Armored Quantum

A successful first Global Dialogue would do three things. First, it would shift the conversation from isolated AI issues to stack-level governance. The defining risks of this decade do not sit only inside models; they sit at the intersections of AI, identity, security, data, and emerging quantum capabilities. Success would mean broad recognition that governance must address these control points, not just individual tools. Second, it would produce practical convergence, not just principles. That includes a shared roadmap for high-priority areas such as trustworthy AI, human-rights protections, transparency and accountability, interoperable identity safeguards, and preparation for quantum-resilient security. This would align well with the Global Digital Compact's call for international AI governance in the public interest. Third, it would establish the Dialogue as a credible, inclusive coordination mechanism between governments, technical experts, industry, civil society, and underrepresented regions. The UN's stated aim is to ensure AI governance reflects the priorities of all countries, not only the most technologically advanced. Success would therefore mean meaningful participation from the Global South, smaller states, children's advocates, linguistic minorities, and security practitioners—not merely symbolic inclusion. In short, the first Dialogue will be successful if it makes clear that the future is not a winner-take-all contest between firms, but a governance challenge over the stack that shapes behavior, trust, and civilization-scale infrastructure.

• Transparency, accountability, and human oversight
• Protection and promotion of human rights
• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI

I selected these four priorities because they address the most urgent risks created by the emerging AI governance stack. Safe, secure and trustworthy AI is foundational because AI is now embedded in systems that influence decisions, access, and public trust. Safety must include not only model performance, but resilience, misuse prevention, and security-by-design. This is consistent with the UN's public-interest framing and with existing international work on trustworthy AI. The social, economic, ethical, cultural, linguistic and technical implications of AI matter because AI does not operate in a vacuum. It affects labor markets, education, cultural expression, language inclusion, and social cohesion. Governance that ignores these dimensions will be incomplete. UNESCO's Recommendation specifically emphasizes impacts on human lives, culture, communication, and the human mind. I prioritized protection and promotion of human rights because the convergence of AI with identity and surveillance capabilities can directly affect dignity, privacy, non-discrimination, freedom of expression, and remedy. UNESCO's AI ethics framework places human rights and dignity at its core. Finally, transparency, accountability, and human oversight are essential because opaque systems can shape outcomes without meaningful recourse. Human oversight must exist not only at deployment, but at the key control points where AI intersects with identity, access, and security. Without this, governance remains reactive rather than structural.

Yes. A major cross-cutting issue is stack control: the concentration of power at the intersections of AI, identity, data infrastructure, cybersecurity, and quantum transition. Much of today's governance discussion evaluates AI as if it were a standalone technology. In practice, the highest-impact risks arise when AI is linked to identity systems, authentication layers, behavioral data, and critical infrastructure. AI plus identity can shape behavior. Quantum plus identity will shape the future of security. Together, these convergences form civilization-scale infrastructure. Governance should therefore pay closer attention to the control points between systems, not only to the systems themselves. A second underemphasized issue is post-quantum readiness. As quantum capabilities develop, current cryptographic foundations may become more vulnerable, including the security of digital identity and long-lived sensitive data. The AI Dialogue should help connect AI governance discussions with broader digital trust and quantum-resilience planning. The Global Digital Compact's emphasis on digital cooperation and AI governance provides a useful foundation for this more integrated view. A third issue is behavioral sovereignty: the right of individuals and communities not to be invisibly manipulated by systems optimized to predict, nudge, and influence choices at scale. This concern sits across safety, rights, culture, and accountability, but deserves clearer treatment as an emerging governance priority.

