UNESCO Chair in Public Communication for Social Justice, Human Rights and Territorial Development (UNVM - Argentina)

A successful first Global Dialogue on AI Governance should deliver outcomes that are both substantively meaningful and actionable for public policy at the national and international levels. First, it should produce a shared vision for global AI governance that can effectively inform and be translated into public policies by Member States. This implies moving beyond general principles toward guidance that is adaptable to different national contexts, while maintaining common standards grounded in human rights, inclusion, and accountability. A key measure of success would be the Dialogue's ability to bridge global discussions with concrete policy implementation. Second, the Dialogue should explicitly address the societal risks associated with AI, particularly the amplification of disinformation and hate speech. It should promote coordinated approaches that combine regulatory frameworks, transparency requirements, and public interest safeguards, while respecting freedom of expression. Advancing international cooperation in this area is essential to mitigate harms that transcend national borders. Third, it should ensure that global governance efforts reflect diverse perspectives and actively reduce not only digital divides, but also cultural, linguistic, and epistemic gaps. This includes fostering multilingual AI, supporting locally grounded approaches, and strengthening the participation of underrepresented groups, including women and Indigenous peoples, in both governance and development processes. Fourth, the Dialogue should lead to concrete cooperation mechanisms, particularly in capacity-building. This includes AI literacy initiatives, infrastructure access, and support for knowledge-sharing networks such as academic and multistakeholder partnerships, enabling countries—especially in the Global South—to actively shape and implement AI governance. Finally, success would mean delivering tangible outputs—such as a policy-oriented roadmap or a set of actionable recommendations—that ensure continuity beyond the event. Ultimately, the Dialogue should lay the foundations for a more inclusive, coordinated, and effective global AI governance framework, capable of guiding public policy and addressing shared challenges in a rapidly evolving technological landscape.

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

My selection reflects a focus on ensuring that AI governance is both inclusive and actionable, particularly in contexts where inequalities in access, representation, and capacity remain significant. Safe, secure and trustworthy AI is a fundamental priority, as trust is a precondition for the meaningful adoption of AI systems. This requires not only technical robustness, but also mechanisms to address risks such as bias, discrimination, disinformation, and the amplification of harmful content, including hate speech. Building trustworthy AI must therefore integrate social and ethical safeguards alongside technical standards. AI capacity-building is essential to enable all countries-especially those in the Global South-to actively participate in and shape AI development and governance. This includes strengthening infrastructure, skills, and institutional capacities, as well as promoting AI literacy across societies. Without this, existing asymmetries risk deepening. The social, economic, ethical, cultural, linguistic and technical implications of AI are central to my priorities. AI systems are not neutral; they reflect and can reinforce existing power structures. Addressing these dimensions is key to ensuring that AI systems are context-sensitive, culturally diverse, and inclusive of different knowledge systems. Promoting multilingual and "situated" AI approaches is particularly important in this regard. Finally, open-source software, open data and open AI models are critical to democratizing access to AI and fostering innovation beyond a small number of dominant actors. Open ecosystems can support transparency, collaboration, and local adaptation, while also enabling more equitable participation in the AI landscape. Together, these priorities aim to advance a model of AI governance that is trustworthy, inclusive, and capable of being translated into concrete public policies and sustainable development outcomes.

Yes. While the listed themes are comprehensive, a key cross-cutting issue that deserves more explicit attention is "gender and diversity in AI governance and development". Although aspects of inclusion can be inferred within existing categories-such as human rights or social implications-the absence of a specific focus risks overlooking the structural nature of gender-based and intersectional inequalities in AI systems. There is growing evidence that AI applications can reproduce and even amplify existing biases, particularly affecting women and gender-diverse communities in areas such as hiring, access to services, content moderation, and representation. Addressing this challenge requires going beyond general principles and adopting targeted strategies. This includes promoting the participation of women and diverse groups in STEM fields, AI research, and decision-making processes, as well as supporting inclusive education and AI literacy policies from early stages. Without this, the design and governance of AI will continue to reflect limited perspectives, reinforcing existing power imbalances. In addition, incorporating gender-sensitive and intersectional approaches into data governance, model design, and evaluation processes is essential to identify and mitigate cognitive and algorithmic biases. This also implies strengthening accountability frameworks that can detect and address discriminatory outcomes. Making gender and diversity a more explicit priority would enhance the effectiveness of all thematic areas, contributing to more equitable, representative, and trustworthy AI systems. It would also ensure that AI governance frameworks are better aligned with broader commitments to equality and non-discrimination in the international system. In this sense, recognizing gender and diversity as a cross-cutting and strategic dimension is not only a matter of inclusion, but a necessary condition for developing fair and socially integrated AI.

