United Way Worldwide

A successful first Global Dialogue on AI Governance would be defined by its ability to move the global conversation from principles to practical implementation across diverse institutional contexts. As discussions across the United Nations system continue to advance responsible AI governance, an important outcome would be the clear identification of implementation gaps, particularly for institutions that adopt and deploy AI systems rather than develop them. Civil society and mission-driven organizations represent one of the largest global ecosystems delivering essential public services, yet governance guidance for responsible AI adoption within this sector remains significantly underdeveloped. A successful Dialogue would therefore: Recognize and elevate civil society as a core stakeholder in AI governance discussions Identify priority governance challenges in real-world service environments, including data stewardship, transparency, algorithmic accountability, and institutional trust Advance practical guidance that translates responsible AI principles into operational governance structures suitable for mission-driven organizations. In addition, the Dialogue should produce actionable outputs, such as: a set of priority governance recommendations for AI adoption contexts a roadmap for developing implementation-focused governance guidance and mechanisms for ongoing multi-stakeholder collaboration. Finally, success would be reflected in the establishment of a continuing platform or working structure that ensures insights from the Dialogue inform future policy development, including contributions from practitioners operating in real-world environments. Strengthening governance frameworks for institutions that implement AI in public service contexts will be essential to ensuring that AI innovation advances social progress while reinforcing public trust, accountability, and equitable outcomes globally.

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

The selected priorities reflect the need to ensure that artificial intelligence is implemented responsibly in real-world service environments, particularly within civil society and mission-driven organizations. Safe, secure and trustworthy AI is foundational, as these institutions operate in high-trust contexts and manage sensitive data related to vulnerable populations. Ensuring reliability and security is essential to maintaining public confidence. AI capacity-building is critical because many nonprofit and humanitarian organizations are adopters-not developers-of AI systems. They require practical skills, governance structures, and operational guidance to implement AI responsibly. Transparency, accountability, and human oversight directly address the governance challenges these organizations face when deploying third-party AI systems. Clear oversight mechanisms are necessary to ensure ethical use, mitigate bias, and support informed decision-making. Protection and promotion of human rights is central to mission-driven work. AI systems used in public service delivery must align with human rights principles to ensure equitable outcomes and prevent harm, particularly for marginalized communities. Together, these priorities emphasize the importance of moving beyond high-level principles toward practical governance approaches that support responsible AI adoption. Strengthening these areas will help ensure that AI enhances service delivery while reinforcing public trust, accountability, and social impact.

Yes. While the listed themes are comprehensive, an important cross-cutting issue that warrants greater attention is the governance of AI adoption in real-world service environments, particularly within civil society and mission-driven institutions. Much of the current AI governance landscape focuses on AI development, model safety, and regulatory oversight, yet a growing number of organizations are adopting and deploying AI systems created by third-party providers. These institutions-including nonprofits, humanitarian organizations, and public service entities-operate as stewards of sensitive data and trusted intermediaries between governments and communities. As a result, there is an emerging need for implementation-focused governance, including: Procurement and due diligence standards for third-party AI tools Operational oversight mechanisms for AI-enabled decision-making Lifecycle governance practices, including monitoring, evaluation, and risk management after deployment Context-specific safeguards to protect vulnerable populations. Additionally, there is a need to address institutional trust as a governance objective, particularly in environments where AI systems influence access to essential services. Another cross-cutting issue is the translation of high-level AI principles into practical, actionable guidance that organizations of varying sizes and capacities can implement. Without this, there is a risk of widening the gap between global governance frameworks and real-world adoption. Addressing these emerging issues would strengthen the effectiveness of global AI governance efforts by ensuring that responsible AI is not only defined at the policy level but also operationalized in the environments where it directly impacts people and communities

In the nonprofit and mission-driven sector, governance gaps in trustworthy AI, capacity-building, transparency and accountability, and human rights protections are already shaping both risks and opportunities. A primary challenge is that many organizations are adopters of AI systems developed by third parties, yet lack practical governance guidance to evaluate, implement, and oversee these tools. This creates exposure in areas such as data stewardship, where sensitive beneficiary information must be protected, and algorithmic accountability, where decision-making processes may not be fully transparent or explainable. Capacity constraints further compound these challenges. Many organizations do not have dedicated resources or expertise to operationalize responsible AI principles into policies, oversight structures, and risk management practices, increasing the likelihood of inconsistent or ungoverned AI use. There is also a growing concern around institutional trust. Mission-driven organizations operate in high-trust environments, and any misuse or unintended consequences of AI could undermine confidence among the communities they serve. At the same time, there are significant opportunities. AI has the potential to enhance service delivery, improve resource allocation, and expand access to critical services, particularly in underserved communities. Emerging developments in responsible AI frameworks and governance models provide a foundation for more structured adoption. However, to fully realize these benefits, there is a need for practical, implementation-focused governance guidance tailored to organizations that deploy AI in real-world service environments. Strengthening governance in these areas will be essential to ensuring that AI adoption advances equitable outcomes, protects human rights, and reinforces public trust across the sector.

