AIgnite Women

Three outcomes would signal meaningful progress for the first Global Dialogue. First, the Dialogue should establish a shared vocabulary and common understandings on priority governance gaps. The current AI governance landscape is fragmented across regional and sectoral instruments. A successful Dialogue would map where frameworks converge and where critical gaps remain, particularly on issues that existing mechanisms have not adequately addressed, such as the gendered impacts of AI systems, algorithmic accountability in cross-border deployment, and governance of increasingly autonomous AI agents. Second, the Dialogue should generate actionable commitments to inclusive governance design. Resolution A/RES/79/325 emphasizes closing digital divides between and within countries. The Dialogue should translate this mandate into specific mechanisms for ensuring that governance frameworks are not designed exclusively by the most technologically advanced nations but reflect the needs and capacities of developing countries. This includes meaningfully integrating the perspectives of women and girls, who remain significantly underrepresented in AI governance and development. Third, the Dialogue should lay groundwork for the intergovernmental consultations that the Co-Chairs of the second Dialogue are mandated to conduct under paragraph 8 of the Resolution. This means identifying concrete priority areas where international cooperation can add value beyond what existing bilateral or regional processes already provide, and establishing a clear process for tracking progress between sessions. A successful Dialogue would not attempt to centralize global AI governance. Rather, it would function as the Co-Chairs have described it: a space where existing initiatives converge, gaps are identified, and all stakeholders, especially those historically excluded from technology governance, can meaningfully shape the rules that will affect their communities.

• Safe, secure and trustworthy AI
• Transparency, accountability, and human oversight
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches

AIgnite Women works to advance women's participation, leadership, and influence in the global AI ecosystem. Our selection reflects the thematic areas where governance gaps most directly affect women and girls. Safe, secure and trustworthy AI is foundational. AI safety cannot be achieved without addressing the specific risks AI systems pose to women and girls, including technology-facilitated gender-based violence, non-consensual deepfakes, and algorithmic profiling. Research has documented gender bias across AI systems, from large language models reinforcing regressive stereotypes (UNESCO, 2024) to AI-generated imagery biased against older working women (Stanford/Nature, 2025). Safety frameworks that ignore gendered harms are incomplete. The social, economic, ethical, cultural, linguistic and technical implications of AI are inseparable from gender equality. AI-driven automation may reshape jobs in sectors where women constitute a significant share of the workforce, while women remain significantly underrepresented where AI systems are designed and governed. Interoperability of governance approaches matters because AI systems operate across jurisdictions. Incompatible regulatory standards leave affected communities without recourse. The Dialogue is uniquely positioned to map where regional frameworks, from the EU AI Act and Council of Europe Convention, to the African Union Continental AI Strategy, ASEAN's Guide on AI Governance and Ethics, and the Santiago and Montevideo Declarations in Latin America, can be made more compatible without imposing uniformity. Transparency, accountability, and human oversight are prerequisites for gender-responsive AI governance. Communities facing algorithmic discrimination in credit, employment, healthcare, and justice cannot challenge decisions they cannot understand. Mandatory transparency obligations and accessible redress mechanisms are essential.

Gender-responsive AI governance should be treated as a cross-cutting lens, not siloed into a single thematic cluster. The Resolution references social, economic, and ethical implications, but gender is not named as a standalone consideration. Yet AI systems produce differentiated impacts across hiring, credit, healthcare, facial recognition, and content generation, compounded by technology-facilitated gender-based violence. The UN Working Group on Discrimination Against Women and Girls' forthcoming 2026 thematic report on AI and substantive gender equality should inform the Dialogue, with gender analysis mainstreamed across all thematic clusters. Second, the governance of agentic AI systems, meaning AI that executes multi-step tasks autonomously with limited human oversight, cuts across safety, accountability, and human rights. As AI agents are deployed for hiring, healthcare triage, credit assessments, and legal analysis, responsibility for harm becomes urgent. Current governance frameworks were largely designed for tool-like AI with clearer human control. Agentic AI disrupts that assumption. Third, AI governance literacy is a prerequisite for meaningful participation. AI literacy is often framed as learning to use AI tools, but governance literacy is more urgent: understanding how AI is regulated, where accountability gaps exist, what rights affected individuals hold, and how to engage in governance processes. This is especially critical for women and communities in developing countries affected by AI-driven decisions in employment, credit, and services. Addressing these issues requires moving beyond voluntary principles toward enforceable obligations, including gender impact assessments for high-risk AI systems, clear liability frameworks for autonomous or agentic AI, and consequences for non-compliance.

