Ministry of ICT and National Guidance

A successful first Global Dialogue on AI Governance should move beyond principles to practical, actionable outcomes that support countries at different levels of readiness. First, it should produce a shared, implementation-oriented framework—including guidance on AI risk management, accountability, and institutional coordination—that countries can realistically adopt and adapt. Global principles already exist; the priority now is operationalization. Second, the Dialogue should result in concrete commitments on capacity building, particularly for developing countries. This includes strengthening policy and regulatory expertise, technical skills, and institutional capabilities required to design, implement, and oversee AI systems. Third, success should be reflected in stronger alignment across global initiatives and standards, reducing fragmentation and helping countries navigate the growing landscape of frameworks, guidelines, and tools. Fourth, the Dialogue should establish sustained multi-stakeholder engagement mechanisms, ensuring continued collaboration between governments, industry, academia, and civil society beyond the event itself. Finally, it should recognize AI governance as a development issue, not only a technology issue—by advancing approaches that are inclusive, context-aware, and responsive to local economic, social, and cultural realities. In essence, the Dialogue will be successful if it shifts the global conversation from "what should be done" to "how it can be done in practice," particularly for countries working to translate AI opportunities into tangible public value.

• Safe, secure and trustworthy AI
• AI capacity-building
• Transparency, accountability, and human oversight
• Interoperability of governance approaches

The selected priorities reflect a practical, implementation-oriented approach to AI governance, particularly from a developing country perspective. AI capacity-building is foundational, as effective governance depends on strengthening policy, regulatory, and technical capabilities across institutions. Without this, even well-designed frameworks cannot be implemented. Safe, secure and trustworthy AI is critical to building public confidence and ensuring that AI systems are reliable, resilient, and aligned with national and international standards. This is especially important as countries adopt AI in sensitive public sector domains. Interoperability of governance approaches is increasingly important in a fragmented global landscape. Aligning national efforts with international frameworks and standards helps avoid duplication, supports cross-border collaboration, and enables consistency in AI oversight and innovation. Transparency, accountability, and human oversight are essential to operationalizing AI governance. Moving beyond high-level principles, countries need practical mechanisms to ensure explainability, assign responsibility, and maintain meaningful human control over AI systems. Together, these priorities address the full spectrum from capacity → systems → coordination → accountability, which is necessary to translate AI governance from policy into effective, real-world implementation.

Yes. Several cross-cutting and emerging issues warrant greater attention beyond the listed themes. First is the implementation gap-the challenge of translating AI principles into operational systems. Many countries have adopted frameworks, but lack practical tools, institutional capacity, and financing to implement, monitor, and enforce them effectively. Second is the governance of AI systems across the lifecycle, including procurement, deployment, auditing, and decommissioning. Public sector adoption of AI requires clear guidance on how to integrate governance into real service delivery environments. Third is data ecosystem readiness, including data availability, quality, interoperability, and governance. AI outcomes are only as reliable as the underlying data, yet many countries face structural data limitations. Fourth is compute and infrastructure inequality, which risks widening the global digital divide. Access to affordable compute, cloud infrastructure, and foundational models remains uneven, limiting participation by developing countries. Fifth is AI assurance and certification mechanisms, including independent validation, auditing, and conformity assessment. There is a growing need for globally recognized but locally implementable assurance frameworks. Finally, coordination of global initiatives remains a challenge. The proliferation of frameworks, standards, and guidelines can overwhelm countries unless better aligned and contextualized. Addressing these cross-cutting issues will be critical to ensuring that AI governance is not only principled, but also implementable, inclusive, and sustainable.

In Uganda and across Africa, the selected thematic areas—AI capacity-building, safe and trustworthy AI, interoperability of governance approaches, and transparency and accountability—highlight both critical governance gaps and emerging opportunities. A major challenge is limited institutional and technical capacity to design, deploy, and regulate AI systems effectively. While policy development is progressing, there remains a significant gap between frameworks and implementation, including limited availability of practical tools for risk management, auditing, and lifecycle governance. Data ecosystem constraints—such as fragmented data systems, limited availability of high-quality datasets, and weak interoperability—affect the reliability, inclusiveness, and scalability of AI solutions. In addition, institutional fragmentation across government entities complicates coordination and slows decision-making. Unequal access to compute infrastructure, digital connectivity, and advanced technologies further risks widening both global and regional disparities. At the same time, there are important positive developments. Uganda is advancing a National AI and Emerging Technologies Strategy, aligned with the Digital Transformation Roadmap, and supported by a multi-stakeholder National AI Task Force. At the continental level, the African Union and regional bodies are promoting harmonized approaches to AI governance, creating opportunities for alignment and shared learning. Engagement with international partners, including UNESCO and International Telecommunication Union, is strengthening capacity-building, ethics frameworks, and readiness assessments. Significant opportunities exist to leverage AI for improved public service delivery in sectors such as health, agriculture, and education. By adopting standards-based, interoperable, and accountable governance approaches, Uganda and Africa can move beyond adoption to become active contributors to global AI governance, while ensuring that AI systems are inclusive, context-aware, and development-oriented.

