KAAD

A successful first Global Dialogue on AI Governance should move beyond broad principles and deliver concrete, equity-centered outcomes aligned with the priorities raised by KAAD at the UN Forum 2025 on Digital Justice. First, it should establish a shared framework for digital justice that explicitly recognizes structural inequalities in AI development and deployment. This includes acknowledging how data extraction, algorithmic bias, and unequal infrastructure disproportionately affect communities in the Global South, particularly Afro-descendant populations. Second, the Dialogue should produce actionable commitments on data governance—including fair data access, community consent, and mechanisms for value redistribution. Data must be treated not merely as a resource, but as a collective asset tied to rights, dignity, and sovereignty. Third, success would mean creating inclusive governance mechanisms. This implies institutionalizing the participation of civil society, grassroots organizations like KAAD, and underrepresented regions in decision-making processes—not as symbolic actors, but as co-authors of policy. Fourth, the Dialogue should advance capacity-building and knowledge transfer, ensuring that countries historically excluded from technological innovation can meaningfully shape and regulate AI systems. This includes funding, education, and infrastructure commitments. Finally, it should define accountability structures, with clear benchmarks, monitoring tools, and consequences for non-compliance. Without enforcement, principles risk remaining aspirational. In essence, success lies in shifting AI governance from a model driven by technological power to one rooted in justice, inclusion, and global equity—where those most affected by AI systems have the authority to shape their future.

Social, economic, ethical, cultural, linguistic and technical implications of AI;Protection and promotion of human rights

I prioritize the social, economic, ethical, cultural, linguistic and technical implications of AI because these dimensions capture how AI systems affect people's lived realities. In many contexts, particularly in the Global South and among Afro-descendant communities, AI can reinforce existing inequalities, through biased data, limited access to digital infrastructure, and the marginalization of local languages and cultural expressions. For me, addressing these implications is essential to ensure that AI contributes to inclusion rather than exclusion. I also prioritize the protection and promotion of human rights as a fundamental anchor for AI governance. A human rights-based approach provides a clear and universal framework to address risks such as discrimination, surveillance, and lack of accountability. It ensures that AI systems respect dignity, equality, and fundamental freedoms. Together, these priorities reflect my commitment to an approach that connects real-world impacts with enforceable principles. AI governance should not only focus on innovation, but also on justice, inclusion, and accountability, ensuring that those most affected by these technologies have their rights protected and their voices represented.

Yes. While the listed themes are essential, several cross-cutting and emerging issues remain insufficiently captured and deserve explicit attention. First, data justice and data sovereignty should be more clearly articulated. Beyond technical governance, there is a need to address who owns, controls, and benefits from data especially data originating from the Global South. Current AI systems often rely on extractive data practices that reproduce historical inequalities without fair compensation or consent from affected communities. Second, power asymmetries in AI development represent a critical structural issue. A small number of corporations and countries dominate AI infrastructure, research, and standard-setting. Without addressing this concentration of power, global governance risks reinforcing dependency rather than enabling equitable participation. Third, environmental and resource impacts of AI remain underexplored. The energy consumption of large-scale AI systems, as well as the extraction of raw materials needed for digital infrastructure, raises important questions about sustainability and environmental justice, particularly in regions already vulnerable to climate change. Fourth, cultural and linguistic diversity in AI systems requires more explicit recognition. Many languages and cultural contexts remain underrepresented in AI models, leading to exclusion and potential erasure. This is not only a technical issue but also one of identity, representation, and epistemic justice. Finally, there is a need to emphasize meaningful participation and co-governance. Beyond inclusion as a principle, mechanisms must ensure that civil society, grassroots organizations, and marginalized communities have real decision-making power in shaping AI policies. Addressing these cross-cutting issues would strengthen the Dialogue by ensuring that AI governance is not only comprehensive, but also structurally responsive to global inequalities.