In my sector, the main governance gaps appear where AI, identity, and security infrastructure converge. First, there is growing pressure to deploy AI quickly in high-stakes settings—cybersecurity, digital trust, public services, compliance, education, finance, and child-protection environments without equivalent maturity in oversight, auditability, or identity safeguards. This creates a mismatch between deployment speed and governance readiness. Second, fragmented governance is increasing risk. Safety, human rights, accountability, and security are often handled in separate policy streams, while the operational reality is highly integrated. In practice, AI systems are being connected to identity data, behavioral signals, and security controls faster than common governance frameworks are being built. Third, there is a widening preparedness gap between well-resourced institutions and those with limited technical capacity. This affects smaller organizations, lower-resourced jurisdictions, and communities whose languages and cultural contexts are not reflected in mainstream AI systems. As a result, benefits are unevenly distributed while risks can be concentrated. Finally, the lack of coordinated planning for future security challenges including quantum-resilient transition creates long-term exposure for identity, trust, and critical infrastructure. In sectors that depend on secure digital systems, governance gaps are no longer abstract; they affect confidence, resilience, and sovereignty. For these reasons, governance needs to move from model-level concern to system-level coordination.

The AI Dialogue can serve as the global coordinating layer for AI governance. Its most important role is to create a trusted space where governments and stakeholders can move from fragmented debate to shared operating understanding, especially on issues that cross borders, sectors, and technical domains. I appreciate that the UN has positioned the Dialogue as a universal and inclusive platform, which is critical because AI governance cannot be shaped only by a small set of technologically advanced actors. The Dialogue can also help translate broad principles into practical interoperability. Many frameworks already exist, but there is still a gap between norms and implementation. The Dialogue can connect human rights, safety, accountability, cultural inclusion, and technical governance into more coherent international action. A further role is to elevate shared risk awareness around emerging issues such as concentration of stack control, identity-linked harms, cross-border model deployment, and longer-term security implications. The Dialogue should help the international community govern not only AI systems, but the infrastructure and control points around them. Finally, the Dialogue can strengthen cooperation by encouraging capacity-building and mutual support, especially for countries and communities that are not yet fully represented in technical standard-setting or policy design. In that sense, it can become a bridge between principles, policy, and implementation, advancing international cooperation in a way that is inclusive, practical, and forward-looking.

The AI Dialogue should build on existing efforts rather than duplicate them. Key foundations include the Global Digital Compact, which explicitly frames AI governance as part of global digital cooperation; UNESCO's Recommendation on the Ethics of AI, which provides a widely adopted normative baseline centered on human rights, dignity, transparency, and oversight; and the OECD AI Principles, which offer a practical framework for trustworthy AI and have been updated to reflect rapid technological change. The Dialogue should also connect with the Independent International Scientific Panel on AI, so that policy discussions are informed by credible scientific and technical evidence, and with UNESCO's AI Ethics Observatory and readiness tools, which can support implementation and benchmarking. Its added value would be threefold: First, it can provide a universal political venue inside the UN system where states and stakeholders align across regions and levels of technical capacity. Second, it can connect currently separate tracks: ethics, human rights, safety, standards, digital inclusion, and security into a more integrated governance conversation. Third, it can surface gaps that existing mechanisms do not fully address, especially around stack-level risks such as AI plus identity, infrastructure concentration, and future quantum-security implications. In that sense, the Dialogue can act as the place where fragmented initiatives become coherent international governance.

Different stakeholders should contribute based on their distinct strengths. Governments should bring policy priorities, implementation realities, and regional perspectives. Technical experts and researchers should clarify capabilities, limits, and emerging risks. Industry should provide operational insight, deployment experience, and concrete commitments. Civil society, human-rights advocates, educators, youth, and affected communities should help ensure that lived experience, public interest, and accountability remain central. For structure, the Dialogue should combine high-level plenaries with smaller working sessions organized around specific governance challenges. I would recommend a format with: 1. A strategic plenary on global priorities and emerging risks 2. Thematic roundtables on safety, rights, accountability, inclusion, and security 3. Cross-cutting sessions on identity, infrastructure, and future trust challenges 4. Regional breakout sessions to capture context-specific realities 5. A synthesis session focused on practical next steps and areas of convergence 6. Panels of all ages including Youth panels are important so no demographic is left behind The Dialogue should also allow for written inputs, expert evidence sessions, and implementation showcases. To avoid staying purely conceptual, each track should be asked to produce a small number of concrete outputs: priority issues, governance gaps, examples of good practice, and proposals for cooperation. It would be great to have schools across the globe join in on the sessions from their classrooms. Most importantly, the structure should reward substantive participation, not only formal statements. The best Dialogue will be one where governments hear practitioners, technologists hear affected communities, and all participants engage not just on AI models, but on the wider stack shaping trust, identity, and power.

