Elite Yassmin Cooperative (Natural Cosmetic & Aromatic Products Initiative)

A successful first Global Dialogue on AI Governance should go beyond general discussions and lead to concrete, practical outcomes. To me, one of the most important indicators of success would be creating a shared understanding of AI governance that truly reflects different global realities. Today, many frameworks are shaped by a limited number of regions, while others especially Africa are still underrepresented. A one-size-fits-all approach simply doesn't work in such a diverse global context. Another key outcome would be stronger commitments around AI capacity-building. Many countries still lack the technical expertise, infrastructure, and policy readiness needed to actively participate in both AI development and governance. Supporting these areas is essential if we want more balanced global participation. It would also be important to ensure that this Dialogue is not a one-time conversation. There should be clear mechanisms for ongoing, meaningful collaboration between governments, civil society, academia, and technical experts not just participation, but real involvement in shaping decisions. In addition, more attention should be given to data representation. If AI systems are trained on limited or unbalanced datasets, they risk reinforcing bias and excluding entire regions. Improving data inclusivity should be part of the global agenda. Finally, the Dialogue should lead to practical guidance whether in the form of principles or a roadmap that promotes transparency, accountability, and human-centered AI, while still allowing flexibility for different regional contexts. In the end, success would mean moving from discussion to coordinated global action that is inclusive, realistic, and sustainable.

• AI capacity-building
• Transparency, accountability, and human oversight
• Protection and promotion of human rights
• Safe, secure and trustworthy AI

The priorities I selected reflect what I believe is the biggest challenge today: making sure AI develops in a way that is both innovative and responsible, while also being inclusive. First, AI capacity-building is essential, especially for developing regions. Many countries still don't have the resources, infrastructure, or expertise needed to fully participate in AI not just as users, but as contributors to how it is governed. Without this, global discussions risk becoming unbalanced. Transparency, accountability, and human oversight are also critical. As AI systems are increasingly used in decision-making, people need to understand how they work and trust that there is still human control behind them. Otherwise, it becomes difficult to prevent misuse or unintended consequences. Human rights should remain at the center of any AI governance effort. If not carefully managed, AI can reinforce existing inequalities or create new ones. Making sure these systems respect fundamental rights is not optional it's essential. Finally, safe and trustworthy AI is key for long-term adoption. People will only accept and rely on AI technologies if they feel confident that these systems are secure, reliable, and aligned with societal values. Overall, these priorities are closely connected. Together, they help create a more balanced and realistic approach to AI governance one that works not just for a few countries, but globally.

One issue I think doesn't get enough attention is who actually gets to shape AI governance. A lot of discussions focus on principles like safety or transparency, which are important, but less attention is given to participation itself. Many regions, especially in the Global South, are still not meaningfully involved in these conversations. For me, this is not just a gap it affects the legitimacy of the entire governance process. Another point that feels important is data. AI systems depend heavily on data, but not all regions are equally represented. This can lead to systems that don't reflect local realities or, in some cases, reinforce bias. I think there needs to be more focus on how data is collected, who owns it, and how it represents different communities. I also believe we should look more closely at the link between AI and economic opportunity. AI can create growth, but it can also deepen inequalities if access to technology and infrastructure is uneven. This is something that deserves more attention in global discussions. Finally, I feel that youth perspectives are still not fully integrated into AI governance. Young people will be among those most affected by these technologies, yet they are often not part of the decision-making spaces. Including them is not just symbolic it's necessary if we want governance that is relevant for the future.

From my perspective, the current gaps in AI governance are already having visible effects, especially in regions like Africa. One of the biggest challenges is limited capacity. Many countries are still in the early stages of understanding and regulating AI, which makes it difficult to keep up with how fast the technology is evolving. This creates a situation where AI systems are adopted without clear frameworks to guide their use or manage risks. Another challenge is related to data. A large number of AI systems are developed using datasets that do not reflect local contexts. As a result, these systems can be less accurate or even biased when applied in African environments. This raises concerns not only about effectiveness, but also about fairness. There is also a gap in terms of participation. Most global discussions on AI governance are still dominated by a few regions, which means that local perspectives are not always taken into account. This can lead to policies that don't fully align with regional needs or priorities. At the same time, there are real opportunities. AI has the potential to support development in areas like education, healthcare, and entrepreneurship. There is also growing interest among young people to learn and engage with these technologies, which is a strong advantage for the future. If governance gaps are addressed, especially through capacity-building, better representation, and more inclusive data practices, regions like Africa could move from simply adopting AI to actively shaping how it is developed and used.

I think the AI Dialogue can play an important role by creating a space where different regions and stakeholders can actually learn from each other, not just share positions. Right now, a lot of AI governance discussions are happening in parallel, often led by a small number of countries or institutions. The Dialogue could help connect these efforts and make them more coherent. Instead of having fragmented approaches, it could encourage more alignment while still respecting different regional contexts. It can also help make cooperation more practical. For example, countries that are more advanced in AI could share not only high-level principles, but also real experiences what worked, what didn't, and what challenges they faced. This kind of exchange can be very valuable for countries that are still developing their approaches. Another important role is making sure that cooperation is inclusive. It's not just about governments civil society, young people, and smaller actors should also have space to contribute. If the Dialogue manages to create that kind of environment, it can make global governance discussions feel more balanced and representative. In the end, I see the AI Dialogue as a bridge not just between countries, but between different levels of experience and different perspectives. If it succeeds in doing that, it can help move from isolated efforts toward more coordinated and meaningful international cooperation.