Governance gaps in AI are already shaping outcomes in unequal ways across my areas of work, particularly in relation to communities in the Global South and Afro-descendant populations, as highlighted in discussions at the UN Forum 2025 on Digital Justice. One of the most significant challenges is the persistence of structural bias and exclusion. AI systems trained on non-representative datasets often reproduce racial, cultural, and linguistic biases, leading to misrepresentation or invisibility of certain communities. This is compounded by limited access to digital infrastructure, which restricts participation in both the development and governance of AI technologies. Another key challenge is the lack of accountability and enforceable human rights safeguards. In many contexts, there are insufficient regulatory frameworks to address harms such as algorithmic discrimination, surveillance, or misuse of personal data. This creates an environment where affected populations have little recourse. At the same time, there are important opportunities. AI governance, if shaped inclusively, can become a tool for empowerment and representation. It offers the possibility to elevate underrepresented languages, document cultural heritage, and improve access to services such as education and information. There is also an opportunity to advance more equitable global standards, where perspectives from the Global South influence how AI is designed and regulated. Initiatives that support capacity-building, knowledge-sharing, and local innovation can help shift current power imbalances. Overall, the impact of these governance gaps is significant: they risk deepening inequalities, but they also create a critical moment to advocate for a more just, inclusive, and rights-based approach to AI governance.

The AI Dialogue can play a pivotal role as a bridge between principles and coordinated global action. While many frameworks already exist, cooperation remains fragmented and often dominated by a limited number of actors. The Dialogue can help align these efforts around a shared vision of equitable and rights-based AI governance. First, it can serve as an inclusive multilateral platform where states, civil society, and actors from the Global South meaningfully participate in shaping norms and priorities. This is essential to ensure that international cooperation reflects diverse realities, not only those of technologically advanced countries. Second, the Dialogue can promote policy coherence and interoperability by facilitating convergence across regional and international approaches. By identifying common standards and areas of mutual recognition, it can reduce regulatory fragmentation while respecting local contexts. Third, it can advance collective accountability by encouraging the development of monitoring mechanisms, benchmarks, and reporting processes. This would help move from voluntary commitments to more structured forms of cooperation. Finally, the Dialogue can catalyze capacity-building partnerships, connecting resources, expertise, and institutions to support countries with limited technical and regulatory capacities. In this sense, it can transform cooperation into a tool for redistribution of knowledge and power. Overall, the AI Dialogue has the potential to become a space where global governance is not only discussed, but actively co-produced in a more inclusive and balanced way.

The AI Dialogue should build upon and connect with existing global and regional initiatives to avoid duplication and strengthen coherence. A key reference point is the work emerging from the UN Forum 2025 on Digital Justice, which emphasizes equity, human rights, and the inclusion of marginalized communities in digital governance. It should also engage with frameworks developed by UNESCO, particularly on the ethics of AI, as well as ongoing discussions within United Nations processes on digital cooperation. Regional efforts such as those led by the African Union or the European Union also provide important regulatory and policy models that reflect different governance approaches. In addition, multi-stakeholder initiatives involving civil society, academia, and the private sector should be integrated, as they often bring practical insights and grounded perspectives that are missing from purely intergovernmental processes. The added value of the AI Dialogue lies in its ability to connect these fragmented efforts into a more coherent ecosystem. It can serve as a coordination hub that amplifies underrepresented voices, particularly from the Global South, and ensures that their contributions shape global standards. Moreover, the Dialogue can introduce a stronger focus on implementation and accountability, translating existing principles into actionable commitments. By linking normative frameworks with practical cooperation—such as funding, capacity-building, and knowledge exchange, it can help close the gap between ambition and impact.