First, countries and communities with less technical infrastructure or less influence in standard-setting are often present only at the margins, even though they will be deeply affected by governance outcomes. Second, linguistic and cultural minorities are underrepresented, despite the fact that AI systems can reshape language access, cultural expression, and informational inclusion. UNESCO's work highlights the importance of cultural and linguistic dimensions in AI governance. Third, the field needs more participation from children's advocates, educators, disability communities, frontline public-interest practitioners, and communities exposed to digital exploitation or identity harms. These groups often see risks earliest, but are least represented in global forums. Fourth, more voices are needed from sectors responsible for trust infrastructure including digital identity, cybersecurity, and long-term resilience because AI governance is increasingly inseparable from those domains. Inclusion should not be limited to invitations. It requires structural support: multilingual participation, travel and remote-access support, preparatory briefings, regional consultations, open calls for evidence, and formats that do not privilege only large institutions. The Dialogue should also reserve speaking and drafting roles for underrepresented groups, not only observer roles. If governance is to be legitimate, it must reflect not only those building the systems, but also those whose rights, security, language, and daily life are shaped by them.

The most effective formats will be those that make the Dialogue interactive, evidence-based, and problem-centered. One strong option is to use scenario-based governance labs. Instead of discussing AI in the abstract, participants could work through realistic cross-border scenarios—such as identity-linked harms, AI deployment in public services, misinformation, child safety, or critical infrastructure risks. This helps reveal where current governance breaks down and where cooperation is needed most. A second useful format is a stack-mapping session, where participants identify risks and governance gaps across layers: models, data, identity, infrastructure, security, and oversight. This would reflect the reality that the most significant risks often arise between systems rather than inside a single tool. Third, the Dialogue could include regional listening sessions and community testimony segments to ensure that lived experience informs technical and policy discussion. Fourth, a curated set of implementation showcases could highlight policies, public-interest tools, audit mechanisms, and accountability practices that are already working. Finally, I would recommend an interactive synthesis format at the end of each day: a moderated session that identifies areas of convergence, open questions, and practical follow-up items. This would help prevent the Dialogue from becoming a series of disconnected statements. Innovative engagement matters because the value of the Dialogue will depend not only on who is in the room, but on whether the format helps participants jointly govern the future rather than simply describe it.

Several existing approaches offer strong building blocks for effective AI governance. At the international level, UNESCO's Recommendation on the Ethics of AI provides a concrete rights-based framework centered on dignity, transparency, fairness, accountability, and human oversight. UNESCO has also developed readiness and observatory tools that help translate principles into practice. The OECD AI Principles are another valuable example because they combine trustworthy-AI values with practical governance guidance and are designed to support interoperable policy development across countries. The OECD has also expanded practical guidance, including due diligence for responsible AI. The Global Digital Compact is important as a broader governance approach because it places AI within international digital cooperation rather than treating it as an isolated field. In practice, effective governance approaches tend to share several features: • risk-based oversight for higher-impact uses • human-rights impact assessment • transparency and documentation requirements • meaningful human oversight and review • independent audit or evaluation mechanisms • strong security and privacy protections • inclusion of affected communities in design and governance Going forward, the strongest solutions will be those that govern not only models, but also the wider stack-especially where AI intersects with identity, infrastructure, and digital trust.