Digital Cooperation Organization

A successful first Global Dialogue should deliver four core outcomes. First, it should ground discussions in evidence-based insights reflecting real-world readiness and gaps. Data-driven frameworks such as the Digital Economy Navigator (DEN) 2025 (https://den.dco.org/) show that AI development is uneven, with significant disparities in governance capacity, institutional readiness, and technical capabilities. Anchoring the Dialogue in such evidence would enable more targeted, differentiated, and outcome-oriented cooperation. Second, it should produce a shared, prioritized operational understanding of immediate cooperation areas, including human-rights-based safeguards, interoperability across governance approaches, and practical capacity-building for countries at different levels of AI readiness. Success would mean translating inclusive dialogue into a concrete, implementable agenda. Third, it should identify a practical cooperation architecture that enables implementation rather than merely restating principles. This should include a light, action-oriented roadmap linking policy guidance, capacity-building, knowledge-sharing, and voluntary tools. Practical, country-tested instruments are essential to demonstrate implementation. For example, the DCO AI-REAL Toolkit (https://ai-real.dco.org), including the AI-REAL Adoption Playbook, provides an end-to-end approach—from assessing national AI readiness across governance, infrastructure, Data, and skills, to guiding tailored adoption pathways. Similarly, the DCO Ethical AI initiative (https://dco.org/initiatives/ethical-ai/) integrates the DCO Principles for Ethical AI with operational tools such as the DCO AI Ethics Evaluator, alongside policy guidance and capacity-building resources. Together, these approaches demonstrate how global standards can be translated into actionable strategies and measurable progress. Fourth, the Dialogue should lay the groundwork for progressively more coordinated global approaches to managing advanced AI risks, including exploring common guardrails, shared safety expectations, and mechanisms for collective oversight. This would help move toward greater international alignment on high-impact risks, while remaining inclusive, flexible, and responsive to different national contexts.

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

These priorities reflect DCO's mandate, practical experience across Member States, and evidence from the Digital Economy Navigator (DEN), which highlights uneven AI development and persistent gaps in capacity, governance, and implementation. Effective AI governance must therefore move beyond normative frameworks toward capability, coordination, and operational mechanisms. AI capacity-building is a primary priority. Countries are at different stages of readiness, and governance ambitions cannot be realized without institutional capability, technical skills, and implementation support. The DEN underscores gaps in skills and public sector capacity. DCO addresses this through the DCO AI-REAL Toolkit, which enables countries to assess readiness across governance, Data, infrastructure, and skills, and translate insights into targeted actions and implementation roadmaps, complemented by Ethical AI governance and capacity-building programs. Interoperability of governance approaches is critical and closely linked to data governance as the foundation of AI systems. Diverging regulatory models-particularly around data access, protection, and cross-border flows-create barriers to innovation and cooperation. The DCO Interoperability Mechanism for Cross-Border Data Flows, combining Privacy Principles, Model Contractual Clauses, and accreditation models, demonstrates how trusted data governance can enable cross-border AI ecosystems. Protection and promotion of human rights remain central. DCO's Ethical AI initiative is grounded in a human-rights-based approach, operationalized through the DCO Principles for Ethical AI, the DCO AI Ethics Evaluator, and the "Rights by Design" report. Safe, secure, and trustworthy AI is essential to address systemic risks. DCO's integrated approach promotes risk-aware adoption, system reliability, and institutional safeguards, while underscoring the need for greater international alignment on managing high-impact AI risks through shared expectations and coordinated oversight. Overall, these priorities reflect a shift toward AI governance as a systemic challenge requiring sustained investment, international coordination, and practical implementation tools.

