Pathfinders for Peaceful, Just, and Inclusive Societies, International Center for Transitional Justice (NYU-CIC), Hague Institute for Innovation in Law (HiiL), American Arbitration Association (AAA)

1. Meaningful, diverse participation. The Dialogue should enable substantive engagement across local, regional, and global perspectives, including those who can voice lived experiences of effective, or failed, AI governance. Participation must also span disciplines. AI governance cannot be designed by technologists alone. It requires collaboration with experts in the sociopolitical dimensions of AI adoption, including justice and dispute resolution actors who play a central role in enforcement, accountability, and access to remedy. 2. Clear, forward-looking priorities. Beyond thematic clusters, the Dialogue should define a focused set of priority areas for sustained international cooperation. These priorities should guide multistakeholder efforts beyond 2026 and anchor future Dialogues. A human rights approach should underpin all substantive areas. These priorities should be designed to evolve over time with built-in mechanisms for periodic review and refinement as AI capabilities and risks continue to develop. 3. Sustained engagement architecture. The Dialogue should establish mechanisms to maintain inclusive, continuous engagement between civil society, academia, private sector, and government agencies. This could include ongoing regional consultations and thematic consultations that feed into future convenings and enable iterative updates to governance approaches as AI evolves. 4. Public, trackable commitments. Governments, international organizations, civil society partners and private sector should be encouraged to make measurable commitments tied to positive impact. These commitments should be transparent, trackable, and adaptable over time, creating accountability beyond the Dialogue itself and allowing for recalibration as technologies and their impacts evolve. 5. Action-oriented learning and exchange. The Dialogue should elevate best practices and successful policies in different contexts, with an emphasis on experimentation, iteration, and evidence. As AI evolves rapidly, governance approaches must be continuously tested, refined, and updated to remain effective. Maintaining a shared repository of proven and emerging practices can support continuous learning, broader engagement and use outside of the Dialogue.

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

Safe, secure, and trustworthy AI directly aligns with advancing fair, transparent, and accessible dispute resolution systems. As AI becomes increasingly embedded in legal processes, the need for clear governance, ethical guardrails, and accountability is urgent. Innovation must be guided by principles of trust, equity, and due process to build scalable, human-centered systems that expand access while maintaining the integrity of outcomes. AI capacity building is an institutional challenge requiring stronger political, economic, and social systems to safeguard human rights and protect livelihoods. Regardless of where a country sits along the AI adoption curve, there is an immediate opportunity to build today against the disruptions of tomorrow. The justice sector is a priority area for capacity building. As enforcers of legislated regulations, justice actors will be among the first to encounter gaps in AI governance design, functioning as an early-warning system for rights violations and unintended consequences from real-world deployment. Building coordination across justice, technology, and AI governance sectors ensures policies are practical, enforceable, and responsive to evolving risks. AI adoption across legal systems is accelerating, but unevenly, presenting risks for rights, accountability, and public trust - compounded by AI's role in shaping information ecosystems, where disinformation, exclusionary narratives, and epistemic inequities can distort access to truth, entrench distrust, and create new barriers to justice, particularly in fragile or conflict-affected contexts. Building capacity among justice actors to make informed, principled decisions about AI is essential to transparency, accountability, and meaningful human oversight. In a world where over 5 billion people lack meaningful access to justice, safeguarding human rights in AI governance must be grounded in how people experience justice problems and seek resolution. As domestic legal systems implement AI governance frameworks, human rights protections should be anchored in a people-centered approach that advances transparency, strengthens accountability, and embeds human oversight across all pathways to justice.

A critical cross-cutting issue is the gap between AI governance in principle and its implementation in practice. AI presents a powerful opportunity to advance the common good, but only if governance frameworks are translated into systems that meaningfully to safeguard rights for everyone. This requires focused capacity-building in the points of friction between AI adoption and existing social, economic, and political systems. In practice, domestic legal systems and dispute resolution mechanisms will be responsible for operationalizing AI governance and enforcing rights. Their ability to identify emerging risks, respond to rights violations, and shape responsible AI use will determine whether governance frameworks function in people's lives or remain aspirational. These gaps are intensified in information ecosystems shaped by AI, where misinformation and epistemic harms can undermine accountability, obscure rights violations, and limit individuals' ability to seek and obtain effective remedies. Closing this gap demands deeper collaboration between AI governance experts, mission-driven technologists, and justice actors. It also requires embedding a people-centered justice perspective into AI governance discussions, grounded in how individuals actually experience harm, seek redress, and navigate legal systems. Justice is not a downstream consideration; it is the mechanism through which rights are realized. As such, it should be treated as a cross-cutting priority across all thematic areas of the Dialogue.