There are already several important initiatives in AI governance, such as UNESCO's work on AI ethics, the OECD AI Principles, and different regional strategies being developed in Europe and other parts of the world. These efforts have helped establish useful foundations, especially around principles like transparency, accountability, and human rights. At the same time, many of these initiatives remain somewhat disconnected from each other, and their implementation varies a lot depending on the region. This is where the AI Dialogue could add real value. I think one of its strengths could be acting as a space that connects these existing efforts, rather than creating something entirely new. By bringing different initiatives together, it can help identify common ground, share practical experiences, and reduce duplication. Another added value is inclusion. Some regions, particularly in the Global South, are not always fully represented in existing frameworks. The Dialogue could help amplify these perspectives and ensure that global governance reflects a wider range of realities. It could also focus more on practical cooperation not just principles, but how to apply them. For example, supporting capacity-building, sharing tools or best practices, and encouraging collaboration across regions. Overall, the Dialogue has the potential to turn existing fragmented efforts into something more connected, inclusive, and actionable.

I think different stakeholders can contribute in very complementary ways, but the structure of the Dialogue needs to make space for all of them to be heard meaningfully. Governments can share policy approaches and regulatory experiences, while the private sector can bring practical insights from developing and deploying AI systems. Academia and the technical community can help clarify complex issues and provide evidence-based perspectives. Civil society, on the other hand, plays an important role in highlighting real-world impacts, especially on communities that are often overlooked. To make this effective, the Dialogue should avoid being too formal or one-directional. Instead of only high-level panels, it could include smaller, more interactive sessions where participants can exchange ideas more openly. For example, roundtables or breakout discussions could allow for more honest and detailed conversations. It would also help to create continuity beyond the event itself. The Dialogue could include follow-up mechanisms, such as working groups or ongoing consultations, so that discussions lead to concrete progress over time. Overall, the structure should make participation feel real, not symbolic, where different voices are not just present, but actually influence the conversation.

In my view, several important voices are still underrepresented in global AI governance discussions. First, perspectives from the Global South, particularly Africa, are not always sufficiently included. Even when these regions are part of the conversation, their role is often limited, and their specific challenges are not always fully reflected in the outcomes. Youth voices are another example. Young people are among those most affected by AI in the long term, but they are rarely involved in shaping the policies that will impact their future. Local communities and grassroots innovators are also often missing. Many discussions happen at a high level, while the people directly experiencing the effects of AI whether positive or negative are not always part of the conversation. To improve inclusion, there needs to be more intentional outreach and support. This could include providing access to resources, reducing barriers to participation, and creating spaces where these voices feel comfortable contributing. It's not only about inviting them, but about making sure their input is taken seriously. A more inclusive approach would make AI governance more relevant, balanced, and effective.

To encourage more meaningful engagement, I think it's important to move beyond traditional formats like panels and formal speeches. More interactive formats could make a big difference. For example, small group discussions or breakout sessions can create space for more open and honest exchanges. In these settings, people are often more comfortable sharing their perspectives, especially those who might not speak in larger forums. Another idea could be to include scenario-based discussions or case studies. Instead of talking only in general terms, participants could explore real situations and discuss how different governance approaches would apply. This makes the conversation more practical and easier to relate to. It could also be interesting to include mixed stakeholder sessions, where people from different backgrounds for example, policymakers, developers, and civil society work together on a shared topic. This can help bridge gaps in understanding and create more balanced discussions. Finally, allowing for both in-person and online participation is important to make the Dialogue more accessible. Overall, the goal should be to create an environment where people are not just listening, but actively engaging and learning from each other.

There are several existing approaches that offer useful lessons for effective AI governance, even if no single model is perfect on its own. For example, frameworks like UNESCO's Recommendation on the Ethics of AI and the OECD AI Principles have helped establish common reference points, especially around transparency, accountability, and human rights. What makes them valuable is that they provide guidance without being too rigid, allowing countries to adapt them to their own contexts. In practice, I think some of the most effective approaches are those that combine principles with implementation. For instance, regulatory sandboxes where new AI systems can be tested in controlled environments allow innovation while still managing risks. This kind of approach feels more realistic than trying to regulate everything in advance. Multi-stakeholder collaboration is also an important practice. When governments, private companies, researchers, and civil society work together, the outcomes tend to be more balanced and better aligned with real-world needs. Another area that deserves attention is capacity-building initiatives. Training programs, open educational resources, and knowledge-sharing platforms can help more countries and communities engage with AI governance, rather than just observe it. Overall, what stands out to me is that effective AI governance is not based on one single solution. It's more about combining flexible principles, practical tools, and inclusive collaboration to respond to a rapidly evolving technology.