Several cross-cutting issues merit greater attention. First, implementation readiness should be treated as a distinct governance priority. Many countries do not lack principles; they lack institutional capacity, technical expertise, policy coordination, and governance tools to operationalize them. The DCO AI-REAL Toolkit was developed precisely to address these gaps, providing structured diagnostics, maturity benchmarking, and actionable pathways from policy to execution. Second, governance of cross-border AI ecosystems requires more explicit focus. AI increasingly depends on data flows, cloud infrastructure, compute access, and global supply chains. National approaches alone are insufficient. DCO's work emphasizes interoperable frameworks that enable trusted cross-border data and technology flows, linking AI governance with practical mechanisms such as interoperable data governance and alignment with global standards. Third, public-sector deployment and procurement governance should be elevated. Governments are becoming major AI adopters, yet governance discussions often remain abstract. Practical questions around procurement, auditing, and oversight are critical. DCO's implementation-oriented resources-such as the Ethical AI Guidebook for Policymakers, the AI-REAL Adoption Playbook, and the DCO AI Ethics Evaluator-support governments in translating strategy into accountable, real-world deployment. Fourth, equitable participation in AI governance and standard-setting remains underemphasized. Without stronger inclusion of developing countries, global frameworks risk reinforcing existing asymmetries. This aligns with DCO's focus on enabling countries at different levels of readiness to participate meaningfully in the digital economy. Finally, there is a growing need for progressively coordinated global approaches to managing advanced AI risks, including the development of shared guardrails and common expectations. Evidence from the Digital Economy Navigator (DEN) reinforces that gaps in governance, capacity, and infrastructure persist, underscoring the importance of moving toward more structured and collective mechanisms that translate principles into consistent and effective global practice.

For DCO Member States and similarly placed digital economies, the central challenge is the asymmetry between the pace of AI deployment and the maturity of governance and implementation capacity. DCO research highlights uneven AI readiness and the coexistence of divergent governance models, creating uncertainty for policymakers and businesses. This manifests in several ways. First, gaps in readiness and limited assurance mechanisms constrain the ability to ensure safe, secure, and trustworthy AI, particularly in high-impact applications. Second, capacity constraints—including skills, institutional capability, and public sector readiness—slow adoption across key sectors such as government services, education, and healthcare. Third, fragmented governance approaches complicate cross-border investment, trade, and digital cooperation. Fourth, insufficiently operationalized safeguards increase risks related to bias, privacy, and weak accountability. The Digital Economy Navigator (DEN) reinforces these challenges. It shows low overall AI maturity (average score of 44.3), significant governance gaps (AI governance scoring 30.6), and a high concentration of research and compute capacity in advanced economies. This concentration risks deepening global disparities, limiting many countries' ability to develop, deploy, and govern AI independently, and increasing reliance on external technologies. At the same time, there is a strategic opportunity to build governance by design. DCO's work—including the "Rights by Design" report, the DCO Ethical AI Guidebook for Policymakers, the DCO AI Ethics Evaluator, and the DCO AI-REAL Toolkit and AI-REAL Adoption Playbook—demonstrates how countries can move from principles to implementation through structured, practical approaches. These dynamics also point to the need for more coordinated global approaches to managing advanced AI risks, including shared guardrails and mechanisms for collective oversight. Strengthening cooperation in this direction—while addressing capacity and compute asymmetries—will be critical to ensuring inclusive, secure, and trustworthy AI ecosystems.

The AI Dialogue can play a distinctive role as an inclusive multilateral bridge across fragmented AI governance efforts. First, it can create a common platform for convergence, where governments and stakeholders compare approaches, identify alignment areas, and reduce duplication across existing initiatives. This is critical in a landscape shaped by diverse principles, regulatory models, standards processes, and emerging discussions on more coordinated global approaches to managing advanced AI risks. Second, the Dialogue should go beyond policy exchange and serve as a platform for implementation-focused cooperation. This includes promoting practical support through toolkits, readiness assessments, capacity-building, and peer learning. Experience from DCO demonstrates that governments benefit most when governance is delivered as an operational toolbox—combining principles with practical instruments that enable real-world deployment. Third, the Dialogue can strengthen interoperability and global legitimacy. By ensuring broad participation—including developing countries, regional organizations, and diverse stakeholder groups—it can help avoid concentration of influence in a limited set of jurisdictions. This is essential to making AI governance globally applicable, trusted, and sustainable. Fourth, the Dialogue can contribute to the gradual development of shared guardrails and coordinated oversight mechanisms for high-impact AI risks. Without pre-empting formal processes, it can provide a space to explore common expectations, risk thresholds, and cooperation models that may, over time, inform more structured international approaches. Finally, it can serve as a bridge between science and policy, leveraging inputs from scientific bodies while anchoring outcomes in inclusive multistakeholder deliberation. Overall, the Dialogue's value lies in connecting fragmented efforts, advancing practical cooperation, and laying the foundation for more coherent and coordinated global AI governance.

