OnPoint

A successful first Global Dialogue on Artificial Intelligence Governance should move beyond general principles toward practical, inclusive, and measurable outcomes. First, it should establish a shared global baseline framework for AI governance that balances innovation with responsibility. This includes clear guidance on safety, ethics, transparency, and accountability, while remaining adaptable to different national contexts—especially for developing countries. Second, success would require meaningful inclusion of the Global South, not only as participants but as contributors to policy design. Countries like Egypt and others in similar contexts must be enabled to shape governance models that reflect their healthcare, economic, and societal realities. Third, the Dialogue should produce sector-specific implementation pathways, particularly in critical domains such as healthcare. AI in healthcare must be governed with a focus on patient safety, data protection, and equitable access, while enabling innovation in areas such as early diagnosis, simulation-based training, and system optimization. Fourth, it should initiate multi-stakeholder partnerships that go beyond discussion into action. This includes collaboration between governments, UN agencies, academia, and institutions on pilot projects, capacity building, and knowledge transfer. Finally, a successful outcome would be the creation of a clear roadmap with milestones, including mechanisms for follow-up, monitoring, and continuous engagement. The Dialogue should not be a one-time event, but the foundation of an ongoing, structured global process. In essence, success lies in translating dialogue into actionable frameworks, inclusive participation, and tangible impact, particularly for sectors that directly affect human well-being.

• Safe, secure and trustworthy AI
• AI capacity-building
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches

The selected priorities reflect the need to balance responsible governance with practical implementation, particularly from the perspective of healthcare systems and developing countries. "Safe, secure and trustworthy AI" is fundamental, especially in high-impact sectors such as healthcare, where patient safety, data protection, and system reliability are critical. Trust is the cornerstone for adoption. "AI capacity-building" is equally essential to ensure that countries with limited resources are not left behind. Building local expertise, infrastructure, and institutional capabilities is key to enabling equitable participation and sustainable implementation. The inclusion of "social, economic, ethical, cultural, linguistic and technical implications of AI" reflects the importance of contextualizing AI governance. Solutions must be adaptable to different societal realities, ensuring that AI systems are inclusive and culturally relevant, particularly in diverse regions such as the Middle East and Africa. Finally, "interoperability of governance approaches" is critical to avoid fragmentation. A harmonized yet flexible global framework allows countries to align with international standards while adapting to national priorities. This is particularly important for cross-border collaboration, data exchange, and international partnerships. Together, these priorities support a governance approach that is inclusive, practical, and globally coordinated, while enabling real-world impact. From a healthcare and institutional development perspective, this alignment ensures that AI is not only governed responsibly but also effectively implemented to improve outcomes, enhance capacity, and support innovation across different regions.

Governance gaps in AI are already having a tangible impact on healthcare systems and institutional development in emerging regions such as the Middle East and Africa. A key challenge is the absence of harmonized regulatory frameworks, which creates uncertainty around data use, cross-border collaboration, and deployment of AI solutions. This limits the ability of institutions to adopt advanced technologies, particularly in sensitive areas such as patient data, clinical decision support, and training systems. Fragmentation also affects partnerships with international organizations and technology providers. Another significant gap is limited AI capacity and infrastructure. While there is strong interest in adopting AI, many institutions lack the technical expertise, standardized training frameworks, and digital readiness required for safe and effective implementation. This creates a risk of uneven adoption, where some entities advance rapidly while others fall behind. There are also ethical and contextual challenges, including ensuring that AI systems are culturally appropriate, unbiased, and aligned with local healthcare realities. Many existing AI models are developed in different contexts and may not fully reflect regional needs. At the same time, these gaps present major opportunities. AI has the potential to transform healthcare delivery, particularly in areas such as early diagnosis, operational efficiency, and simulation-based education and training. For institutions, this creates an opportunity to leapfrog traditional limitations and build more advanced, scalable systems. Furthermore, there is a strong opportunity to position regional institutions as active contributors to global AI governance, particularly by providing use cases from real-world healthcare environments. Addressing these gaps through capacity-building, international collaboration, and adaptable governance frameworks can enable countries and institutions not only to adopt AI responsibly, but to play a meaningful role in shaping its global future.

