Moi University

A successful first Global Dialogue on AI Governance would be one that delivers practical, inclusive, and enforceable outcomes, particularly for regions like Kenya and the broader Global South. From this perspective, success would mean establishing clear, globally recognized ethical AI principles that are adaptable to local contexts, while also addressing persistent challenges such as bias, data privacy and accountability. Equally important is regional inclusivity, ensuring that African nations and other underrepresented regions are not just participants, but active contributors in shaping AI governance frameworks. This includes capacity-building initiatives, knowledge transfer and equitable access to AI infrastructure and research opportunities. Another key outcome would be the development of policy guidelines that are accessible and understandable, not only to policymakers but also to the general public. A layman-friendly approach is critical to demystify AI and encourage public trust and engagement. Furthermore, the dialogue should result in strong commitments from governments to adopt and implement ethical AI policies, supported by funding, education and institutional backing. This includes sensitizing leaders across Africa on the long-term societal and economic implications of AI. Ultimately, success would be measured by the creation of a collaborative global framework that balances innovation with responsibility, ensuring that no region is left behind in the AI revolution while safeguarding human rights and promoting sustainable development.

• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Transparency, accountability, and human oversight
• Protection and promotion of human rights

From the perspective of Moi University, the following four thematic areas are the most urgent priorities for action and engagement: 1. Safe, secure and trustworthy AI Ensuring AI systems are safe and reliable is critical, especially in developing contexts where regulatory frameworks are still evolving. Moi University can contribute through research, testing, and validation of AI systems to ensure they are robust, secure, and suitable for local deployment. 2. Social, economic, ethical, cultural, linguistic and technical implications of AI AI must be contextualized to local realities. In Kenya and across Africa, issues such as language diversity, cultural representation, and economic inequality are key. The university can play a role in developing inclusive AI solutions that reflect local needs and values while addressing ethical concerns such as bias and fairness. 3. Protection and promotion of human rights AI systems must uphold fundamental rights, including privacy, non-discrimination, and access to information. Moi University can support this through interdisciplinary research, policy advocacy, and training programs that emphasize rights-based approaches to AI development and deployment. 4. Transparency, accountability, and human oversight Building trust in AI requires systems that are explainable and accountable. The university can contribute by advancing research in explainable AI, promoting ethical standards, and training students and professionals to ensure responsible human oversight in AI systems. These priorities align with Moi University's role in advancing ethical, inclusive, and impactful AI for sustainable development in Kenya and the wider Global South.

Yes, an important cross-cutting issue not properly captured in the listed themes is the sustainability and environmental impact of AI. AI systems, particularly large-scale models and data centers, require significant computational power, which translates into high energy consumption and carbon emissions. For countries in the Global South, including Kenya, this raises concerns about the long-term environmental costs of adopting AI technologies, especially where energy resources may already be limited or reliant on non-renewable sources. Addressing sustainability is therefore essential to ensure that AI development does not come at the expense of environmental degradation or climate goals. In addition, there is a need to promote green AI practices, such as energy-efficient algorithms, sustainable data infrastructure, and the use of renewable energy sources in powering AI systems. This is particularly relevant for institutions like Moi University, which can contribute through research and innovation in low-resource and energy-efficient AI solutions tailored to local contexts. Sustainability also intersects with issues of equity and access. If AI development continues to be resource-intensive, it risks widening the gap between well-resourced regions and those with limited infrastructure. Therefore, integrating environmental considerations into AI governance frameworks is crucial for ensuring both responsible innovation and global inclusivity. Incorporating sustainability as a core pillar would strengthen the AI governance agenda by aligning technological advancement with environmental responsibility and long-term development goals.

Governance gaps in AI are already having visible effects in Kenya, particularly in relation to safety, ethics, human rights and accountability. One of the most pressing challenges is the rise of misinformation and deepfakes, where AI is used to generate false images, videos and publications involving government officials and the public. This undermines trust in digital information ecosystems and highlights the urgent need for safe, secure and trustworthy AI frameworks. Ethical concerns are also significant. The rapid adoption of AI technologies has outpaced the development of clear ethical guidelines, leading to issues such as biased algorithms, misuse of personal data and lack of informed consent. In the Kenyan context, where digital literacy levels vary, this increases the risk of exploitation and reinforces existing inequalities. From a human rights perspective, gaps in governance expose citizens to risks related to privacy violations, surveillance and discrimination, especially in sectors such as finance, healthcare and public services. The lack of transparency and accountability mechanisms further complicates efforts to hold developers and deployers of AI systems responsible. However, these challenges also present important opportunities. Institutions like Moi University are well-positioned to lead in research, capacity building and policy development for responsible AI. By fostering interdisciplinary collaboration, the university can contribute to the development of locally relevant ethical frameworks, promote digital literacy and support innovation that aligns with Kenya's development priorities.

The AI Dialogue can play a critical role in advancing international cooperation on AI governance by serving as a platform for harmonizing policies, sharing knowledge and fostering inclusive participation across regions. One of its most important contributions would be the development of coherent and interoperable global policy frameworks that guide the safe, ethical and responsible use of AI while allowing flexibility for local adaptation. A key priority should be ensuring that marginalized regions, particularly in the Global South, are actively included in shaping these frameworks. This can be achieved through targeted support such as funding, subsidies and capacity-building initiatives that enable countries like Kenya to develop infrastructure, research capabilities and regulatory systems. Without such support, there is a real risk of widening the digital and technological divide. The Dialogue can also facilitate technology transfer and collaborative research, allowing institutions such as Moi University to partner with global stakeholders in developing context-aware AI solutions. This would not only strengthen local innovation ecosystems but also ensure that AI systems are more representative and equitable. It can promote shared standards for transparency, accountability, and ethical AI, helping to build trust across borders and reduce the risks associated with unregulated AI deployment. By encouraging open dialogue between governments, academia and industry, the initiative can bridge gaps in understanding and align global efforts.