The AI Dialogue should connect with existing global and regional efforts rather than replicate them. Relevant foundations include the UNESCO Recommendation on the Ethics of AI, which provides a global normative baseline centered on human rights, dignity, transparency, fairness, and human oversight; the OECD AI Principles, which remain a widely used intergovernmental reference point for trustworthy AI; and the Council of Europe Framework Convention on AI, which establishes a legally binding international instrument focused on human rights, democracy, and the rule of law. It should also connect with ITU's AI for Good as a major UN-linked convening platform linking standards, policy, and practical AI applications. From DCO's perspective, the Dialogue should also recognize and draw from DCO's multilateral implementation-oriented initiatives, including the DCO Principles for Ethical AI, the DCO AI Ethics Evaluator, the "Rights by Design" and "Responsible AI Governance" reports, the Ethical AI Guidebook for Policymakers, and the AI-REAL Toolkit, which provides an implementation layer by translating governance principles into AI readiness assessments, adoption pathways, and lifecycle guidance, bridging the gap between AI policy and real-world deployment. These instruments were designed as global public goods to help operationalize ethical AI, human-rights-based safeguards, and readiness-based AI adoption. The added value of the AI Dialogue would be to connect these tracks: norms, regulation, technical cooperation, and implementation support. Its comparative advantage is not to become another standalone framework, but to become the place where different initiatives are mapped, linked, and translated into more coherent international cooperation.

Different stakeholders should contribute according to their comparative strengths. • Governments should bring policy experience, current AI readiness, implementation needs, and lessons from national strategies, laws, procurement, and institutional design. • International Organisations should help map standards, frameworks, and capacity-building resources. • Industry should contribute to deployment experience, technical constraints, safety practices, and transparency lessons, development of AI tools, and customized AI roadmaps. • Academia and technical experts should provide evidence on impacts, evaluation methods, and future risks. • Civil society, human rights organizations, workers' groups, youth, and affected communities should bring lived experience, accountability demands, and inclusion perspectives. • Financial institutions should help define the roadmap for inclusivity, scalability and predictability of investments in AI, enabled by governance instruments. In terms of format, the AI Dialogue would benefit from a three-layer structure: 1. Plenary sessions to identify strategic priorities and points of convergence; 2. Thematic working sessions focused on concrete issues such as human rights, interoperability, capacity-building, and transparency/accountability; 3. Implementation clinics or solution labs where stakeholders present tools, case studies, and practical cooperation proposals. To improve continuity, the process should include written inputs before the meeting, structured synthesis outputs after the meeting, and intersessional workstreams between Geneva and New York. DCO's experience suggests that implementation-oriented exchanges are strongest when principles (e.g., DCO Principles for Ethical AI, DCO Privacy Principles tec.) are paired with tools (e.g., DCO Interoperability Mechanism, DCO AI Ethics Evaluator, DCO AI REAL Toolkit for AI readiness assessment and adoption support, etc.), guidance (e.g., DCO Ethical AI Policy Guidebook and DCO's AI Adoption Playbook) , and peer learning AI-enabled digital transformation opportunities (e.g., DCO IMPACT Platform). The AI-REAL Toolkit in particular can support structured stakeholder engagement by aligning readiness diagnostics with practical implementation pathways across different sectors and governance contexts.