Different stakeholders should be engaged as co-creators, not just participants. Governments can provide regulatory perspectives and commit to implementation pathways; civil society organizations can bring grounded insights on social impacts and rights; academia can contribute research and evidence; and the private sector can share technical expertise while being held accountable for real-world impacts. To enable this, the AI Dialogue should adopt a multi-layered structure. This could include: * Plenary sessions to define shared priorities and political commitments * Thematic working groups where stakeholders collaborate on specific issues (e.g., human rights, data governance, cultural inclusion) * Regional consultations to ensure context-specific inputs, especially from the Global South * Outcome-oriented tracks focused on producing recommendations, toolkits, or policy guidelines Importantly, participation must be supported by access mechanisms such as funding for underrepresented groups, language accessibility, and hybrid (online/offline) formats. This would ensure that contributions are not limited to well-resourced actors

Global AI governance discussions continue to underrepresent communities from the Global South, particularly Afro-descendant populations, Indigenous groups, and marginalized linguistic communities. Grassroots organizations, informal sector workers, and cultural actors are also often excluded, despite being directly affected by AI systems. Their exclusion is not only geographic but also structural linked to barriers such as limited access to funding, technical expertise, and global policy spaces. As highlighted in discussions at the UN Forum 2025 on Digital Justice, this leads to governance models that do not fully reflect diverse realities. To address this, inclusion must go beyond symbolic representation. It requires dedicated funding mechanisms, capacity-building initiatives, and partnerships that enable these actors to participate meaningfully. Language accessibility is also critical—AI governance spaces should not be limited to dominant global languages. In addition, mechanisms for continuous engagement, not just one-off consultations, should be established, allowing these communities to shape agendas, not only respond to them.

To foster dynamic and meaningful engagement, the AI Dialogue should move beyond traditional conference formats and adopt more interactive and participatory models. One approach is the use of co-creation labs, where diverse stakeholders work together in small groups to design concrete policy solutions or frameworks. These labs can be problem-driven and produce tangible outputs. Another format is scenario-based dialogues, where participants engage with real or hypothetical cases (e.g., AI in public services, surveillance, or cultural production) to explore trade-offs and governance responses. This helps bridge abstract principles with practical realities. Storytelling and lived-experience sessions can also be powerful. By centering voices directly affected by AI—such as workers, artists, or community leaders—these sessions bring human impact into policy discussions and align with justice-centered approaches. Additionally, digital participation platforms can enable broader, asynchronous engagement, allowing stakeholders from different regions and time zones to contribute. Finally, multi-stakeholder roundtables with decision-making mandates can ensure that discussions lead to actionable outcomes, rather than remaining purely deliberative. Together, these formats can make the AI Dialogue more inclusive, grounded, and results-oriented.

Several existing policies and approaches offer valuable lessons for more effective and inclusive AI governance. At the global level, the Recommendation on the Ethics of AI developed by UNESCO provides a comprehensive normative framework grounded in human rights, inclusion, and sustainability. Its emphasis on impact assessments and accountability mechanisms is particularly relevant for operationalizing ethical principles. In Europe, the AI Act adopted by European Union introduces a risk-based regulatory model, classifying AI systems according to their potential harm and imposing corresponding obligations. This approach offers a concrete method for translating abstract risks into enforceable rules, especially in areas such as biometric surveillance and high-risk applications. From a data governance perspective, emerging practices around data stewardship and data trusts provide innovative ways to manage data collectively, ensuring that communities retain a degree of control and can benefit from how their data is used. These models are particularly relevant for addressing concerns around data extraction and inequitable value distribution. In addition, algorithmic impact assessments (AIAs), increasingly used in public sector contexts, offer practical tools to evaluate potential harms before deploying AI systems. When combined with transparency requirements and public oversight, they can strengthen accountability. Finally, multi-stakeholder initiatives highlighted in forums such as the UN Forum 2025 on Digital Justice demonstrate the importance of inclusive governance processes. They show that effective AI governance is not only about regulation, but also about participation, capacity-building, and aligning technological development with social justice goals. Together, these examples illustrate that effective AI governance requires a combination of normative frameworks, regulatory tools, and inclusive practices.