI governance requires aligning policy, regulation, capacity, and investment to deliver real-world impact.