The AI Dialogue can play a pivotal role in advancing international cooperation by shifting global engagement from fragmented discussions to coordinated, implementation-focused collaboration. First, it can serve as a convergence platform that aligns diverse global initiatives, standards, and policy frameworks. By bridging efforts across institutions such as UNESCO, International Telecommunication Union, and standards bodies, the Dialogue can reduce duplication and provide clearer pathways for countries navigating the complex AI governance landscape. Second, it can facilitate peer learning and exchange of practical experiences, particularly between countries at different levels of readiness. Sharing real-world implementation models—including regulatory approaches, institutional arrangements, and public sector use cases—will be critical to accelerating adoption. Third, the Dialogue can promote inclusive participation, ensuring that developing countries are not only recipients of global norms but active contributors to shaping them. This is essential for ensuring that AI governance frameworks are context-aware and globally representative. Fourth, it can catalyze capacity-building partnerships and resource mobilization, linking countries to technical assistance, financing, and skills development initiatives. Finally, the Dialogue can support the development of interoperable governance approaches, enabling cross-border collaboration, trust, and innovation. In essence, the AI Dialogue should function not only as a forum for discussion, but as a platform for coordination, capacity-building, and sustained global cooperation, translating shared principles into collective action.

The AI Dialogue should build on and connect with a range of existing global, regional, and standards-based initiatives. At the global level, frameworks such as the UNESCO Recommendation on the Ethics of AI and programs by the International Telecommunication Union provide important foundations for ethical, inclusive, and human-centered AI governance. Standards development efforts, particularly through ISO/IEC, are critical for translating principles into operational and certifiable systems. Development partner support is also playing a key role. In Uganda, the World Bank is supporting the development of the National AI and Emerging Technologies Strategy, demonstrating the importance of linking governance frameworks to financing, implementation, and institutional capacity. Emerging cooperation mechanisms such as the Global Network on AI supervision by competent authorities provide a platform for collaboration among regulators and competent authorities, strengthening oversight, supervision, and enforcement capacity. Similarly, the Global Partnership on Artificial Intelligence offers an important forum for policy coordination, research collaboration, and practical guidance on responsible AI. At the regional level, African Union initiatives and related frameworks are promoting harmonization and shared priorities, which are essential for cross-border coherence. However, these efforts often remain fragmented and difficult for countries to navigate. The added value of the AI Dialogue lies in its ability to: Connect and align these initiatives into a coherent global ecosystem Provide practical, implementation-focused guidance Serve as a clearinghouse for tools, standards, and best practices Facilitate coordination between policy, standards, and development financing Amplify the perspectives and priorities of developing countries By linking existing efforts and focusing on practical, context-aware implementation, the AI Dialogue can significantly strengthen the coherence, inclusiveness, and effectiveness of global AI governance.

Effective participation in the AI Dialogue requires a structured, multi-layered approach that enables meaningful contributions from all stakeholders. Governments should contribute policy experiences, regulatory approaches, and implementation lessons, particularly from real-world deployments. The private sector can provide technical expertise, innovation insights, and operational perspectives on AI development and deployment. Academia and research institutions should contribute evidence-based analysis, risk assessments, and emerging research, while civil society brings critical perspectives on ethics, inclusion, and societal impact. To support this, the Dialogue should adopt a hybrid and iterative format: Pre-dialogue consultations at national and regional levels to gather grounded inputs Thematic working groups focused on specific governance areas (e.g., risk management, standards, capacity-building) Case-based sessions where countries and organizations present practical implementation experiences Multi-stakeholder roundtables to foster dialogue across sectors The structure should move beyond plenary discussions to include smaller, action-oriented groups that produce concrete outputs such as policy recommendations, toolkits, or pilot initiatives. In addition, the Dialogue should establish ongoing engagement mechanisms, such as communities of practice or knowledge-sharing platforms, to ensure continuity beyond the event. A well-structured Dialogue will enable stakeholders not only to share perspectives, but to co-create practical solutions, bridging the gap between global principles and national implementation.