The AI Dialogue can play a pivotal role by acting as a neutral, action-oriented platform that translates global discussions into structured international cooperation. First, it can serve as a bridge between developed and developing countries, ensuring that governance frameworks are not only globally aligned but also context-sensitive. From a healthcare and institutional perspective, this is critical to ensure that AI solutions are safe, applicable, and scalable across different systems. Second, the Dialogue can facilitate practical collaboration through pilot initiatives, particularly in priority sectors such as healthcare. Joint projects between countries, institutions, and technology partners can demonstrate how governance principles are applied in real-world settings—such as AI-enabled training, clinical decision support, and operational optimization. Third, it can promote capacity-building and knowledge transfer, enabling institutions in emerging regions to develop the necessary technical, regulatory, and operational capabilities. This includes sharing best practices, developing standardized training frameworks, and supporting institutional readiness. Fourth, the Dialogue can contribute to interoperability of governance approaches by aligning standards, guidelines, and regulatory principles. This is essential to enable cross-border data exchange, international research collaboration, and coordinated responses to emerging AI risks. Finally, it can establish sustained engagement mechanisms, including working groups and sector-specific platforms, ensuring that cooperation continues beyond initial discussions and evolves into long-term partnerships. From a practical standpoint, success will depend on the Dialogue's ability to move from general alignment to implementation-driven cooperation, where institutions actively collaborate, share expertise, and contribute to shaping globally relevant, inclusive, and actionable AI governance frameworks.

The AI Dialogue should build upon and connect with existing global and sectoral initiatives to avoid duplication and accelerate impact. At the global level, frameworks such as UNESCO's Recommendation on the Ethics of AI, the OECD AI Principles, and ongoing efforts under the Global Digital Compact provide important normative foundations. Similarly, the work of international organizations such as the ITU on AI standards and the WHO on AI in healthcare offers sector-specific guidance that can be further operationalized. There are also emerging multi-stakeholder platforms and partnerships that focus on practical implementation, capacity-building, and responsible AI adoption. These include collaborations between governments, academia, and industry, as well as regional initiatives aimed at strengthening digital infrastructure and AI readiness in developing countries. From a healthcare perspective, initiatives related to digital health transformation, simulation-based education, and data governance frameworks provide valuable entry points for applying AI responsibly in real-world settings. The added value of the AI Dialogue lies in its ability to act as a unifying and coordinating mechanism. Rather than creating new frameworks, it can align existing efforts, identify gaps, and promote interoperability across different governance approaches. Importantly, the Dialogue can shift the focus from principles to implementation and impact by: Connecting global frameworks with institutional-level application Facilitating pilot projects and cross-border collaboration Supporting capacity-building in underserved regions Enabling continuous knowledge exchange By integrating existing initiatives into a coherent, action-oriented ecosystem, the AI Dialogue can accelerate progress toward inclusive, practical, and globally coordinated AI governance, while ensuring that all regions—particularly developing countries—are actively engaged and represented.

Inclusive participation requires moving beyond representation toward structured, meaningful engagement of all stakeholder groups. Governments should contribute by aligning national strategies with global frameworks and sharing regulatory experiences. International organizations can provide coordination, standards, and evidence-based guidance. The private sector should contribute technical expertise, innovation pathways, and responsible deployment models. Academia and research institutions can support with scientific validation, evaluation frameworks, and foresight analysis. Civil society plays a critical role in ensuring ethical oversight, inclusivity, and public trust. From an institutional perspective—particularly in sectors such as healthcare—operational entities should be actively engaged to provide real-world use cases, implementation feedback, and impact assessment. This ensures that governance is grounded in practice, not only policy. In terms of format and structure, the AI Dialogue would benefit from a multi-layered, action-oriented design: First, thematic working groups focused on priority sectors (e.g., healthcare, education, infrastructure), bringing together diverse stakeholders to develop practical recommendations and pilot initiatives. Second, regional tracks to ensure that local contexts, challenges, and priorities are reflected, particularly for developing countries. Third, public-private collaboration platforms to translate policy into implementation through joint projects, capacity-building programs, and technology transfer. Fourth, a continuous engagement model, combining periodic high-level meetings with ongoing virtual collaboration, knowledge-sharing platforms, and progress tracking mechanisms. Finally, the Dialogue should include clear outputs and accountability mechanisms, such as implementation roadmaps, measurable milestones, and regular reporting. This structure ensures that participation is not only inclusive, but also productive, sustained, and impact-driven, enabling all stakeholders to contribute effectively to global AI governance.

