University of São Paulo

A successful outcome for the first Global Dialogue on AI Governance would be the alignment around a set of shared baseline principles to guide the responsible development and use of AI across different national contexts. These should include commitments to safety, transparency, accountability, and the protection of human rights. Ensuring meaningful participation from developing countries and addressing technological asymmetries is essential for achieving inclusive and equitable global governance. Additionally, success would involve defining actionable priorities with clear short- and medium-term goals, particularly in areas such as risk assessment, auditing of AI systems, and regulatory interoperability. Ultimately, the dialogue must move beyond discussion toward implementation, with measurable commitments and follow-up mechanisms.

• Safe, secure and trustworthy AI
• AI capacity-building
• Protection and promotion of human rights
• Transparency, accountability, and human oversight

The selected priorities reflect the need to balance innovation with responsibility. Ensuring safe, secure, and trustworthy AI is fundamental to mitigating systemic risks and building public trust. AI capacity-building is critical to reducing global inequalities and enabling broader participation in the digital economy, preventing the concentration of technological power in a few regions. The protection and promotion of human rights must remain a central guiding principle, ensuring that AI systems do not reinforce discrimination, violate privacy, or exacerbate social exclusion. Finally, transparency, accountability, and human oversight are essential to ensure that AI systems are understandable, auditable, and aligned with societal values.

An important cross-cutting issue is the governance of foundation models and general-purpose AI systems, which pose unique challenges in terms of risk assessment, control, and accountability across the value chain. Another critical concern is the concentration of power among a small number of companies and countries, raising questions about competition, digital sovereignty, and equitable access to infrastructure and data. The environmental impact of AI, including energy consumption and resource use, is also an emerging issue that deserves greater attention in governance discussions.

Governance gaps in AI are creating both significant challenges and emerging opportunities in my country/region/sector. One of the main challenges is the lack of clear, harmonized regulatory frameworks, which leads to uncertainty for organizations deploying AI systems. This is particularly evident in areas such as risk classification, accountability, and auditability, where guidance remains fragmented or underdeveloped. Another critical gap is limited institutional and technical capacity to effectively oversee AI systems. Many public institutions and smaller organizations lack the expertise and resources needed to assess risks, ensure compliance, and implement robust safeguards. At the same time, the rapid pace of technological development is outstripping existing governance mechanisms, especially in relation to general-purpose and foundation models.. However, these challenges also present important opportunities. Increased awareness of ethical and human rights implications is driving demand for more responsible AI governance. Additionally, there is an opportunity to foster international cooperation and regulatory interoperability, which can reduce fragmentation and support innovation. By addressing governance gaps proactively, the region/sector can build a more inclusive, trustworthy, and resilient AI ecosystem.

One of its key contributions would be facilitating alignment around shared principles and promoting interoperability between different regulatory approaches. Rather than imposing uniform rules, the Dialogue can support mutual understanding and compatibility across frameworks, reducing fragmentation and enabling cross-border collaboration.

Regional and national AI strategies, as well as initiatives led by the private sector and civil society, are also important foundations. Connecting these efforts can help avoid duplication, promote coherence, and leverage existing expertise and resources. The added value of the AI Dialogue lies in its ability to act as a convening and coordinating platform at the global level, particularly within the UN context. It can help bridge gaps between policy and implementation, as well as between different regions and levels of development.

Different stakeholders can contribute by bringing complementary expertise, perspectives, and practical experience to the AI Dialogue. Governments can provide policy direction and regulatory insights, while the private sector can share technical knowledge, implementation challenges, and innovation pathways. Academia can contribute independent research and evidence-based analysis, and civil society can ensure that societal impacts, ethical concerns, and human rights considerations remain central. To support meaningful contributions, the Dialogue should adopt a structured, multistakeholder format. This could include thematic working groups, regional consultations, and sector-specific roundtables that allow for more focused and in-depth discussions.

Several voices remain underrepresented in global AI governance discussions, particularly stakeholders from developing countries, small and medium-sized enterprises, grassroots organizations, and marginalized communities. Indigenous groups, local communities, and non-English-speaking populations are often excluded, despite being directly affected by AI systems.

Innovative engagement formats can significantly enhance participation and outcomes in the AI Dialogue. Interactive, problem-solving formats—such as policy labs, scenario-based simulations, and case study workshops—can help stakeholders collaboratively address real-world challenges and explore trade-offs in AI governance. Multistakeholder "co-creation sessions" can be used to jointly develop guidelines, frameworks, or recommendations, fostering ownership and alignment across different groups.

Several existing policies and approaches offer valuable examples for effective AI governance. The OECD AI Principles and UNESCO's Recommendation on the Ethics of AI provide widely endorsed, human rights-based frameworks that guide national strategies and regulatory efforts. The EU AI Act represents a concrete risk-based regulatory approach, offering a structured model for classifying and governing AI systems according to their potential impact. In practice, algorithmic impact assessments (AIAs) have emerged as an important tool to evaluate risks prior to deployment, particularly in the public sector. Similarly, independent auditing and certification mechanisms are gaining traction as ways to ensure accountability and build trust in AI systems. Technical standards and frameworks, such as NIST's AI Risk Management Framework, contribute practical guidance for organizations to identify, measure, and mitigate risks.