LocaleNLP

A successful first Global Dialogue on AI Governance would establish a truly inclusive, action-oriented global framework that reflects the priorities of all nations especially those historically underrepresented in AI development. Success means moving beyond discussion toward practical commitments that expand equitable access to AI infrastructure, data, talent, and governance participation. Key outcomes should include: • A shared roadmap for inclusive, interoperable AI governance. • Concrete commitments to bridge AI capacity gaps in developing countries. • Recognition of linguistic diversity and the urgent need to support low-resource languages in AI systems. • Mechanisms to promote open, safe, and trustworthy AI models accessible to researchers, startups, and public institutions globally. • Stronger international cooperation on standards, transparency, and responsible innovation. For Africa and other emerging regions, success also means ensuring AI governance addresses digital sovereignty, local data ownership, and equitable participation in the global AI economy. The Dialogue should create pathways for countries not only to regulate AI, but also to build, deploy, and benefit from it. Ultimately, the first Dialogue should lay the foundation for a global AI ecosystem where innovation is shared, governance is collaborative, and AI serves humanity across all languages, cultures, and communities.

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

These priorities are deeply interconnected and essential for ensuring AI benefits all of humanity. AI capacity-building is critical because many developing countries lack the infrastructure, talent, and computing resources necessary to participate meaningfully in the AI economy. Without deliberate investment, the global AI divide will continue to widen. The social, cultural, linguistic, and economic implications of AI are especially important for regions like Africa, where thousands of languages and diverse cultural systems remain underrepresented in current AI models. Inclusive AI must reflect humanity's full linguistic and cultural richness. Safe, secure, and trustworthy AI is fundamental to building public confidence and ensuring AI systems are reliable, fair, and aligned with societal values. Finally, open-source software, open data, and open AI models are vital for democratizing innovation. Open ecosystems enable researchers, startups, governments, and civil society, particularly in emerging markets, to build locally relevant AI solutions without prohibitive barriers to entry. Together, these priorities can help create an AI future that is equitable, participatory, and globally representative.

Yes. Several critical cross-cutting issues deserve greater attention. First, linguistic inclusion and language equity should be explicitly recognized. AI systems today disproportionately serve high-resource languages, leaving billions of people excluded from the benefits of AI. Language is foundational to participation, access, and digital rights. Second, digital sovereignty and local data governance are increasingly important. Countries need the ability to govern, own, and benefit from their data, models, and AI infrastructure while maintaining alignment with global standards. Third, environmentally sustainable AI must be prioritized. As AI adoption accelerates, its energy consumption and carbon footprint are becoming significant concerns. Efficient, low-power, and edge-based AI systems are essential for sustainable and inclusive deployment. Finally, representation of low-resource communities, including indigenous peoples, rural populations, and speakers of underrepresented languages, must be embedded across all governance discussions. AI governance should not only protect these communities from harm but also actively empower them as co-creators of the AI future.

In Africa, governance gaps in AI are creating both significant risks and missed opportunities. The absence of inclusive AI frameworks has resulted in limited representation of African languages, cultures, and contexts in global AI systems. This perpetuates digital exclusion, restricts access to AI-powered services, and limits local innovation. For startups like LocaleNLP, the lack of open datasets, compute infrastructure, and harmonized policy frameworks creates barriers to developing AI solutions tailored to local needs. Many African innovators face challenges accessing funding, high-performance computing resources, and regulatory clarity. At the same time, these gaps present enormous opportunities. Africa can help shape a more inclusive AI future by developing context-specific governance models rooted in equity, linguistic diversity, and public benefit. AI has transformative potential across education, healthcare, agriculture, financial inclusion, and public service delivery. With the right governance structures, Africa can move from being a consumer of imported AI technologies to becoming a creator of globally relevant, locally grounded AI innovation.

