Tenzro Labs

A successful first Global Dialogue on AI Governance would help establish a practical and internationally inclusive foundation for long-term cooperation around AI systems that are increasingly autonomous, cross-border, and embedded into critical digital and economic infrastructure. Success should not be measured only by high-level principles, but by whether the Dialogue helps advance shared understanding around implementation challenges, technical standards, interoperability, accountability mechanisms, and governance coordination across jurisdictions. It would also be valuable for the Dialogue to create meaningful engagement between governments, researchers, civil society, technical builders, and infrastructure operators, particularly as many governance discussions remain disconnected from the operational realities of rapidly evolving AI systems. The emergence of AI agents and autonomous systems capable of coordinating workflows, accessing tools, conducting transactions, and interacting across platforms introduces new governance considerations beyond traditional model safety discussions. This includes identity, delegation, verification, auditability, policy enforcement, and cross-system accountability. A successful outcome would therefore include continued international collaboration toward interoperable governance approaches, open technical standards, and mechanisms that enable trustworthy AI deployment while preserving innovation, openness, and broad participation across regions and ecosystems.

• Transparency, accountability, and human oversight
• Open-source software, open data and open AI models
• Safe, secure and trustworthy AI
• Interoperability of governance approaches

Our selections reflect the increasing importance of governance infrastructure for AI systems that are becoming more autonomous, interconnected, and operationally embedded across industries and jurisdictions. Safe, secure and trustworthy AI remains foundational, particularly as AI systems move beyond passive assistance toward decision-making, coordination, and economic activity. Interoperability of governance approaches is increasingly important because AI systems operate globally across different legal, technical, and institutional environments. Fragmented governance models may create challenges around accountability, compliance, and coordination. Transparency, accountability, and human oversight are critical as AI systems gain greater autonomy and interact with financial systems, public infrastructure, enterprise workflows, and digital services. Governance mechanisms must increasingly support auditability, verifiable actions, delegated authority structures, and clear responsibility boundaries. Open-source software, open data, and open AI models also remain important for innovation, accessibility, transparency, scientific collaboration, and reducing concentration risks within the global AI ecosystem. Open standards and interoperable infrastructure can help ensure broader participation in the development of trustworthy AI systems.

One emerging area that may require greater attention is the governance of autonomous AI agents and machine-to-machine coordination systems. Much of today's governance discussion focuses on AI models themselves, while less attention has been given to the operational infrastructure surrounding autonomous systems, including identity, delegation, permissions, verification, settlement, and cross-platform coordination. As AI systems increasingly act on behalf of users, organizations, or other systems, new questions emerge around: * verifiable AI identity * delegated authority and permissions * accountability for autonomous actions * secure machine-to-machine transactions * auditability of AI-driven decisions and workflows * cross-border interoperability of governance and compliance systems Another important issue is the growing concentration of AI infrastructure, compute resources, and model access within a small number of organizations. Supporting open standards, interoperability, and broader infrastructure participation may become increasingly important for resilience, innovation, and global inclusion. Finally, governance discussions should consider how technical mechanisms such as cryptographic verification, trusted execution environments, programmable policy controls, and interoperable identity systems may complement regulatory and institutional approaches to trustworthy AI governance.

The rapid advancement of AI systems is creating both major opportunities and significant governance challenges across digital infrastructure, finance, enterprise operations, research, and public sector environments. One of the most important challenges is that existing governance models were largely designed for software systems with direct human control, while emerging AI systems are increasingly capable of autonomous coordination, decision support, workflow execution, and cross-platform interaction. This creates growing gaps around accountability, verification, transparency, delegated authority, and cross-border interoperability. Fragmented regulatory approaches across jurisdictions may also create operational and compliance complexity for organizations building globally interoperable AI systems. Greater international coordination and technical interoperability between governance frameworks will become increasingly important as AI systems operate across multiple legal and institutional environments. At the same time, there are substantial opportunities. AI has the potential to improve scientific research, productivity, environmental monitoring, healthcare, logistics, education, and access to digital services at global scale. Advances in open models, verification technologies, privacy-preserving computation, and interoperable digital infrastructure may also enable more transparent and accountable AI ecosystems. There is also an opportunity to embed governance directly into technical infrastructure through mechanisms such as verifiable identity, programmable permissions, cryptographic auditability, secure execution environments, and interoperable policy enforcement systems. From our perspective, one of the key priorities is ensuring that governance frameworks remain practical, internationally interoperable, innovation-friendly, and adaptable to increasingly autonomous and decentralized AI ecosystems.

The AI Dialogue can play an important role as a neutral and internationally inclusive platform for coordination between governments, technical communities, researchers, civil society, and industry participants working across different areas of AI governance. As AI systems increasingly operate across borders, sectors, and institutional environments, governance challenges are becoming inherently international. Issues such as interoperability, accountability, safety, verification, transparency, identity, and cross-border compliance cannot be effectively addressed through isolated national approaches alone. The AI Dialogue can help advance shared understanding around emerging governance challenges while also encouraging practical collaboration on standards, technical safeguards, governance frameworks, and interoperable approaches that support both innovation and trust. An important contribution of the Dialogue would be connecting policy discussions with operational and technical realities. Many governance conversations remain focused on principles, while rapidly evolving AI infrastructure introduces practical questions around autonomous systems, delegated decision-making, auditability, secure execution environments, and machine-to-machine coordination. The Dialogue can also help ensure broader global participation, particularly from emerging regions, open-source communities, smaller innovators, and technical infrastructure builders who may otherwise have limited influence in global AI governance discussions. Over time, the AI Dialogue could help support greater convergence around international best practices, technical interoperability, and collaborative governance mechanisms while preserving flexibility for different regional and institutional approaches.