In my country, current governance gaps in AI are shaped by a limited engagement of the State in advancing or adopting global AI governance frameworks. This creates a challenging context, where public policy development on AI remains fragmented and insufficient to address emerging risks and opportunities. In this scenario, one of the most significant challenges is the lack of coordinated strategies to address issues such as disinformation and the spread of hate speech, which are increasingly amplified through digital platforms and AI systems. At the same time, there are persistent inequalities in access to AI technologies and in the development of digital and AI-related skills, limiting the possibility of a truly inclusive and widespread adoption. These gaps are particularly visible in relation to capacity-building and the social and cultural dimensions of AI. Without strong public policies, there is a risk that existing structural inequalities—especially those linked to language, geography, and socioeconomic conditions—will deepen. The absence of a clear national framework also makes it more difficult to align with international standards on safe, trustworthy, and human-centered AI. At the same time, this context highlights important opportunities. Civil society and academia are playing a crucial role in advancing this agenda. Through empirical research and critical perspectives, they have consistently demonstrated the need for a more active engagement in global discussions on AI governance, as well as the importance of developing locally grounded and inclusive approaches. In particular, there is growing momentum around promoting AI literacy and digital education as a means to democratize access and empower communities. Strengthening these efforts, and ensuring that diverse stakeholders are meaningfully included in governance discussions, represents a key opportunity to bridge current gaps and move toward a more inclusive, coordinated, and globally connected approach to AI governance.

The AI Dialogue can play a critical role in advancing international cooperation by serving as a bridge between global principles and their effective implementation through public policies, with a strong focus on universal access and the protection of human rights. First, it can help build a shared framework that guides countries in developing AI policies aligned with international human rights standards. This includes promoting safeguards against discrimination, protecting freedom of expression while addressing disinformation and hate speech, and ensuring accountability and human oversight in AI systems. By fostering convergence around these principles, the Dialogue can strengthen coherence across national approaches. Second, the Dialogue can act as a catalyst for cooperation aimed at universalizing access to AI. This involves supporting capacity-building initiatives, advancing AI literacy, and promoting equitable access to infrastructure, data, and tools—particularly in developing countries. International cooperation is essential to avoid deepening existing inequalities and to ensure that all regions can benefit from and contribute to AI development. Third, it can elevate the role of civil society, academia, and other stakeholders in global governance discussions. In contexts where state engagement may be limited, these actors are key to advancing evidence-based debates, highlighting risks, and proposing inclusive and context-sensitive approaches. Finally, the Dialogue can facilitate the exchange of best practices and the development of collaborative initiatives—such as open and interoperable frameworks, shared resources, and multi-stakeholder partnerships—that support both innovation and rights-based governance. Overall, the AI Dialogue has the potential to strengthen international cooperation by promoting a more inclusive, rights-based, and development-oriented approach to AI governance, grounded in the principles of equity, participation, and universal access.

The AI Dialogue should build upon existing multilateral and multi-stakeholder initiatives that have already advanced normative frameworks, capacity-building, and inclusive approaches to AI governance—particularly those led by UNESCO. A key reference is UNESCO's Recommendation on the Ethics of Artificial Intelligence, which provides a globally agreed normative framework grounded in human rights, inclusion, and sustainability. In addition, UNESCO has developed tools such as readiness assessment methodologies, as well as initiatives on AI capacity-building and governance support for Member States. These efforts offer a strong foundation for translating global principles into national policies. The Dialogue could also connect with networks such as UNESCO Chairs and the UNITWIN programme, which bring together academic institutions worldwide to promote research, knowledge-sharing, and capacity-building with regional and cultural diversity. These platforms are particularly valuable for amplifying perspectives from the Global South and fostering interdisciplinary approaches. Beyond UNESCO, the Dialogue should engage with broader multi-stakeholder ecosystems, including academic research networks, civil society coalitions, and open-source communities working on AI transparency, accountability, and inclusive innovation. The added value of the AI Dialogue lies in its ability to connect these existing initiatives within a more cohesive and action-oriented global framework. It can serve as a coordination space to reduce fragmentation, align efforts across institutions, and promote interoperability between governance approaches. Importantly, the Dialogue can elevate these initiatives to a higher level of political visibility, facilitating their uptake in public policy and strengthening international cooperation. By linking normative frameworks with practical implementation and inclusive participation, it can help accelerate progress toward human rights-based, accessible, and globally coordinated AI governance.