Inclusive participation requires moving beyond representation toward structured, meaningful engagement of all stakeholder groups. Governments should contribute by aligning national strategies with global frameworks and sharing regulatory experiences. International organizations can provide coordination, standards, and evidence-based guidance. The private sector should contribute technical expertise, innovation pathways, and responsible deployment models. Academia and research institutions can support with scientific validation, evaluation frameworks, and foresight analysis. Civil society plays a critical role in ensuring ethical oversight, inclusivity, and public trust. From an institutional perspective—particularly in sectors such as healthcare—operational entities should be actively engaged to provide real-world use cases, implementation feedback, and impact assessment. This ensures that governance is grounded in practice, not only policy. In terms of format and structure, the AI Dialogue would benefit from a multi-layered, action-oriented design: First, thematic working groups focused on priority sectors (e.g., healthcare, education, infrastructure), bringing together diverse stakeholders to develop practical recommendations and pilot initiatives. Second, regional tracks to ensure that local contexts, challenges, and priorities are reflected, particularly for developing countries. Third, public-private collaboration platforms to translate policy into implementation through joint projects, capacity-building programs, and technology transfer. Fourth, a continuous engagement model, combining periodic high-level meetings with ongoing virtual collaboration, knowledge-sharing platforms, and progress tracking mechanisms. Finally, the Dialogue should include clear outputs and accountability mechanisms, such as implementation roadmaps, measurable milestones, and regular reporting. This structure ensures that participation is not only inclusive, but also productive, sustained, and impact-driven, enabling all stakeholders to contribute effectively to global AI governance. Which voices, communities, or perspectives are currently underrepresented in global discussions on AI governance? How could they be included? (Max. 300 words) Proposed Answer (≤300 words): Global discussions on AI governance remain disproportionately shaped by developed economies, large technology companies, and policy institutions. Several critical voices and perspectives are still underrepresented. First, developing countries and regions, particularly in Africa and parts of the Middle East, are not sufficiently included in shaping governance frameworks. Their realities—limited infrastructure, different regulatory maturity, and unique societal needs—are often not fully reflected. Second, sector-level practitioners, especially from high-impact domains such as healthcare, education, and public services, are underrepresented. These stakeholders bring essential insights on real-world implementation, risks, and operational challenges that cannot be captured through policy discussions alone. Third, institutional and operational leadership from hospitals, training centers, and public institutions are often missing, despite being directly responsible for deploying AI systems and managing their impact. Fourth, end-user communities, including patients, vulnerable populations, and non-technical users, are not sufficiently engaged. Their perspectives are critical for ensuring that AI systems are inclusive, accessible, and aligned with societal values. To address these gaps, the AI Dialogue should adopt a structured inclusion approach: Establish regional representation mechanisms to ensure balanced geographic participation Create sector-specific engagement tracks to include practitioners and operational leaders Integrate user-centered consultation processes, including civil society and community representatives Support capacity-building programs to enable me

To foster meaningful and dynamic engagement, the AI Dialogue should adopt interactive, implementation-oriented formats that go beyond traditional panel discussions. First, scenario-based simulation sessions can be highly effective. Stakeholders from different sectors can engage in structured simulations of real-world AI governance challenges—such as data sharing in healthcare or cross-border AI deployment—allowing participants to explore decision-making, risks, and trade-offs in a practical setting. Second, co-creation labs (policy and implementation labs) should be introduced, where diverse stakeholders collaboratively design solutions, frameworks, or pilot projects. These labs can produce tangible outputs such as draft guidelines, operational models, or partnership concepts. Third, sector-specific roundtables (e.g., healthcare, education, infrastructure) can ensure focused, in-depth discussions that are directly linked to implementation. These should include both policymakers and practitioners to bridge the gap between strategy and execution. Fourth, innovation showcases and live demonstrations can allow institutions and companies to present real use cases of AI in action. This helps ground discussions in reality and facilitates knowledge exchange on what works and what does not. Fifth, interactive digital platforms should support continuous engagement before, during, and after the Dialogue. These platforms can enable live polling, collaborative drafting, knowledge sharing, and follow-up tracking. Finally, multi-stakeholder challenge sessions or "AI governance hackathons" can encourage rapid problem-solving around specific issues, generating innovative ideas and fostering collaboration across disciplines. By combining these formats, the AI Dialogue can become more participatory, solution-driven, and outcome-oriented, ensuring that engagement leads to concrete results and sustained collaboration rather than purely conceptual discussions.

Several policies and practices have demonstrated effective approaches to advancing responsible and practical AI governance. At the global level, UNESCO's Recommendation on the Ethics of AI provides a comprehensive, values-based framework that emphasizes human rights, transparency, and inclusivity. Similarly, the OECD AI Principles have been widely adopted and translated into national strategies, offering a flexible yet coherent foundation for governance. From a regulatory perspective, the European Union's AI Act represents a significant step toward a risk-based approach, categorizing AI systems based on their potential impact and applying proportionate requirements. This model provides clarity while allowing innovation to continue. In the healthcare sector, emerging AI governance frameworks for clinical use-such as guidelines on data governance, validation of AI models, and clinical safety protocols-offer practical examples of how to operationalize governance. Institutions are increasingly adopting structured processes for evaluating AI tools before deployment, including ethical review, performance validation, and continuous monitoring. Another effective practice is the development of multi-stakeholder platforms that bring together governments, academia, and industry to co-develop solutions. These platforms enable knowledge exchange, joint research, and pilot initiatives, ensuring that governance evolves alongside technological advancements. Additionally, capacity-building initiatives and simulation-based training approaches are proving valuable in preparing professionals to work with AI systems safely and effectively. These approaches help bridge the gap between policy and practice by enabling hands-on understanding of AI applications and risks. Overall, the most effective approaches share common characteristics: they are risk-based, inclusive, adaptable, and implementation-oriented. Combining global principles with sector-specific applications and continuous capacity development is essential to ensuring that AI governance is both responsible and actionable.