AI is rapidly reshaping legal systems, creating significant opportunities alongside real governance challenges. OECD analysis of 200 use cases shows that justice administration and access to justice are among the leading domains for public sector AI deployment. Governments are using AI to streamline case management, improve service delivery, and generate insights to inform system performance and reform. The World Bank similarly identifies key applications across civil justice systems, including administrative efficiency, judicial decision support, expanded access through guided tools, data privacy protections, enhanced alternative and online dispute resolution, and system-level analytics. At the same time, governance frameworks are lagging behind adoption. A 2023 UNESCO survey found that while 44% of judicial actors report using AI in their work, only 9 percent have received guidance on responsible use. This gap is compounded by uneven infrastructure: many jurisdictions lack the capacity to develop or adapt AI systems such as locally relevant large language models (LLMs) that reflect applicable laws, languages, and social contexts. These gaps directly impact progress toward legal inclusion -ensuring individuals can access affordable, effective tools to resolve their legal problems in a responsible and sustainable way. Without intentional design, AI risks reinforcing, rather than reducing, existing inequities in access to justice. Without clear standards, oversight, and technical capacity, these gaps threaten to undermine fairness, amplifying bias, and eroding trust in legal systems. These risks are amplified in digital information environments where AI can propagate misleading or divisive content, weakening confidence in legal institutions and complicating efforts to establish truth and accountability. Strengthening governance in practice, alongside adoption, is essential to ensuring that AI advances legal inclusion and meaningfully expands access to justice.

The AI Dialogue can play a critical role in advancing international cooperation by expanding inclusivity, shaping shared norms, and grounding AI governance in human rights. It has the opportunity to move beyond high-level principles and drive alignment in how governance is implemented in practice. Drawing on the principles of people-centered justice, as outlined in the Hague Declaration for Equal Access to Justice for All by 2030, the Dialogue can encourage an approach that begins with evidence to understand how people experience AI in their daily lives, and uses those insights to inform policy design. This includes investing in robust, people-centered data infrastructure to understand how AI is affecting individuals' rights and how they seek remedies through legal systems, drawing on tools such as legal needs surveys and other empirical methods to inform national AI strategies. Equally important is ensuring meaningful participation of those most affected by AI, including individuals experiencing harm or exclusion, so they are not only represented but engaged as contributors to policy development and implementation. In addition, the Dialogue can help elevate the intersection of AI, information integrity, and justice by promoting approaches that mitigate misinformation risks and support the development of AI systems that strengthen transparency, truth-seeking, and access to remedy. Finally, the Dialogue can strengthen transparency, accountability, and human oversight by fostering continuous feedback loops between justice actors and AI governance policymakers, and by encouraging the development of clear, accessible mechanisms for redress when AI-related harms occur.

The AI Dialogue can build on and connect with existing multi-stakeholder initiatives that are already advancing people-centered justice and access to justice globally. Greater coordination with the Justice Action Coalition (JAC) would be particularly valuable in reinforcing the AI Dialogue's human rights priorities. The JAC brings together a broad alliance of countries and organizations committed to achieving measurable progress in justice outcomes. Engaging with this network would provide the AI Dialogue with a strong foundation of expertise at the intersection of AI, justice systems, and societal impact. Through this collaboration, the Dialogue can better integrate practical insights on both the opportunities and risks of AI deployment in legal systems. This includes strengthening domestic capacity to implement AI governance frameworks, ensure accountability, and provide access to remedy for AI related harms – helping translate global principles into operational reality. At the regional level, the African Alliance for People-Centered Justice and the Ibero-American Alliance for Access to Justice offer established platforms for inclusive, context-specific engagement. These networks can support the Dialogue in surfacing diverse perspectives, sharing data and best practices, and ensuring that governance approaches reflect local legal, cultural, and institutional realities. By connecting these efforts, the AI Dialogue can add value as a global convening and coordination mechanism, linking policy, practice, and lived experience to advance more coherent, inclusive, and actionable AI governance.

