AI for ALL Inc.

Success, to me, is when this Dialogue moves from talking about AI governance to actually operationalizing it. We don't need another set of high-level principles—we need alignment that works in the real world. That means creating a shared baseline across countries on safety, transparency, and accountability, while still allowing for local context. Not identical systems, but interoperable ones. What would really make this meaningful is if we leave with a *living coordination layer*—something that continues beyond the event. Ongoing councils or working groups with clear mandates, timelines, and accountability. A system that connects governments, industry, academia, and civil society in a way that actually gets things done. I would also expect to see *real pilot initiatives emerge*—cross-border collaborations where governance is tested in action. Whether it's healthcare, climate, or education, we need to prove what works, refine it, and scale it. Governance should be experienced, not just theorized. Another critical outcome is the adoption of *decision intelligence tools*—the ability to simulate second- and third-order consequences before policies are implemented. If we can see the ripple effects in advance, we can make better, faster, and more trusted decisions. Inclusion is non-negotiable. The Global South, youth, and underrepresented voices must be part of shaping this. Otherwise, we risk building systems that don't reflect the world they're meant to serve. And finally, success looks like clear commitments—who is doing what, by when, and how progress will be measured. If we get this right, this isn't just a dialogue, it becomes a global execution engine for responsible AI. Done well, this Dialogue becomes the foundation for a new kind of global coordination system—one designed for the speed and complexity of AI.

• AI capacity-building
• Protection and promotion of human rights
• Open-source software, open data and open AI models
• Interoperability of governance approaches

My priorities reflect a shift from isolated efforts to coordinated global capability. AI capacity-building is foundational. Without broad access to knowledge, tools, and infrastructure, we risk concentrating power in a small number of regions and organizations. Capacity-building ensures that all countries and communities can meaningfully participate-not just as users of AI, but as contributors to its development and governance. Interoperability of governance approaches is critical. AI is inherently borderless, yet governance remains fragmented. We need frameworks that can work together across jurisdictions-aligned enough to collaborate, but flexible enough to respect local contexts. Interoperability is what enables real coordination-and the ability to test, adapt, and improve approaches across systems over time. Protection and promotion of human rights must remain central. As AI systems scale, they influence decision-making at every level of society. Embedding human rights into the design, deployment, and oversight of AI is essential to ensure these systems serve humanity equitably and responsibly. Finally, open-source software, open data, and open AI models are key to transparency, innovation, and trust. Open ecosystems accelerate progress, enable auditability, and allow a broader range of stakeholders to participate in shaping AI systems. Together, these priorities support a globally coordinated, inclusive, and actionable approach to AI governance-one that can evolve as quickly as the technologies it is designed to guide.

Yes; one of the most important cross-cutting gaps is the need for decision intelligence and coordination infrastructure. Much of the current conversation focuses on principles, risks, and domain-specific impacts. What's less developed is how we actually make better decisions, faster, across complex systems. AI governance is not only about regulating technology-it's about navigating interconnected outcomes across economies, societies, and environments. We need the ability to model, simulate, and understand second- and third-order consequences before policies and systems are deployed. A related emerging issue is coordination at scale. Governance today is fragmented across institutions, sectors, and geographies. Without shared mechanisms to align stakeholders-governments, industry, academia, and civil society-we risk duplication, delays, and unintended consequences. The challenge is not just setting rules, but enabling continuous, adaptive collaboration. Another gap is the concept of execution accountability. Many frameworks stop at recommendations, but do not define how actions are tracked, measured, and iterated over time. Governance systems need to evolve into living processes, where commitments are visible, progress is measurable, and learning is continuous. Finally, there is an emerging need to recognize AI as part of a broader system of global infrastructure, similar to financial systems or the internet. This requires thinking beyond isolated policies toward integrated, interoperable systems that can support real-time decision-making and coordination. Addressing these cross-cutting issues would help shift AI governance from static frameworks toward dynamic, adaptive systems capable of responding to the speed and complexity of change.