The AI Dialogue can serve as the world's most inclusive platform for multilateral AI cooperation. Its greatest value lies in creating a universal forum where all countries, not only technologically advanced nations, can shape the future of AI governance. It can help align global principles, facilitate knowledge exchange, and promote interoperability across national and regional governance frameworks. The Dialogue can also foster collaboration on shared challenges, including safety standards, capacity-building, open-source development, and responsible innovation. Importantly, it can amplify the voices of emerging economies, ensuring that AI governance reflects diverse realities, priorities, and development needs. By connecting governments, industry, academia, and civil society, the Dialogue can bridge policy gaps and catalyze practical partnerships. Its ultimate role should be to transform fragmented AI governance efforts into a coordinated global ecosystem that is inclusive, transparent, and future-ready.

The AI Dialogue should build upon existing global and regional initiatives, including UNESCO's Recommendation on the Ethics of AI, the Global Digital Compact, the OECD AI Principles, the African Union AI Strategy, and the ITU's AI for Good platform. It should also connect with open-source communities such as Hugging Face, Mozilla, and LAION, as well as multistakeholder initiatives like the Global Partnership on AI (GPAI). Regional innovation ecosystems, research networks, and startup communities across Africa, Latin America, and Asia should also be integrated. The added value of the AI Dialogue lies in its universal legitimacy and convening power. Unlike existing initiatives, it provides a truly global forum under the United Nations where all countries and stakeholders can collaborate on equal footing. It can serve as the connective tissue between technical, policy, and implementation efforts, translating principles into coordinated action and ensuring that no region is left behind.

A successful AI Dialogue must be genuinely multistakeholder by design. Governments should provide policy leadership and ensure alignment with public interest. The private sector can contribute technical expertise, innovation, and practical implementation insights. Academia and the technical community can offer evidence-based research, standards development, and independent evaluation. Civil society can ensure that human rights, inclusion, and accountability remain central. The Dialogue should include plenary sessions, thematic working groups, regional consultations, and year-round virtual engagement. Dedicated tracks for youth, startups, indigenous communities, and underrepresented regions would strengthen inclusivity. Outputs should be action-oriented, including policy recommendations, partnership frameworks, and implementation roadmaps. Participation should be accessible both physically and virtually, with multilingual support to ensure global engagement.

Underrepresented voices in global AI governance include African innovators, indigenous communities, low-resource language speakers, rural populations, women, youth, persons with disabilities, and small startups from emerging markets. These groups are often most affected by AI systems yet least represented in their design and governance. Their exclusion risks reinforcing existing inequalities and embedding systemic bias into global AI ecosystems. Inclusion requires intentional design: funding support for participation, multilingual engagement, regional consultations, fellowship programs, and formal representation mechanisms. Community-led consultations and local ecosystem partnerships are essential. AI governance must be built not merely for the communities it aims to serve.

To foster meaningful engagement, the AI Dialogue should adopt participatory and interactive formats beyond traditional plenaries. These could include: • Regional innovation showcases and policy labs • AI governance simulation exercises • Multistakeholder roundtables and design sprints • Startup and research demonstration sessions • Citizen assemblies and youth forums • Collaborative policy hackathons • Scenario planning workshops Digital participation tools, real-time polling, multilingual AI-enabled translation, and hybrid engagement platforms can further expand access and inclusivity. These formats would encourage practical problem-solving, deepen collaboration, and ensure the Dialogue remains dynamic, accessible, and action-oriented.

Promising examples include UNESCO's Recommendation on the Ethics of AI, which provides a globally endorsed normative framework, and the African Union's Continental AI Strategy, which emphasizes inclusion, capacity-building, and local innovation. Open-source initiatives such as Hugging Face and Mozilla Common Voice demonstrate how collaborative data and model development can democratize AI access. Federated learning and privacy-preserving AI approaches offer practical solutions for balancing innovation with data protection. At LocaleNLP, we are advancing a model of inclusive AI governance through community-based data collection, ethical language dataset creation, and energy-efficient AI systems designed for low-resource environments. Our work ensures that underrepresented languages and communities are not merely consumers of AI but active contributors to its development. Effective AI governance must combine ethical principles, open innovation, local ownership, and practical implementation pathways that empower all regions to participate in the AI future. https://localenlp.org/