As the format of the AI Dialogue is already established, our recommendation would be to not just open the door to participation, but actually guide and support its realization. Without financial sponsorship and visa support, many underrepresented stakeholders may be unable to join the AI Dialogue in practice. Additional innovative engagement formats have been shared in question 17.

AI is reshaping how people experience rights, accountability, and access to justice, yet justice actors and institutions remain underrepresented in global AI governance conversations. This gap has real consequences. Effective and inclusive AI governance depends on the ability to enforce rights in practice, but billions of people worldwide still lack meaningful access to justice. Without deliberate integration of justice systems, AI governance frameworks risk remaining aspirational: lacking enforceability, weakening accountability, and failing to provide accessible pathways to remedy when AI-related harms occur. In this context, governance gaps do not just limit effectiveness: they risk deepening existing inequalities, eroding public trust, and allowing new forms of harm to persist unaddressed. Addressing this requires more intentional inclusion of justice-sector voices as well as the communities most affected by AI-related harms. Existing platforms like the Justice Action Coalition and the Digital Justice Working Group of the World Bank would offer immediate entry points for engagement. The AI Dialogue could formalize these connections by inviting structured input, commissioning joint outputs, or integrating their expertise into its working processes. By elevating justice as a core pillar of AI governance, and embedding it into the Dialogue's structure, the process can better ensure that global frameworks translate into real-world accountability and access to remedy.

The AI dialogue should adopt a mix of structured and participatory formats that enable continuous, multi-stakeholder input. This could include targeted surveys across stakeholder groups aligned with the Dialogue's themes, as well as a centralized repository to capture and share emerging best practices in AI governance. Equally important is creating space for lived experience. Dedicated speaking opportunities for individuals and communities affected by AI-related harms would ensure that policy discussions remain grounded in real-world impact. Cross-cutting thematic sessions such as a dedicated track on the sociopolitical impacts of AI can help bridge silos and connect insights across domains. To broaden participation, the Dialogue could also include regional and satellite events enabling more inclusive engagement across geographies and ensuring that diverse legal, cultural, and institutional perspectives are reflected. Taken together, these engagement formats support an iterative and adaptive approach to governance that evolves alongside AI capabilities, rather than producing static recommendations that can become obsolete with the rapid pace of AI change, the Dialogue can function as a living platform for learning, feedback, and continuous improvement, strengthening both the relevance and durability of its outcomes.

• Here is a list of relevant resources on interventions and implications at the intersection of AI, AI governance, and justice: A/77/219: Promotion and protection of human rights: human rights questions, including alternative approaches for improving the effective enjoyment of human rights and fundamental freedoms
• A/79/296: Human rights in the administration of justice: Report of the Secretary-General
• A/80/169: AI in judicial systems: promises and pitfalls - Report of the Special Rapporteur on Independence of Judges and Lawyers
• A People-Centered Justice Approach to Implementing AI Governance: How the Justice Sector Can Support Inclusive, Equitable, and Enforceable AI Governance, NYU Center on International Cooperation
• AI for Justice and Justice for AI: Why Access to Justice Enables Better AI Governance, NYU Center on International Cooperation
• AI Justice Atlas, Oxford Institute of Technology and Justice
• Building Trust in AI Through Justice, NYU Center on International Cooperation
• Governing with Artificial Intelligence: The State of Play and Way Forward in Core Government Functions, Chapter 5: Deep dive: The current status and future potential of AI in government [Section: AI in justice administration and access to justice, AI in justice administration and access to justice], OECD
• AI and Access to Justice: A Snapshot of Current Trends and Future Potential, Hague Institute of Innovation in Law (HiiL)
• Regulating AI in the Delivery of Consumer-Facing Legal Services: Unlocking Legal Regulation, Institute for the Advancement of the American Legal System (IAALS)
• The Digital Divide in Justice: Consistent Progress and Persistent Challenges, Justice Technology Association
• The Pact for the Future (II.39.g)
• Our Common Agenda (p. 27)
• The New Vision for the Rule of Law (p. 2, 7)
• A New Agenda for Peace (p. 6)