The AI Dialogue should build upon existing global and regional AI initiatives to create a more cohesive and inclusive governance ecosystem. Key initiatives include: 1. UNESCO's Recommendation on the Ethics of AI - Provides a framework for ethical AI development and human rights protection. The Dialogue can extend this by promoting implementation support, especially in underrepresented regions like Africa. 2. OECD AI Principles - Offers guidance on trustworthy AI, transparency, and accountability. The Dialogue could help contextualize these principles for developing countries and support adoption through regional partnerships. 3. Global Partnership on Artificial Intelligence (GPAI) - Focused on collaborative AI research and policy. The AI Dialogue could leverage this network to include universities and research institutions from the Global South, enabling knowledge exchange and joint innovation. 4. Regional initiatives such as the African Union's AI Strategy - Focused on AI capacity-building and regulation. The Dialogue could connect these efforts with global frameworks, ensuring local priorities are recognized in international governance discussions. Added value of the AI Dialogue: 1.Inclusivity: Bringing marginalized regions and underrepresented voices into decision-making processes. 2.Policy support: Helping governments implement ethical AI, backed by funding, training, and technical guidance. 3.Collaboration: Encouraging partnerships between academia, industry, and policymakers to co-develop AI solutions tailored to local needs. 4.Standardization and trust: Promoting consistent governance practices and transparency mechanisms that increase public trust and responsible AI adoption globally.

Academia, including institutions like Moi University, can contribute through research, data analysis and evidence-based recommendations. Universities can provide insights into AI's social, ethical and technical implications, develop innovative solutions tailored to local contexts and support capacity-building initiatives for governments and industry. Government stakeholders can provide policy guidance, regulatory frameworks and incentives to promote safe and ethical AI adoption. They can also facilitate collaboration with regional partners, ensuring that governance structures reflect local needs while aligning with international standards. Industry can offer practical perspectives on deployment, scalability and commercial feasibility, while civil society can advocate for transparency, accountability and human rights protections. For the format and structure, the Dialogue should be multi-tiered and interactive, combining plenary sessions for policy discussions, technical workshops for research exchange, and breakout groups to address sector-specific issues. Inclusion of regional representation, multilingual platforms and virtual participation would enhance accessibility. Additionally, establishing follow-up mechanisms and reporting structures will ensure commitments are tracked and outcomes translated into actionable strategies globally. This approach ensures the AI Dialogue is collaborative, practical and inclusive, bridging gaps between innovation, governance and societal needs.

Currently, voices from the Global South, marginalized communities and underrepresented demographic groups like Maasais, are significantly underrepresented in global AI governance discussions. This has contributed to AI systems that often reflect racial, cultural and socio-economic biases, reinforcing inequities. For example, many AI models are trained on data from Europe or North America, leading to assumptions such as high car ownership or certain work habits, that do not hold true in contexts like Kenya. Such biases can result in AI solutions that are irrelevant, exclusionary, or even harmful in local settings. To address this, global AI governance must actively include researchers, policymakers and civil society from underrepresented regions, particularly the Global South. This could involve regional representation in international AI forums, funding for local AI research and collaborative projects that reflect diverse realities. Additionally, AI systems should be trained on inclusive, representative datasets that capture local languages, cultures and socio-economic contexts. By incorporating these perspectives, AI governance can become truly global, equitable and context-aware, ensuring that technological development benefits all populations rather than privileging a few regions or demographics. This inclusivity is crucial for ethical, safe and socially relevant AI deployment worldwide.

Hands-on workshops are highly effective, allowing participants to directly experiment with AI tools, understand their capabilities and limitations and explore practical solutions to context-specific challenges. These workshops can focus on ethical AI, inclusive data collection, or local problem-solving, providing tangible skills and knowledge. Hackathons and global AI challenges, such as the EY AI for Good Data Challenge, are another powerful format. They encourage cross-border collaboration, innovation and creative problem-solving, while actively including participants from underrepresented regions like Africa. These events generate actionable prototypes, foster networking and create a sense of global AI community. Panel discussions and roundtables can bring together policymakers, academics, industry leaders and civil society to debate critical issues such as bias, ethics, and regulation. When structured interactively, they allow participants to ask questions, co-create solutions and identify best practices. Virtual meetings and hybrid sessions ensure accessibility and inclusivity, enabling participants from remote regions, resource-limited institutions, and marginalized communities to engage fully without geographic constraints.

Policy frameworks such as UNESCO's Recommendation on the Ethics of AI and the OECD AI Principles establish clear guidelines on transparency, fairness, accountability and human rights, offering governments a foundation for ethical AI deployment. National AI strategies, like the African Union's AI Strategy, emphasize regional capacity building, ethical oversight and context-specific applications , ensuring local priorities are addressed. Practices such as explainable AI and algorithmic audits enhance transparency and trust, helping organizations detect bias and improve accountability. Similarly, inclusive data collection ensures AI models are representative and reduce systemic inequalities. Platforms and approaches like the Global Partnership on AI and AI4Good challenges provide collaborative spaces for research, innovation and knowledge exchange, enabling diverse stakeholders, including universities like Moi University, to co-develop AI solutions aligned with societal needs.