SHIFTERLABS

A successful first Global Dialogue on AI Governance should establish a shared global understanding that AI governance is not only a technical or regulatory challenge, but also an educational, institutional, and capacity-building challenge. From the perspective of ShifterLabs, an EdTech and AI education organization based in Ecuador, success would mean that the Dialogue recognizes AI literacy as a foundational condition for meaningful participation in AI governance. Countries, institutions, educators, workers, and communities cannot shape the future of AI if they do not have the capacity to understand, evaluate, and responsibly use AI systems. A strong outcome would be a co-chair summary that clearly identifies AI capacity-building, human oversight, transparency, institutional readiness, and culturally relevant education as essential pillars of responsible AI governance. The Dialogue should also highlight the needs of the Global South, where the AI divide is not only about infrastructure, but also about knowledge, language, access, and institutional preparedness. The first Dialogue should create a credible foundation for long-term cooperation among governments, UN agencies, academia, civil society, the private sector, and education-focused organizations. Its success should be measured by whether underrepresented regions and communities can see their realities reflected in the global AI governance agenda for 2027 and beyond.

• AI capacity-building
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Protection and promotion of human rights
• Transparency, accountability, and human oversight

We selected these priorities because inclusive AI governance depends on people and institutions having the capacity to understand, question, and use AI responsibly. AI capacity-building is essential for developing countries. The AI divide is not only a technological divide; it is also an educational and institutional divide. Without scalable AI literacy programs, many communities will remain consumers of AI systems designed elsewhere, rather than active participants in shaping their use and governance. The social, economic, ethical, cultural, linguistic, and technical implications of AI are also central. In Latin America and other regions of the Global South, AI governance must consider local languages, educational realities, cultural diversity, labor markets, and unequal access to digital infrastructure. We also selected protection and promotion of human rights because AI systems increasingly affect education, employment, public services, information access, and democratic participation. Governance frameworks must ensure that AI strengthens human dignity, inclusion, and agency. Finally, transparency, accountability, and human oversight are necessary to move from abstract principles to responsible implementation. However, these principles require practical literacy. People cannot exercise oversight over systems they do not understand. For this reason, ShifterLabs believes that AI literacy should be treated as public infrastructure for responsible AI governance.

Yes. One cross-cutting issue that deserves stronger attention is the relationship between AI governance and AI literacy. Many governance frameworks assume that individuals, educators, workers, institutions, and communities can understand AI systems well enough to make informed decisions. In practice, this is not always the case. Without broad AI literacy, transparency may not lead to real understanding, human oversight may become symbolic, and accountability mechanisms may remain inaccessible to the people most affected by AI systems. A second cross-cutting issue is institutional readiness. Schools, universities, companies, public agencies, and civil society organizations often want to adopt AI responsibly, but lack clear internal policies, training pathways, evaluation criteria, and governance protocols. The Dialogue should therefore consider how global AI governance can support practical institutional adoption frameworks, especially in developing countries. A third emerging issue is culturally and linguistically relevant AI capacity-building. Communities in Latin America, indigenous territories, rural regions, and low-resource educational contexts require approaches that respect local languages, cultural realities, and digital access conditions. Finally, mobile-first and low-bandwidth learning models should be considered part of the AI capacity-building agenda. In many developing contexts, the most scalable path to AI literacy may not be traditional platforms, but accessible systems that work through tools people already use, such as messaging applications.

In Ecuador and across Latin America, AI governance gaps are affecting education, work, public institutions, and small and medium-sized organizations in a very direct way. The main challenge is not only access to AI tools, but the lack of structured AI literacy, institutional policies, and practical adoption frameworks. Many educators, workers, and leaders are already using generative AI, but often without clear guidance on privacy, transparency, accountability, intellectual property, bias, human oversight, or responsible evaluation. This creates risks of superficial adoption, misinformation, dependency on external systems, and unequal access to productivity gains. In the education sector, the gap is especially urgent. Schools and universities need support to move beyond banning or informally using AI. They need policies, teacher training, authentic assessment models, and student-centered AI literacy programs that promote critical thinking, creativity, and responsible use. At the same time, the opportunity is significant. AI can help reduce educational gaps, improve productivity, expand access to high-quality learning, support multilingual and culturally relevant content, and strengthen institutional capacity. In regions such as the Amazon and other underserved areas, mobile-first and low-bandwidth learning models can make AI education more accessible. For ShifterLabs, the key opportunity is to transform AI governance into practical capacity-building: helping institutions adopt AI responsibly, training educators and workers, and creating learning infrastructure that allows communities in the Global South to participate actively in the AI transformation rather than only adapt to decisions made elsewhere.

The AI Dialogue can play a crucial role by becoming a universal space where international cooperation on AI governance is connected to the realities of all countries, not only those with advanced technological infrastructure. Its value is to bring governments, academia, civil society, the private sector, educators, and the technical community into the same conversation. This is especially important because AI governance requires coordination across many dimensions: technical standards, human rights, education, economic development, data governance, safety, institutional adoption, and public trust. The Dialogue can also help countries and stakeholders identify shared principles while respecting different levels of capacity, cultural contexts, and development priorities. For the Global South, this means ensuring that AI governance is not only about risk management, but also about access, opportunity, capacity-building, and meaningful participation. A key role of the Dialogue should be to translate high-level governance principles into practical pathways for implementation. This includes supporting AI literacy, institutional readiness, responsible adoption frameworks, and mechanisms for human oversight. The Dialogue should also serve as a bridge among existing initiatives, helping reduce fragmentation and build coherence across global, regional, national, and sector-specific AI governance efforts. Its greatest contribution would be to create a trusted and inclusive foundation for long-term cooperation, where developing countries can help shape the rules, capacities, and priorities of the AI era.