Different stakeholders should contribute according to their comparative strengths, within a genuinely inclusive and balanced multi-stakeholder structure. Member States play a key role in translating global discussions into public policies and regulatory frameworks. The private sector contributes technical expertise and innovation capacity, while also bearing responsibility for ensuring transparency and accountability. Civil society is essential to bring rights-based perspectives, particularly regarding the societal impacts of AI, including disinformation and hate speech. Academia contributes independent, evidence-based research and critical analysis, helping to identify risks, evaluate impacts, and propose context-sensitive solutions. To enable meaningful participation, the Dialogue should adopt a structure that goes beyond formal statements. It should include interactive and deliberative formats, such as thematic working groups, roundtables, and collaborative sessions focused on concrete policy challenges. These spaces should aim to produce actionable outputs, such as policy recommendations or cooperative initiatives. Additionally, the Dialogue should ensure continuity through intersessional work, allowing stakeholders to remain engaged beyond the main event. Hybrid and multilingual formats are also essential to guarantee accessibility and broader participation. Overall, the structure should prioritize inclusiveness, practical cooperation, and policy relevance, ensuring that diverse stakeholders can actively shape outcomes.

Global discussions on AI governance continue to reflect significant imbalances in representation. Voices from the Global South remain underrepresented, particularly from regions with limited access to technological infrastructure and policy development resources. In addition, Indigenous peoples, local communities, and speakers of non-dominant languages are often excluded from shaping AI systems that directly affect them. Gender disparities also persist, with women underrepresented in both technical development and governance spaces. This lack of diversity risks reinforcing existing biases and limiting the ability to design inclusive and context-sensitive AI systems. To address these gaps, inclusion must move beyond symbolic participation. It requires targeted support for capacity-building, funding mechanisms to enable participation, and the adoption of multilingual approaches that reflect linguistic diversity. Promoting AI literacy and strengthening local research ecosystems are also critical to ensure that underrepresented communities can engage meaningfully. Existing networks, such as those supported by UNESCO (e.g., UNESCO Chairs and UNITWIN), can play an important role in facilitating more geographically and culturally diverse participation. Ultimately, inclusion should be embedded as a structural principle of the Dialogue, ensuring that diverse perspectives are not only heard, but effectively integrated into decision-making processes.

To foster meaningful and dynamic engagement, the AI Dialogue should incorporate innovative, participatory formats that move beyond traditional plenary discussions. One effective approach would be the use of thematic co-creation labs, where diverse stakeholders collaboratively develop policy proposals or solutions to specific challenges, such as regulating disinformation or expanding AI access. These spaces can encourage practical cooperation and produce tangible outputs. Another valuable format is scenario-based discussions, where participants explore the social, economic, and human rights implications of AI in different contexts. This can help bridge technical and policy perspectives, while grounding discussions in real-world challenges. The Dialogue could also include regional or community-led sessions, ensuring that local perspectives shape global debates. These sessions should be supported by multilingual facilitation and accessible digital tools to broaden participation. In addition, integrating hybrid and asynchronous participation mechanisms—such as online platforms for written inputs, consultations, and collaborative drafting—can ensure continuity and inclusiveness beyond live sessions. Finally, spaces specifically designed for underrepresented groups (e.g., youth, Indigenous communities, and grassroots organizations) can help amplify perspectives that are often excluded. Overall, innovative formats should prioritize interaction, co-creation, and inclusivity, enabling stakeholders not only to exchange views, but to collaboratively shape actionable outcomes.

Several existing policies, practices, and approaches offer valuable insights for advancing more inclusive and effective AI governance. A key example is the work developed by UNESCO on AI and Indigenous peoples, which highlights the importance of designing AI systems that respect cultural diversity, collective rights, and traditional knowledge. This approach promotes community participation, data sovereignty, and culturally grounded governance models, offering concrete guidance for more inclusive and context-sensitive AI. Another relevant initiative is the Feminist AI Manifesto developed by Latin American scholars and practitioners. This framework emphasizes the need to address structural inequalities embedded in AI systems, advocating for gender-sensitive design, intersectionality, and the redistribution of power in technological development. It provides both a critical lens and practical principles to guide more equitable AI governance. In addition, emerging research in regions such as Peru and across Africa illustrates innovative, locally grounded approaches. Efforts to recover and integrate Indigenous computational languages in Peru demonstrate how AI can support linguistic diversity and cultural preservation. Similarly, initiatives in Africa exploring smaller, context-specific programming languages and models offer alternatives to dominant large-scale AI systems, which often reproduce global asymmetries in knowledge production. These approaches prioritize efficiency, accessibility, and cultural relevance over scale. Together, these examples show that effective AI governance does not rely on a single universal model, but rather on plural, situated approaches that reflect diverse realities. The added value of these practices lies in their ability to combine technical innovation with social inclusion, human rights, and cultural diversity. Scaling and connecting these experiences through international cooperation can help build a more balanced and representative global AI ecosystem.