Telecel Ghana

Success should not be measured by how well we define AI risks, but by who gains the power to shape AI after the dialogue. From where I stand, working in tech in Ghana, the biggest gap is not awareness. It is access. Access to infrastructure. Access to data. Access to decision-making tables. Therefore, I think a successful dialogue should shift that. First, it should move African private sector actors from "participants" to "co-builders" of governance. Not consulted after decisions are made but involved at the point where standards are defined. Second, it should unlock real capacity. Not promises. Actual pipelines. Funding for compute. Support for local datasets. Partnerships that allow African developers to build, test, and deploy AI systems that solve local problems. Third, it should make safety practical. Governance should not become a barrier that only large companies can afford to comply with. It should be designed in a way that smaller and growing tech ecosystems can realistically implement. Finally, success means continuity. A system where feedback from regions like Africa actively shapes how governance evolves. To me I would say, if nothing changes on the ground after the dialogue, then it has not succeeded.

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

Safe, secure and trustworthy AI matters because trust is fragile. In my work in Quality Assurance, I have seen how small system flaws can break user confidence. At scale, that risk is even higher with AI. AI capacity-building is not optional for Africa. It determines whether we remain users of imported systems or become creators of our own. Without infrastructure, skills, and support, governance conversations will not translate into real participation. The social and cultural implications are real in our context. AI systems trained on non-African data often fail silently here. They do not understand our languages, our patterns, or our realities. That is not just a technical issue. It is an inclusion issue. Open-source and open models are practical solutions. They reduce dependency and allow local developers to experiment, adapt, and build relevant tools. For many African innovators, openness is the only viable entry point into AI development.

Safe and trustworthy AI is important because people need to trust the systems they use. From my work in Quality Assurance, I have seen how small errors can break user confidence. With AI, the impact can be much bigger. AI capacity-building is very important for Africa. It will decide if we only use tools built by others or if we can build our own. Without skills, infrastructure, and support, we cannot fully take part in AI development. The social and cultural side of AI also matters. Many AI systems are trained on data that does not reflect African realities. Because of this, they may not understand our languages or how we live. This is not just a technical problem. It is an inclusion problem. Open-source tools and open models help solve this. They make it easier for developers to learn, build, and create solutions that fit local needs. For many people in Africa, this is the easiest way to get started with AI.

I think AI governance gaps are already affecting how AI is used and developed in Africa, especially in countries like Ghana. One big challenge I see is limited capacity. Many organizations want to use AI, but they don't have the skills, infrastructure, or access to good data. At the same time, global AI rules are getting more complex. This creates a situation where African companies are expected to follow standards; they aren't fully ready for. Another challenge I notice is that AI systems often perform poorly in local contexts. Many models are trained on non-African data, so they struggle with local languages and real-life conditions. This causes errors, reduces trust, and slows adoption. I also think dependency is a problem. Most AI tools and platforms are built outside Africa, which makes local companies rely on systems they don't control. This affects cost, access, and innovation in the long run. But I also see opportunities. AI can solve real problems in agriculture, healthcare, and financial services. With the right support, local developers can build solutions that actually fit our context. Open-source tools are another opportunity. They let individuals and startups learn, experiment, and build without high costs. Finally, I think growing awareness of ethical AI gives Africa a chance to shape governance in a way that reflects our realities. If these gaps are addressed, I believe Africa can move from just using AI to actually building and contributing to it.

I think the AI Dialogue can be the bridge that finally connects different regions in a practical way. Right now, most global AI rules are made by countries with advanced tech ecosystems, and many African voices are left out. I see the Dialogue as a chance to change that so African companies and developers are not just followers but active contributors. I also think it can help create shared standards that make AI safer and more accessible worldwide. For example, if developers in Ghana, Kenya, and Nigeria can agree on certain principles with partners in Europe and Asia, it becomes easier to build systems that work across borders without compromising safety or fairness. Another role I see is fostering real collaboration, not just discussion. This could be sharing knowledge, open data, and best practices between countries, or creating joint projects that help developing regions catch up in skills and infrastructure. Finally, I think the Dialogue can help build trust. When countries and companies work together transparently, everyone can feel more confident using AI systems, and emerging economies can grow their own AI capacity instead of being left behind. In short, I see it as a platform that can turns international cooperation from a vague idea into real action connecting people, resources, and knowledge across borders.