The Global Dialogue on AI Governance can play a critical role in advancing international cooperation by serving as a bridging platform between policy, practice, and diverse stakeholder perspectives. First, the Dialogue can help align global efforts by identifying shared governance priorities and gaps, particularly in areas where current frameworks remain fragmented. By convening stakeholders across regions and sectors, it can support greater coherence in approaches to trustworthy AI, accountability, and human rights protections. Second, the Dialogue can advance cooperation by emphasizing implementation-focused governance. While many international efforts have established high-level principles, there remains a need for practical guidance that supports how AI is adopted and governed in real-world environments. This is especially relevant for civil society and mission-driven organizations, which often operate across borders and require governance models that are adaptable, scalable, and context-sensitive. Third, the Dialogue can foster collaboration through the exchange of case studies, best practices, and lessons learned, enabling countries and institutions to build on existing work rather than duplicating efforts. Highlighting practical examples of responsible AI adoption can accelerate collective learning and inform more effective policy development. Finally, the Dialogue can support the development of ongoing coordination mechanisms, such as working groups or knowledge-sharing platforms, that sustain engagement beyond the initial convening. This would help ensure that insights generated through the Dialogue contribute to continuous progress in global AI governance. By strengthening alignment, promoting practical implementation, and enabling sustained collaboration, the AI Dialogue can help ensure that international cooperation supports responsible, inclusive, and trustworthy AI adoption worldwide.

The Global Dialogue on AI Governance can build upon several existing international initiatives while adding value by connecting principles to implementation across sectors. Key initiatives include efforts within the United Nations system, particularly the Global Digital Compact, which emphasizes inclusive and human-centered digital governance, and the AI governance work coordinated through the International Telecommunication Union, which brings together governments, industry, and technical experts. In addition, the OECD AI Principles and the UNESCO Recommendation on the Ethics of Artificial Intelligence provide widely recognized frameworks for responsible AI. While these initiatives establish strong normative foundations, a key gap remains in operationalizing these principles in real-world environments, particularly for organizations that adopt AI systems rather than develop them. The Global Dialogue can add value by: Bridging global principles with practical implementation, translating existing frameworks into actionable governance approaches Connecting stakeholders across sectors, including civil society and mission-driven organizations that are underrepresented in current governance discussions Highlighting real-world use cases and lessons learned, particularly in public service delivery contexts Supporting the development of implementation-focused guidance that complements existing frameworks and enables consistent, responsible AI adoption. By building on existing global initiatives while addressing the gap between policy and practice, the Dialogue can strengthen international cooperation and ensure that AI governance frameworks are not only aligned at a high level, but also effective, inclusive, and applicable across diverse institutional contexts.

Different stakeholders can contribute most effectively to the AI Dialogue when their roles are clearly linked to both policy development and practical implementation. Member States can provide policy direction, regulatory insight, and national priorities, while the private sector can contribute technical expertise, responsible innovation practices, and transparency mechanisms. Civil society and mission-driven organizations play a critical role in bringing real-world perspectives on how AI affects communities, particularly in public service delivery contexts. Academia and the technical community can support research, evaluation, and evidence-based approaches, while international organizations can help coordinate global alignment and knowledge-sharing. To maximize impact, the Dialogue should be structured to move beyond high-level discussion and enable practical exchange and collaboration. Recommended format and structure include: Thematic working sessions organized around key governance areas such as trustworthy AI, data stewardship, and accountability Case-based discussions where stakeholders share real-world implementation experiences, challenges, and lessons learned Multi-stakeholder roundtables that bring diverse perspectives into focused dialogue on specific governance questions Interactive formats, such as breakout groups or scenario-based exercises, to encourage deeper engagement Integration of the Scientific Panel's findings into facilitated discussions that translate research into actionable insights In addition, the Dialogue could benefit from establishing ongoing mechanisms, such as working groups or communities of practice, to continue collaboration beyond the initial convening. Structuring the Dialogue in this way would help ensure that it not only fosters inclusive discussion, but also produces practical, implementation-focused outcomes that support responsible and effective AI governance across sector