The AI Dialogue should build on existing UN and multistakeholder efforts, especially those that already connect AI governance with ethics, education, digital inclusion, and sustainable development. Important foundations include the Global Digital Compact, UNESCO's Recommendation on the Ethics of Artificial Intelligence, ITU's AI for Good platform, the work of the Independent International Scientific Panel on AI, and initiatives led by UNDP, OECD, G20, G7, regional organizations, academic networks, civil society, and the technical community. However, the added value of the AI Dialogue should not be to duplicate these efforts. Its role should be to connect them, identify common ground, and help make AI governance more inclusive, interoperable, and actionable. From the perspective of ShifterLabs, the Dialogue should also connect with education and capacity-building initiatives that are already working directly with teachers, students, public institutions, companies, and communities. AI governance will only become meaningful if it reaches the institutions and people who must implement it in practice. The Dialogue can add value by creating a global bridge between high-level governance frameworks and local implementation. It can help translate principles such as transparency, accountability, human rights, and human oversight into practical training models, institutional policies, and accessible learning infrastructure. This is especially important for developing countries, where AI readiness depends not only on regulation, but also on education, digital foundations, and trusted partnerships.

Different stakeholders can contribute by bringing complementary forms of knowledge to the AI Dialogue. Governments can provide policy priorities, regulatory needs, and national development perspectives. Academia and the technical community can contribute evidence, research, evaluation methods, and scientific clarity. Civil society can represent affected communities, rights-based concerns, and inclusion priorities. The private sector can share implementation experience, innovation capacity, and lessons from responsible deployment. Education-focused organizations can help translate AI governance principles into practical capacity-building. To make this participation meaningful, the format should avoid being limited to formal statements. The Dialogue should combine plenary sessions with thematic working groups, regional consultations, practical case studies, and implementation-oriented roundtables. We recommend that each thematic cluster include space for both high-level governance discussion and practical implementation examples. For instance, AI capacity-building should include voices from education providers, teachers, workforce development organizations, and local innovation ecosystems. The structure should also include mechanisms for written input, virtual participation, multilingual access, and post-Dialogue follow-up. This is important for smaller organizations and stakeholders from developing countries who may not be able to participate in person. The Dialogue should be designed not only as a conference, but as a recurring learning and coordination mechanism where diverse stakeholders can help move from principles to practice.

Several voices remain underrepresented in global AI governance discussions. These include educators, students, youth, women in technology, indigenous communities, rural communities, small and medium-sized enterprises, public institutions with limited technical capacity, and organizations from the Global South working directly on digital inclusion and AI literacy. In many cases, the people most affected by AI systems are not the ones shaping the rules, standards, or narratives around them. This creates the risk that AI governance reflects the priorities of highly resourced countries and institutions, while overlooking local realities such as limited connectivity, language diversity, informal labor markets, fragile education systems, and unequal access to training. These voices could be included through regional consultations, multilingual submissions, travel support, virtual participation, community-based listening sessions, and partnerships with local education and civil society organizations. The Dialogue should also create specific spaces for educators and capacity-building organizations, because they are essential to preparing society for responsible AI adoption. From a Latin American perspective, it is especially important to include Amazonian, indigenous, rural, and low-resource educational contexts. Their participation would help ensure that AI governance is not only globally legitimate, but also locally meaningful and culturally grounded.

The AI Dialogue could foster more meaningful engagement by combining formal deliberation with participatory, practical, and learning-oriented formats. One useful format would be thematic implementation labs, where stakeholders work on concrete scenarios such as AI in education, public services, labor markets, health, or small businesses. These labs could identify governance gaps, practical risks, capacity-building needs, and examples of responsible adoption. Another format could be regional listening sessions before and after the Dialogue, especially in the Global South. These sessions would allow local stakeholders to explain how AI governance issues appear in their own contexts and would help connect global principles with local realities. The Dialogue could also include short case-study showcases from different regions, highlighting what is working, what is failing, and what support is needed. This would help the conversation move beyond abstract principles. To support inclusive participation, the Dialogue should offer multilingual virtual sessions, asynchronous written input, youth and educator roundtables, and digital collaboration spaces where stakeholders can continue contributing after the event. Finally, the Dialogue could create an open repository of practical governance resources, including policy templates, AI literacy frameworks, human oversight models, institutional readiness tools, and examples of capacity-building programs. This would make the Dialogue not only a forum for discussion, but also a source of practical support for implementation.

One promising approach is to treat AI literacy as a practical foundation for AI governance. Governance becomes more effective when people and institutions understand how AI systems work, what their limitations are, and how to use them responsibly. At ShifterLabs, we are developing this approach through AI literacy frameworks, educator training, corporate AI adoption programs, and ShiftLoop, a WhatsApp-native microlearning platform designed to reduce friction and expand access to high-quality learning. This model is especially relevant for developing countries because it works through a tool people already use, supports mobile-first learning, and can deliver short lessons, quizzes, progress tracking, and digital certification without requiring complex infrastructure. Good AI governance practices should include institutional AI policies, teacher and workforce training, privacy and transparency guidelines, authentic assessment models, human oversight protocols, and clear criteria for responsible adoption. In education, for example, institutions need to move from informal or reactive AI use toward structured policies that protect students while helping them develop critical, creative, and responsible AI skills. The AI Dialogue could promote concrete solutions by encouraging open repositories of policy templates, AI literacy frameworks, institutional readiness tools, and practical case studies from different regions. It should also support partnerships among governments, universities, civil society, technology providers, and education-focused organizations. Effective AI governance should not remain only at the level of principles. It should become usable, teachable, measurable, and accessible to communities and institutions, especially in the Global South.