I think the Dialogue will be a success if it does more than talk about AI rules and principles. I see real impact happening when African voices are at the table as co-creators, not just observers. Success for me would mean agreements and frameworks that African private sector actors can actually use to build and govern AI locally. I also think actionable support is crucial training programs, funding, and infrastructure that allow companies and developers in Africa to create their own AI solutions. Governance without the ability to act on it is just paper. Another thing I see as important is accountability. AI systems are powerful, and mistakes can affect lives. There needs to be clarity about who is responsible, and those mechanisms must be accessible to people and organizations everywhere. Finally, I think a lasting platform for collaboration is essential. AI changes fast. Governance cannot be a one-time conversation. The Dialogue should create a system where knowledge, lessons, and standards evolve together with technology. For me, success is when this dialogue turns into real action on the ground.

I think every stakeholder being it governments, private sector, academia, civil society, and local communities has a role to play, but the Dialogue should make participation practical, not just symbolic. For private sector actors like me, I see the value in sharing what works on the ground: testing AI systems, understanding local challenges, and pointing out where regulations may create barriers. Governments can share policy goals and enforcement mechanisms, while academics can contribute research, evidence, and frameworks. Civil society can highlight social, ethical, and inclusion issues. In terms of format, I think a mix of plenaries and smaller working groups works best. Plenaries can set the big picture and global standards, while breakout groups allow different regions, sectors, or expertise areas to dig into real challenges and co-create solutions. I also think a rotating leadership or co-chair model could help ensure no single region dominates the conversation. The structure should include ongoing feedback loops. Stakeholders should be able to follow up after the dialogue, report on what works, and update policies and guidelines. I see this as turning a one-time event into a living process that actually changes how AI is built and governed globally.

I think one of the biggest gaps right now is the voice of African developers and small tech companies. Many global discussions are dominated by wealthy countries and large tech corporations, which makes it easy to overlook the realities of emerging markets. Other underrepresented voices include local language experts, women in tech, rural communities, and youth innovators. These groups experience AI differently, but their perspectives are rarely reflected in policy. I think they could be included by intentionally inviting them into working groups, offering translation support, and providing access to online sessions. Funding small participants to join and actively encouraging representation in panels would also help. I also think using surveys or local hackathons as input into the Dialogue could give these communities a way to share ideas without needing a formal seat at the table.

I think the Dialogue should move beyond just panels and presentations. Interactive formats work best. For example, hands-on workshops where participants test AI systems and explore governance challenges can make discussions more concrete. Hackathons or co-creation labs could let different stakeholders, especially developers in emerging markets, work together to solve governance challenges in real time. I also see value in digital platforms where stakeholders can continuously contribute ideas, vote on priorities, or share case studies before, during, and after the Dialogue. This creates a living, dynamic conversation instead of a one-off meeting. Finally, I think storytelling can be powerful. Hearing real experiences from local developers, users, and communities can ground abstract policies in reality and make participants understand the human impact of AI decisions.

I think the AI Dialogue should build on existing initiatives like UNESCO's AI Ethics recommendations, the African Union's digital transformation strategies, and global open-source data platforms. These are creating standards and sharing knowledge but often in separate silos. The Dialogue could connect these efforts and make them practical for emerging markets. For example, it could help African tech companies access training programs, datasets, and regulatory guidance from multiple sources in one place. This turns fragmented efforts into a coordinated network. Another opportunity is to create partnerships beyond government and academia. Private sector actors like startups often struggle to participate because the mechanisms are too complex or expensive. The Dialogue could open the door for these voices, which is critical if AI is going to work for all regions. Finally, I think this Dialogue to be treated like a living feedback loop. Instead of setting rules once, it could collect lessons from real-world deployment in different countries and continuously improve international AI standards. Cooperation becomes real action, not just words.