Artefcat Côte d'Ivoire

A successful first Global Dialogue on AI Governance would be defined by concrete, credible steps toward coordination, trust, and accountability. First, a key outcome would be the establishment of a shared baseline of principles; success would mean convergence on a minimal set of actionable commitments such as transparency standards for advanced AI systems, risk classification frameworks, and mechanisms for independent auditing. These should be adaptable across different regulatory contexts, including both developed and emerging economies. Second, the dialogue should produce a roadmap for international coordination by creating or strengthening a multilateral platform, formal or informal, where governments, industry, and civil society can continuously exchange information. A commitment to interoperability between regional regulations would be a strong signal of progress. Third, a successful outcome would ensure that countries from the Global South are not just participants but also agenda-setters. This could take the form of capacity-building initiatives, technology transfer partnerships, or funding mechanisms to support local AI governance and innovation ecosystems. Fourth, the dialogue should deliver at least one pilot initiative. For example, a cross-border AI safety evaluation program or a shared repository of high-risk AI use cases. Concrete pilots demonstrate seriousness and help avoid the perception of "talk shops." Finally, success would also be measured by trust-building. If stakeholders leave with clearer expectations, reduced uncertainty, and a willingness to collaborate despite geopolitical tensions, the dialogue would have achieved something meaningful.

• AI capacity-building
• Open-source software, open data and open AI models
• Social, economic, ethical, cultural, linguistic and technical implications of AI

AI capacity-building is a priority because access to AI is still very uneven. Many countries do not have the resources, skills, or infrastructure to develop or regulate AI properly. Supporting training, education, and knowledge sharing is important so that more countries can take part in shaping and using AI, rather than just following others. The social, economic, ethical, cultural, linguistic, and technical implications of AI also matter a lot. AI can affect jobs, reinforce biases, and influence how cultures and languages are represented online. It is important to pay attention to these impacts to make sure AI systems are fair, inclusive, and adapted to different societies. Compatibility of governance approaches is another key issue. Right now, different regions are creating their own AI rules, which can lead to confusion and fragmentation. Encouraging alignment between these approaches can make regulation more effective and avoid unnecessary barriers, while still allowing countries to keep their own priorities. Finally, open-source software, open data, and open AI models are important for widening access to AI. Openness allows more people to experiment, innovate, and understand how systems work. It helps smaller actors, such as researchers or startups, to contribute. At the same time, this openness should be managed carefully to limit risks of misuse.

One key gap is the environmental impact of AI. Training and deploying large models require significant energy, water, and hardware resources. As AI adoption grows, its carbon footprint and resource consumption will become major concerns, especially for countries already facing climate and energy challenges. This issue deserves more explicit attention within global AI governance discussions. Data governance across borders is also an emerging challenge. Questions around data sovereignty, cross-border data flows, and fair access to high-quality datasets are becoming more complex. Without clearer frameworks, there is a risk of tension between openness, privacy, and national interests.

In Africa, gaps in AI governance and uneven progress in key areas create both challenges and opportunities. A major challenge is the lack of capacity and infrastructure. Many countries still have limited access to computing power, quality data, and trained professionals. This makes it harder to develop local AI solutions and increases reliance on foreign technologies, which are not always adapted to local needs, especially when it comes to languages or social contexts. There is also an issue with coordination between countries. AI policies are developing at different speeds across the continent, which can lead to fragmentation. This makes regional collaboration more difficult and can slow down investment and innovation. On the social side, AI can reinforce existing inequalities if it is not well managed. For example, biased data can lead to unfair outcomes in areas like hiring or access to services. Africa's cultural and linguistic diversity is often not well represented in AI systems, which can exclude certain groups. At the same time, there are real opportunities. AI can help address key challenges in sectors like agriculture, healthcare, education, and finance. With the right support, African countries can build solutions that are adapted to their own contexts. The growth of open-source tools and open data is also a big opportunity. It allows more people, including startups and researchers, to work on AI without needing huge resources. Finally, stronger cooperation at the regional level, for example through the African Union, could help countries share knowledge, align their approaches, and have more influence in global discussions on AI. Overall, while there are clear gaps, there is also strong potential if the right investments and collaborations are put in place.

The AI Dialogue can play a useful role by simply getting the right people to talk to each other regularly and in a constructive way. Today, many countries are working on AI rules on their own. The Dialogue can help connect these efforts by creating a space where governments, companies, researchers, and civil society can share what they are doing, what works, and what does not. This can save time and avoid repeating the same mistakes. It can also help reduce fragmentation. Right now, AI governance is moving in different directions depending on the region. The Dialogue can encourage some level of alignment, or at least make systems more compatible. This makes cooperation easier and avoids creating barriers between countries. Another important role is to give more voice to countries that are often less represented, especially in Africa and other parts of the Global South. If these countries are included from the start, the rules and standards that emerge will be more balanced and realistic. The Dialogue can also push for concrete collaboration, not just discussions. For example, countries could work together on AI safety, share certain types of data responsibly, or develop common evaluation methods. Even small joint projects can build momentum. Finally, it helps build trust. AI is a sensitive topic, and there is a lot of competition between countries. Having a space to talk openly can reduce tensions and make cooperation more likely over time. In short, the Dialogue can act as a meeting point that makes cooperation more natural, more inclusive, and a bit more practical.

Organizations like the OECD and its AI Principles, or UNESCO's work on AI ethics, have already created widely recognized frameworks. The Global Partnership on Artificial Intelligence also brings together governments and experts to work on practical AI policy issues. In addition, regional regulations such as the EU AI Act are shaping concrete approaches to risk-based governance. There are also technical and multi-stakeholder initiatives, such as the Partnership on AI, as well as growing communities around open-source AI and research collaboration. The value of the AI Dialogue would be to connect these efforts rather than duplicate them. Today, many of these initiatives operate in parallel, with limited coordination. The Dialogue can act as a bridge, helping to align discussions, share outcomes, and avoid fragmentation. It can also bring in missing voices, especially from developing regions that are not always fully represented in existing forums. By doing so, it can make global AI governance more inclusive and balanced. Another added value is the ability to link high-level principles with practical action. While many frameworks already exist, implementation remains uneven. The Dialogue can encourage concrete cooperation, such as joint projects, shared standards, or capacity-building programs. Finally, it can provide a neutral and continuous space for exchange, helping to build trust between actors with different interests. In short, the AI Dialogue can strengthen what already exists by improving coordination, inclusiveness, and practical impact at the global level.

Governments have a key role since they design and implement policies. They can share their regulatory approaches, priorities, and the challenges they face. At the same time, since the private sector is often more advanced in developing and deploying AI, governments should also take this opportunity to learn from companies. Businesses deal with real systems, real risks, and real constraints on a daily basis, so their feedback can help make policies more realistic and effective. Researchers can provide more neutral and long-term perspectives, while civil society can ensure that social impacts, fairness, and rights are properly considered.

Many countries from the Global South, especially in Africa, are still under-represented. They are sometimes invited, but they don't always have a real influence on the discussions. Yet their context is very different, and their input is important if we want global rules that actually work everywhere. Then there are everyday users. Most conversations happen between governments, big companies, and experts, but the people who are directly affected by AI are rarely heard. Whether it's students, workers, or small business owners, their experience could make the discussions much more concrete. Startups and SMEs are also often missing. Large tech companies are well represented, but smaller actors face different challenges and can bring fresh ideas. Their perspective is important for a more balanced view of innovation and regulation.