AIgnite operates globally at the intersection of AI governance, gender equality, and women's leadership. Across the regions where our stakeholders and community are engaged, governance gaps are producing tangible harms. In practice, these failures are visible not only in regulation but also at deployment. Organizations increasingly adopt AI tools without clearly identifying the problem to be solved, assessing operational risks, or establishing accountability structures. This leads to misaligned use cases, unmanaged ethical and operational risks, limited return on investment, and weak accountability. It reflects a gap in implementation discipline and governance literacy, not only policy design. A major challenge is the absence of gender-disaggregated impact assessments in AI governance frameworks. The EU AI Act, while the most comprehensive binding AI regulation to date, does not require gender-specific risk assessments for high-risk systems. The Council of Europe Framework Convention references gender equality as a principle but leaves implementation to national transposition. The African Union Continental AI Strategy emphasizes inclusive development but remains in implementation planning. In practice, AI systems used in hiring, credit scoring, social protection, and criminal justice are being deployed without systematic attention to gender-differentiated impacts. Technology-facilitated gender-based violence is another acute governance gap. Globally, governance remains fragmented, with no widely coordinated framework addressing how AI-powered tools, including deepfake generation, surveillance technologies, and smart home ecosystems, are weaponized against women and girls. The opportunity is significant. The Dialogue can broker alignment where regional approaches are developing in parallel, connecting UNESCO's Women4Ethical AI platform, OECD/GPAI's work on transformative AI policy for gender equality, the AU's Women, Peace and Security engagement on AI-related harms, and the UN Working Group's 2026 report.

The Dialogue's greatest value lies in its universality. Unlike the OECD, G7, or regional bodies, it is the only AI governance forum where every UN Member State has a seat alongside civil society, academia, and the private sector. This positions it to play three roles that no other mechanism can. First, the Dialogue can serve as a "dialogue of dialogues" by synthesizing and connecting existing governance initiatives rather than duplicating them. The current landscape includes the OECD AI Principles, G7 Hiroshima Process, EU AI Act, Council of Europe Convention, AU Continental AI Strategy, UNESCO Recommendation on the Ethics of AI, and national strategies. The Dialogue can identify where these instruments align, where they conflict, and where gaps remain, particularly for populations and issues that existing mechanisms have not adequately addressed. Second, the Dialogue can ensure that developing countries are not passive recipients of governance frameworks designed elsewhere. Resolution A/RES/79/325 was adopted with an explicit commitment to bridging AI divides. The Dialogue should operationalize this by creating structured pathways for countries at varying levels of technological development to contribute to norm-setting, not merely adopt norms set by others. However, structural barriers remain significant. AI development continues to outpace regulatory processes, while countries facing poverty, inequality, or entrenched social hierarchies may struggle to prioritize AI governance alongside immediate economic needs. This creates a widening gap between global governance ambitions and local implementation capacity. Third, the Dialogue can advance cross-cutting issues that no single thematic forum owns, including gender-responsive AI governance. It can move fragmented best practices toward coordinated international commitment, linking gender equality, human rights, safety, accountability, and capacity-building.

The Dialogue should build upon rather than duplicate the following established frameworks and initiatives. On standard-setting, the UNESCO Recommendation on the Ethics of AI (2021), adopted by all 193 Member States, provides the most comprehensive global normative framework, including specific policy provisions on gender. The Council of Europe Framework Convention on AI and Human Rights, Democracy and the Rule of Law (opened for signature September 2024) is the first legally binding international AI treaty, with explicit gender equality obligations. The EU AI Act provides a risk-based regulatory model now entering enforcement. The OECD AI Principles, updated in 2024, offer a widely adopted governance baseline. These instruments represent significant normative capital that the Dialogue should reference and build upon rather than renegotiate. On capacity-building, the African Union Continental AI Strategy (endorsed July 2024), supported by UNESCO, provides a continental framework for inclusive AI development. ASEAN's Guide on AI Governance and Ethics offers a voluntary regional model for Southeast Asia. The Global Partnership on AI's Responsible AI Working Group has produced specific recommendations on gender equality and diversity in AI ecosystems. The ITU's AI for Good platform, provides an established convening space for linking AI governance to sustainable development objectives. On gender-responsive governance specifically, UNESCO's Women4Ethical AI platform, the OECD/GPAI's report on Transformative AI Policy for Gender Equality and Diversity, and the UN Working Group on Discrimination Against Women and Girls' forthcoming 2026 thematic report on AI all represent significant work that the Dialogue should draw upon. The Dialogue's added value is coordination and universality. No existing mechanism connects all these initiatives within a single forum where every country and stakeholder group can participate. The Dialogue can serve as the connective tissue that turns fragmented progress into coherent international cooperation.

The Dialogue's value depends on whether it functions as a genuine multi-stakeholder platform or defaults to traditional intergovernmental plenary formats. AIgnite recommends several structural design choices. Governments should come prepared with concrete examples of governance approaches, their outcomes, and lessons learned, not with general statements of principle. The Dialogue should incentivize specificity by structuring sessions around governance challenges rather than thematic overviews. Civil society organizations, particularly those representing affected communities, should be integrated into substantive discussions, not confined to side events. Women-led and gender-focused organizations should have dedicated speaking roles within the thematic clusters, especially on safety, human rights, and socioeconomic implications. The European Union has called for the Dialogue to ensure cross-stakeholder mix in participation to avoid replicating existing silos, and AIgnite strongly supports this approach. The private sector should be expected to present not only innovation showcases but concrete accountability practices, including the results of gender and bias audits, transparency reports, and documented responses to harms identified in their AI systems. Academia and the technical community should contribute evidence on emerging risks and governance gaps, including on agentic AI, generative AI's gendered harms, and the effectiveness of existing regulatory approaches. On format, AIgnite recommends that the Dialogue include interactive breakout sessions organized around specific governance problems rather than broad themes, with mixed stakeholder composition in each session. The two-day format is limited, and sessions should produce concrete outputs: identified governance gaps, recommended priority actions, and specific commitments that can be tracked at the 2027 session. The Co-Chairs' summary should be structured around actionable findings rather than descriptive summaries of proceedings.