From a mission-driven and implementation perspective, one of the most underrepresented voices in global AI governance discussions is civil society and nonprofit organizations that adopt and deploy AI systems in real-world service environments. Much of the current dialogue is shaped by governments, large technology companies, and academic institutions. While these perspectives are essential, they do not fully capture the experiences of organizations that are on the front lines of service delivery, including nonprofits, humanitarian agencies, and community-based organizations. These institutions often operate as stewards of sensitive data and as trusted intermediaries between systems and the populations they serve, particularly vulnerable or marginalized communities. Additionally, there is limited representation from practitioners responsible for implementing AI in operational settings, including technology leaders within mission-driven organizations. Their insights into challenges such as third-party AI adoption, data governance, transparency, and institutional trust are critical to ensuring that governance frameworks are practical and effective. To better include these perspectives, the AI Dialogue could: Intentionally incorporate civil society and nonprofit technology leaders into speaking roles, working groups, and advisory structures Create dedicated sessions focused on AI adoption in public service and community-based contexts Elevate case studies and lived experiences from organizations deploying AI in real-world environments Support capacity-building and participation mechanisms that enable smaller or under-resourced organizations to engage meaningfully. Expanding participation in this way will help ensure that global AI governance frameworks are not only technically sound, but also inclusive, implementable, and grounded in the realities of those delivering services and supporting communities worldwide

To foster meaningful and dynamic engagement, the AI Dialogue should move beyond traditional panel formats and incorporate interactive, implementation-focused engagement models that reflect the complexity of real-world AI governance. One effective format is case-based working sessions, where organizations present real examples of AI adoption, including governance challenges related to data stewardship, transparency, and accountability. These sessions can ground discussions in practical experience and enable participants to collaboratively explore solutions. Scenario-based simulations are another valuable approach. Participants can work through realistic situations—such as deploying AI in public service delivery or managing third-party AI risks—to examine governance trade-offs and decision-making processes in a structured way. The Dialogue could also include multi-stakeholder roundtables focused on specific governance questions, allowing representatives from different sectors to engage in smaller, facilitated discussions that encourage deeper exchange and shared understanding. In addition, interactive breakout groups organized around thematic areas—such as trustworthy AI, data governance, and human rights—can help participants contribute more actively and bring forward diverse perspectives. Finally, incorporating implementation labs or governance design sprints would provide a space for participants to collaboratively develop practical guidance, frameworks, or policy recommendations in real time. This work can also build on emerging governance approaches designed to support responsible AI adoption in mission-driven and civil society contexts. These formats would enable the Dialogue to move beyond discussion and generate practical insights, shared learning, and actionable outcomes that advance responsible and effective AI governance across sectors.

Effective AI governance is increasingly being shaped by a combination of global policy frameworks and emerging implementation practices. At the international level, initiatives such as the OECD AI Principles and the UNESCO Recommendation on the Ethics of Artificial Intelligence provide foundational guidance on transparency, accountability, and human-centered AI. Within the United Nations system, ongoing efforts such as the Global Digital Compact further emphasize inclusive and responsible digital governance. At an operational level, leading practices are emerging around AI lifecycle governance, including risk assessments prior to deployment, continuous monitoring of AI systems, and the establishment of internal governance structures such as ethics review boards or AI oversight committees. In addition, organizations are increasingly adopting third-party AI risk management practices, including vendor due diligence, data protection safeguards, and requirements for transparency and explainability. Platforms that support governance are also evolving, including tools for algorithmic auditing, model monitoring, and data governance, which help organizations operationalize responsible AI principles in practice. However, a key gap remains in translating these global principles into practical, implementation-focused governance approaches, particularly for organizations that adopt AI systems rather than develop them. In my own work examining responsible AI adoption in mission-driven institutions, this has informed the development of governance approaches designed to support responsible AI implementation in civil society contexts. These approaches emphasize operational governance, including data stewardship, oversight of third-party AI systems, and alignment with organizational mission and public trust. Together, these policies, practices, and emerging approaches demonstrate that effective AI governance requires both strong global principles and practical frameworks that enable responsible adoption across diverse institutional environments. Feel free to reach out for detailed explanation of the framework for mission driven organizations and civil society as a whole.