From my perspective, the governance gaps across capacity-building, interoperability, human rights, and open systems are already shaping outcomes in uneven ways across regions and sectors. Through my work with AIforAll, which operates globally, and my involvement with GAFAI.org, I see these dynamics play out across diverse countries, institutions, and stakeholder groups in real time. One of the most significant challenges is asymmetry in AI capacity. Organizations and countries with access to talent, infrastructure, and capital are advancing quickly, while others are left adopting systems they did not help shape. This creates dependency, limits local innovation, and risks reinforcing existing inequalities—particularly in emerging markets. A second challenge is fragmentation of governance approaches. Without interoperability, organizations operating across borders face conflicting standards, compliance uncertainty, and slower deployment cycles. This not only increases cost and complexity, but also discourages collaboration at the very moment it is most needed. There are also growing concerns around human rights and trust. As AI systems influence decisions in areas like hiring, finance, healthcare, and public services, gaps in transparency and oversight can lead to bias, exclusion, and reduced public confidence. Trust, once lost, is difficult to rebuild. At the same time, there are clear opportunities. Expanding AI capacity-building can unlock participation from new regions and communities, enabling more diverse perspectives and solutions. This is particularly important for addressing global challenges where local context matters. Advances in open-source models and open data are also accelerating innovation and access. They provide a foundation for transparency, collaboration, and shared progress, while reducing reliance on closed systems. Finally, improving interoperability creates the opportunity for more coordinated, cross-border initiatives—where governance approaches can be tested, refined, and scaled more effectively. Overall, these dynamics are pushing us toward a critical inflection point: either fragmentation deepens, or we intentionally build more inclusive, connected, and adaptive systems for AI governance.

The AI Dialogue has the potential to become more than a forum—it can act as a catalyst for coordinated global action. Its most important role is to help move international cooperation from fragmented conversations to aligned execution. Today, many countries and organizations are working on AI governance in parallel, but without sufficient connection. The Dialogue can serve as a bridging layer, bringing together governments, industry, academia, and civil society to align priorities, share learnings, and accelerate progress. It can also play a key role in advancing interoperability across governance approaches. By identifying where frameworks can align—and where flexibility is needed—the Dialogue can help reduce friction for cross-border collaboration and create the conditions for more seamless cooperation. Another critical function is to support the development of shared pilot initiatives. International cooperation becomes meaningful when it is grounded in real-world implementation. The Dialogue can help convene partners around specific use cases—such as healthcare, climate, or education—where governance approaches can be tested, refined, and scaled together. The Dialogue is also well positioned to elevate capacity-building efforts globally, ensuring that countries and communities are not left behind. This includes sharing tools, knowledge, and best practices, while supporting more inclusive participation in shaping AI systems. Finally, it can introduce greater continuity and accountability into global cooperation. By establishing ongoing working groups, clear commitments, and measurable outcomes, the Dialogue can evolve into a living coordination mechanism, not just a one-time exchange. If done well, the AI Dialogue becomes a platform for collective intelligence and coordinated action—helping the global community navigate the speed and complexity of AI together.

There are already a number of important initiatives shaping the AI governance landscape, and the opportunity for the AI Dialogue is not to duplicate them, but to connect, align, and activate them. Efforts such as global alliances, multilateral forums, academic networks, and industry-led initiatives are advancing work on standards, ethics, safety, and policy. Organizations like GAFAI.org, along with UN-led processes, OECD frameworks, and other cross-sector collaborations, are contributing valuable pieces of the puzzle. At the same time, many of these efforts operate in parallel, with limited coordination across domains and regions. The AI Dialogue can add value by acting as a coordination layer across these initiatives—a place where insights, frameworks, and progress can be brought together and translated into more unified action. Rather than creating new standards, it can help identify where alignment already exists and where interoperability can be strengthened. It can also serve as a convening mechanism for multi-stakeholder collaboration, bringing together actors who may not typically work closely—particularly across public, private, and civil society sectors. This is especially important for addressing complex, cross-border challenges that no single organization can solve alone. Another key contribution is the ability to move from dialogue to shared pilot implementations. By connecting existing initiatives to real-world use cases, the Dialogue can help test governance approaches in practice, generate evidence, and accelerate what works. Finally, the AI Dialogue can introduce greater visibility, continuity, and accountability across the ecosystem. By tracking commitments, sharing progress, and supporting ongoing collaboration, it can help ensure that efforts are not siloed or short-lived. In this way, the Dialogue becomes a unifying and activating force—turning a fragmented landscape into a more coordinated, effective global system for AI governance.

Inclusive participation needs to be designed, not assumed. The AI Dialogue can achieve this by clearly defining how each stakeholder contributes and by structuring the process around action, not just discussion. Governments can bring policy authority and implementation pathways. Their role is to share regulatory approaches, identify areas for alignment, and commit to pilot initiatives. Industry contributes technical expertise, infrastructure, and real-world deployment insights, along with the responsibility to operationalize governance standards. Academia and research institutions provide evidence, evaluation frameworks, and independent analysis to inform decisions and assess impact. Civil society and NGOs ensure that human rights, equity, and lived experience remain central, especially for underrepresented communities. Youth and Global South stakeholders should be included not as observers, but as co-creators, shaping priorities and solutions from the outset. In terms of format, the Dialogue should move beyond plenaries to a multi-layered structure: 1. Thematic working groups that operate on an ongoing basis, focused on priority areas such as interoperability and capacity-building, with clear mandates, timelines, and outputs. 2. Action-oriented roundtables that are small and curated to produce specific outcomes, such as partnerships, pilot designs, or policy alignment proposals. 3. Global pilot labs that support cross-border initiatives where governance approaches are tested in real-world contexts, with shared learning loops. 4. An open participation layer, supported by digital platforms, allowing for broader input, transparency, and continuous engagement beyond core participants. 5. An accountability mechanism that tracks commitments, progress, and outcomes to ensure continuity between Dialogues. This structure enables stakeholders to contribute based on their strengths, while ensuring the Dialogue becomes a living system for coordination, experimentation, and execution.