Several communities remain structurally underrepresented in AI governance discussions, and the Dialogue must take deliberate steps to correct this. While inclusivity is widely emphasized in global discussions, in practice it remains a work in progress, particularly for women, children, the elderly, and low-income communities whose perspectives are still not systematically integrated into governance design. Communities in the Global South, particularly in least developed countries, small island developing states, and landlocked developing countries, lack the technical infrastructure, institutional capacity, and financial resources to participate meaningfully in global governance processes. Resolution A/RES/79/325 encourages financial support for developing country participation, and this should be operationalized through dedicated funding for travel, remote participation infrastructure, and translation services. Affected populations who experience AI harms directly, including women denied credit by algorithmic scoring, workers displaced by automation, and survivors of technology-facilitated gender-based violence, are rarely present in governance forums. Their lived experience is essential for ensuring that governance frameworks address real-world impacts rather than theoretical risks. Youth, particularly young women in developing countries, represent a critical but underheard voice. An estimated 1.2 billion young people are entering the global workforce, many of them girls, into labour markets being reshaped by AI. To include these voices, the Dialogue should establish dedicated participation tracks with funded seats, ensure that panels and breakout sessions include representatives of affected communities, and create structured input mechanisms beyond written submissions that accommodate varying levels of technical expertise and institutional capacity.

Traditional UN plenary formats, where delegations deliver prepared statements to largely empty rooms, will not produce the substantive exchange the Dialogue requires. AIgnite recommends several alternative formats. Problem-focused working sessions should replace broad thematic panels. Instead of a session on "safe and trustworthy AI," the Dialogue should convene sessions around specific governance problems, such as how should liability be allocated when an autonomous AI agent causes harm in a cross-border transaction, or what transparency obligations should apply to AI systems used in recruitment. This specificity forces concrete engagement and produces actionable outputs. Case study exchanges should feature concrete examples of governance approaches that have worked, failed, or produced unintended consequences. Countries and organizations should present real cases, including documented instances of AI-related harm and the governance responses that followed, rather than abstract frameworks. This builds a shared evidence base and allows mutual learning. Structured scenario exercises can be used to test proposed governance approaches against realistic situations. For example, participants could work through a scenario involving an AI-powered credit scoring system that systematically disadvantages women entrepreneurs in a developing country, and evaluate which existing governance tools would provide redress and where gaps remain. A "governance gap map" should be produced collaboratively during the Dialogue, identifying where existing instruments provide coverage and where international cooperation is needed. This visual, interactive output would give the Co-Chairs a concrete deliverable to inform their summary and the intergovernmental consultations mandated for the second Dialogue. Finally, the Dialogue should incorporate asynchronous engagement before and after the Geneva sessions, including digital consultation platforms and regional pre-sessions that allow stakeholders who cannot travel to Geneva to contribute substantively.

UNESCO's Recommendation on the Ethics of AI, adopted by all 193 Member States in 2021, provides the most comprehensive global normative framework, with specific policy provisions on gender equality. UNESCO has supported over 40 countries through its Readiness Assessment Methodology, with countries from the Global South leading implementation (CAIDP, Artificial Intelligence and Democratic Values 2026). UNESCO's Red Teaming Playbook offers a practical tool for testing generative AI for gender bias and technology-facilitated gender-based violence. The Council of Europe Framework Convention on AI (CETS No. 225) is the first legally binding international AI treaty, explicitly requiring that AI systems respect equality, including gender equality, and providing legal remedies for AI-related human rights violations. Its openness to non-European accession makes it a genuinely international instrument. Beyond Europe, AI legislation is advancing rapidly. Vietnam enacted a standalone risk-based AI law in 2025, effective in 2026. Peru's AI Law No. 31814 and its 2025 implementing regulation establish risk categories and governance obligations. South Korea enacted its AI Basic Act in 2025, with the law taking effect in January 2026. Japan drew directly on the G7 Hiroshima AI Process to formulate its national AI Act, demonstrating how countries can integrate international frameworks into domestic legislation. The African Union Continental AI Strategy (2024) provides a people-centred, development-oriented continental framework. ASEAN published its Responsible AI Roadmap (2025-2030) and Expanded Guide on AI Governance and Ethics, strengthening regional alignment with foundational governance frameworks. These examples demonstrate that effective AI governance is emerging across every region. The Dialogue's task is to connect these efforts, identify where they fall short, and build the international cooperation needed to fill the gaps.