Several important voices remain underrepresented in global AI governance discussions, particularly from the perspective of developing regions. These include: Policymakers and regulators from developing countries, especially those in early stages of AI adoption Local innovators, startups, and SMEs, who face unique constraints in accessing data, infrastructure, and markets Public sector practitioners responsible for implementing AI in service delivery Communities affected by AI systems, including rural populations, informal sector workers, and marginalized groups Linguistic and cultural communities whose contexts are often not reflected in global datasets and models Inclusion requires deliberate and structured approaches. First, the Dialogue should support regional and national consultations to capture locally grounded perspectives. Second, it should provide financial and logistical support to enable participation from underrepresented groups, including travel, connectivity, and translation services. Third, contributions should be enabled through multiple formats, including written submissions, virtual participation, and asynchronous inputs, to lower barriers to engagement. Fourth, outputs should explicitly reflect diverse perspectives, ensuring that contributions from developing countries are not diluted in global summaries. Finally, there is a need to strengthen capacity for meaningful participation, so that stakeholders can engage not only as observers, but as informed contributors shaping global outcomes.

To foster meaningful and dynamic engagement, the AI Dialogue should adopt innovative, participatory, and outcome-oriented formats. First, case-based learning sessions should be prioritized, where countries and organizations present real-world AI governance challenges and solutions. This grounds discussions in practical experience rather than abstract principles. Second, interactive policy labs or "co-creation workshops" can bring together diverse stakeholders to jointly develop solutions to specific governance challenges, such as AI risk frameworks or regulatory models. Third, the Dialogue can incorporate simulation exercises or scenario-based discussions, allowing participants to explore responses to emerging risks (e.g., AI misuse, system failures, or cross-border data issues). Fourth, digital collaboration platforms should be used to enable continuous engagement before, during, and after the Dialogue. These platforms can support document sharing, feedback loops, and iterative development of outputs. Fifth, regional hubs or parallel sessions can ensure that discussions are context-sensitive while feeding into a global process. Finally, the Dialogue should focus on tangible outputs, such as toolkits, model frameworks, or joint action plans, developed through collaborative formats. By combining these approaches, the Dialogue can move beyond traditional conferences to become a dynamic, solution-oriented platform that drives real progress in AI governance.

Several practical approaches emerging in Uganda and across Africa demonstrate how AI governance can be translated into implementable systems. First, Uganda's ongoing development of a National AI and Emerging Technologies Strategy, supported by the World Bank, reflects a strategy-first, implementation-oriented approach. This is complemented by the planned use of Regulatory Impact Assessment (RIA) to guide policy, legal, and regulatory decisions-ensuring that governance measures are evidence-based, proportionate, and aligned with human rights principles, including fairness, non-discrimination, and accountability. Second, the establishment of a multi-stakeholder National AI Task Force provides a practical coordination mechanism, bringing together government, academia, and private sector actors to guide AI governance and implementation. Third, Uganda's broader digital governance ecosystem offers relevant platforms that can be extended to AI governance. For example: The Digital Service Delivery Standard embeds transparency, accountability, and user-centric design Existing platforms such as e-Government Procurement (eGP) provide a foundation to integrate responsible AI procurement practices, including requirements for transparency, explainability, data protection, and vendor accountability Importantly, Uganda is advancing procurement mechanisms that prioritize locally developed innovative digital solutions, enabling startups and SMEs to participate competitively. This approach helps address structural disadvantages where local innovators would otherwise be outcompeted by large, established international vendors, while also strengthening domestic innovation ecosystems and ensuring context-relevant AI solutions. Fourth, participation in international initiatives-such as UNESCO AI ethics readiness processes and ITU-led capacity-building-supports local adaptation of global frameworks into actionable policies. Regionally, collaboration through the East African Community, the African Union, and the Smart Africa Alliance is promoting harmonization, shared standards, and coordinated AI policy development, essential for cross-border interoperability. Finally, standards-based governance (e.g., ISO/IEC AI management systems) and emerging AI assurance mechanisms provide practical tools for risk management and compliance. Taken together, these approaches position Uganda-and Africa more broadly-not only as adopters, but as active contributors shaping practical, rights-based, and implementable global AI governance models.