Several voices remain consistently underrepresented in global AI governance—and the data shows this is not marginal, it's structural. Women and gender-diverse individuals are still significantly underrepresented. Globally, women make up only about 22% of AI talent and less than 15% of senior leadership roles. This imbalance shapes not only who builds AI systems, but whose perspectives influence governance decisions. The Global Majority—including Africa, Latin America, parts of Asia, and small island states—also remains underrepresented in decision-making. AI governance is still largely shaped by Western nations and corporations, often sidelining local priorities and contexts. This creates a cycle where countries are governed by frameworks they did not help design. Interdisciplinary and non-technical voices, including educators, community leaders, and social sector practitioners, are often missing. Yet AI impacts society broadly, and limiting governance discussions to technical or policy elites reduces the relevance and effectiveness of outcomes. Communities most affected by AI systems—including racialized groups, people with disabilities, and low-income populations—are also underrepresented. Research continues to show that AI systems can disproportionately impact these groups due to biased data and limited representation in design and oversight. To address this, inclusion must be intentional. Participation should be resourced, not symbolic, through funded seats, regional representation targets, and capacity-building programs. Governance processes should include co-creation mechanisms, where underrepresented groups help shape agendas from the outset. Digital participation can broaden access, but must be paired with meaningful influence over decisions. Ultimately, inclusion is not just about fairness—it directly impacts the quality, legitimacy, and effectiveness of AI governance.

To foster meaningful and dynamic engagement, the AI Dialogue needs to move beyond static panels and into formats that are interactive, outcome-driven, and designed for real collaboration. One powerful approach is facilitated round tables—small, curated groups of diverse stakeholders brought together around a specific challenge or opportunity. When designed well, these sessions don't just surface ideas—they produce aligned next steps, partnerships, and shared accountability. The key is intentional composition and clear outcomes. Another format is the use of live scenario exploration. Rather than debating abstract policies, participants can work through real-world scenarios, testing how different governance approaches might play out across sectors or regions. This allows participants to see potential second- and third-order effects and creates a shared understanding of trade-offs. Cross-border pilot design sessions can also be highly effective. These are structured working sessions where participants co-create pilot initiatives in areas like healthcare, climate, or education, with a focus on implementation. The goal is to leave the Dialogue not just with ideas, but with initiatives ready to be tested and scaled. A continuous digital engagement layer is equally important. Not everyone can be in the room, and engagement shouldn't end when the event does. A platform that enables ongoing collaboration, matchmaking, and contribution allows the Dialogue to evolve over time and include a broader range of voices. Finally, introducing a light coordination and feedback loop—where insights, commitments, and progress are captured and revisited—helps ensure that engagement translates into action. The opportunity is to design the Dialogue as a living system, where people don't just participate—they connect, test, and build together in real time.

There are already strong examples of policies, practices, and platforms that point toward more effective and actionable AI governance. The opportunity now is to connect and build on what is working. At the policy level, risk-based approaches and accountability frameworks are gaining traction. These focus on how AI is used, not just how it is built, and require ongoing oversight, transparency, and clear responsibility for outcomes. This creates a more practical pathway for governance that can evolve with the technology. In practice, multi-stakeholder collaboration models are proving effective. When governments, industry, academia, and civil society are involved from the outset, governance becomes more balanced, trusted, and implementable. Open ecosystems are also playing an important role. Open-source models, shared datasets, and collaborative research environments are accelerating innovation while improving transparency and auditability. They allow a broader range of participants to contribute and reduce reliance on closed systems. From a platform perspective, there is growing value in tools that support coordination and decision-making. Platforms that enable stakeholders to map relationships, align around shared goals, and simulate potential outcomes before implementation are helping to move governance from static frameworks into something more dynamic and responsive. Another promising approach is the use of real-world pilot environments, where governance frameworks are tested in sectors such as healthcare, climate, or education. These pilots generate evidence, reveal gaps, and create feedback loops that strengthen policy over time. Overall, the most effective approaches move beyond theory and create systems that can be implemented, tested, and continuously improved in real-world conditions.