Several perspectives remain underrepresented. Most notably, developing countries and smaller digital economies are still not represented at a level proportionate to the impact that global AI rules will have on them. This is especially important for countries that are not leading AI model development but will nevertheless bear major consequences in public services, labor markets, education, and digital trade. DCO's work reflects this reality by focusing on Member States with differing readiness levels and by building tools that support implementation, not just frontier regulation. Also underrepresented are public-sector implementers, including ministries, regulators, procurement officials, and service-delivery agencies; workers and labor voices affected by AI-enabled restructuring; youth and educators; linguistic and cultural communities outside dominant AI languages; and communities most affected by bias, exclusion, or weak digital protections. Inclusion should be improved through deliberate design: travel support and remote participation options; balanced speaking slots; regional consultations feeding into the global meeting; multilingual documentation; open calls for written submissions; and dedicated sessions for practitioners and affected communities, not only senior policymakers and major technology firms. The AI Dialogue should also value contributions in practical formats such as case studies, implementation notes, and policy tool demonstrations, and structured stakeholders' input that can be systematically synthesized into actionable policy insights.

The most effective formats would be those that move beyond standard panel discussions. One useful format would be solution labs where governments, International Organisations, and other stakeholders present concrete AI tools, policy templates, AI governance assessments, or AI adoption case studies. This would allow participants to compare what is actually usable, including instruments such as evaluator tools, readiness toolkits, and AI adoption and policy guidebooks. A second format would be scenario-based policy workshops. Participants could work through shared governance dilemmas, such as public-sector deployment of high-impact AI, cross-border data governance for AI systems, or accountability failures in automated decision-making. This produces deeper engagement than general discussion because it forces trade-offs, prioritization, and problem-solving. A third format would be multistakeholder implementation roundtables organized by theme, each tasked with identifying practical next steps, cooperation needs, and gaps in evidence or capacity. These could feed directly into an outcome summary for follow-up between 2026 and 2027. Finally, interactive demonstrations can be valuable when they are governance-oriented rather than purely promotional. For example, showcasing AI readiness and risk-assessment tools, assurance methods, red-teaming approaches, or capacity-building platforms can make discussions more concrete and useful for policymakers.

A useful example is the DCO Principles for Ethical AI, which provide a shared foundation for governance across Member States, covering accountability, transparency, fairness, privacy, sustainability, and human oversight. These principles promote alignment across jurisdictions and support interoperable governance frameworks. A second example is the DCO AI Ethics Evaluator, which translates principles into a practical, role-based assessment for developers and deployers. By identifying risks and recommending mitigation measures, it bridges the gap between ethical commitments and operational implementation, enabling consistent governance across the AI lifecycle. A third example is DCO's broader Ethical AI Governance toolbox, including the "Rights by Design" and "Responsible AI Governance" reports, and the DCO Ethical AI Guidebook for Policymakers. These resources support governments in embedding human-rights safeguards, strengthening policy coherence, and translating principles into legislation and practice. A fourth example is the DCO AI-REAL Toolkit, which integrates governance with readiness and implementation capacity. It enables countries to assess gaps, strengthen institutional preparedness, and define adoption pathways, complemented by the AI-REAL Adoption Playbook, which translates readiness insights into actionable strategies. At the global level, frameworks such as the UNESCO Recommendation on the Ethics of AI, the OECD AI Principles, and the Council of Europe Framework Convention reinforce convergence around core values while allowing contextual flexibility. The Digital Economy Navigator (DEN) complements these approaches by providing a data-driven assessment of AI maturity across countries, highlighting gaps in governance, infrastructure, and capacity, and enabling targeted interventions. Collectively, these approaches demonstrate how principles, tools, and evidence can be integrated into practical governance systems, while also contributing to emerging efforts toward greater international alignment on managing high-impact AI risks through shared expectations and coordinated approaches.