The AI Dialogue should build upon existing international initiatives and standards efforts already contributing to AI governance, safety, interoperability, and digital trust. This includes work from the United Nations system, UNESCO Recommendation on the Ethics of Artificial Intelligence, OECD AI Principles, the Global Digital Compact, the G7 Hiroshima AI Process, the Council of Europe AI Convention, and ongoing efforts within standards organizations and technical communities focused on identity, interoperability, security, privacy, and trustworthy digital infrastructure. The Dialogue could also benefit from stronger engagement with open technical ecosystems, academic institutions, open-source communities, and infrastructure-level initiatives working on verification, cryptographic trust systems, interoperable identity, secure computation, and governance-aware AI architectures. One important added value of the AI Dialogue would be its ability to connect fragmented conversations across policy, technical implementation, regulation, standards, and operational infrastructure into a more coherent global discussion. Another important contribution would be encouraging interoperability between governance approaches rather than reinforcing isolated regulatory silos. As AI systems increasingly interact across jurisdictions and sectors, practical coordination between governance frameworks will become increasingly important. The Dialogue could also help surface emerging governance questions that are not yet fully addressed in existing frameworks, including autonomous AI agents, delegated machine actions, verifiable AI identity, machine-to-machine coordination, and technical accountability mechanisms for increasingly autonomous systems.

The AI Dialogue should encourage participation from governments, academia, civil society, technical builders, infrastructure operators, standards organizations, startups, enterprise operators, and open-source communities, as each contributes different perspectives on how AI systems are developed, deployed, governed, and integrated into society. To maximize effectiveness, the Dialogue should combine high-level policy discussions with technical and operational working sessions focused on practical implementation challenges, interoperability, governance mechanisms, and emerging infrastructure requirements. A multi-layered structure could be particularly effective: * strategic policy discussions * technical and standards-oriented sessions * sector-specific workshops * regional and cross-border coordination forums * open stakeholder consultations * collaborative working groups producing ongoing recommendations and outputs The Dialogue should also maintain continuity between annual meetings through open consultations, technical working groups, public submissions, and collaborative digital participation mechanisms. Importantly, the Dialogue should avoid becoming exclusively government- or large-company-led. Smaller technical teams, open-source contributors, researchers, and emerging ecosystem participants often play a significant role in shaping the future operational architecture of AI systems and should remain actively included.

Global AI governance discussions often underrepresent technical infrastructure builders, open-source communities, emerging regions, smaller innovators, interdisciplinary researchers, and organizations working directly on operational AI deployment outside major technology hubs. Many governance conversations are also concentrated around large frontier model providers, while less attention is given to the broader infrastructure stack surrounding AI systems, including verification, interoperability, digital identity, delegated authority systems, privacy-preserving computation, and decentralized coordination mechanisms. Emerging economies and smaller states may also face challenges participating consistently in global governance discussions despite being significantly affected by the economic and societal impact of AI systems. Inclusion could be improved through: * open consultation processes * remote-first participation mechanisms * publicly accessible technical working groups * multilingual participation support * stronger collaboration with academic and open-source communities * regional engagement programs * transparent public contribution processes * support for participation from smaller organizations and underrepresented regions The Dialogue should also encourage interdisciplinary participation across technology, governance, law, economics, ethics, security, sustainability, and public infrastructure, as AI governance increasingly affects multiple interconnected domains.

The AI Dialogue could benefit from more interactive and collaborative engagement formats beyond traditional panel discussions. Possible formats include: * technical governance workshops * cross-sector simulation exercises * collaborative standards sessions * scenario-based policy exercises * live interoperability demonstrations * multi-stakeholder roundtables * open technical forums * structured public consultations * collaborative drafting sessions for recommendations or principles Scenario-based exercises may be particularly valuable for exploring emerging governance challenges involving autonomous AI systems, cross-border coordination, AI-assisted decision-making, and machine-to-machine interactions in realistic operational environments. The Dialogue could also incorporate hybrid participation formats with digital collaboration platforms enabling broader international engagement before, during, and after the event itself. Another potentially valuable format would be thematic working groups that continue operating between annual Dialogue meetings and produce ongoing recommendations, research outputs, interoperability proposals, or technical guidance documents. Maintaining an open and iterative structure may help the Dialogue evolve alongside the rapidly changing technical and governance landscape surrounding AI systems.

Several emerging governance approaches demonstrate the importance of combining policy frameworks with technical mechanisms that enable accountability, transparency, interoperability, and verifiable trust in AI systems. The UNESCO Recommendation on the Ethics of Artificial Intelligence and the OECD AI Principles have helped establish globally recognized foundations around human rights, transparency, accountability, and responsible AI deployment. The EU AI Act and related regulatory sandbox initiatives also represent important efforts to balance innovation with risk-based governance approaches. At the technical level, growing work around interoperable digital identity standards, cryptographic verification, privacy-preserving computation, secure execution environments, and auditable infrastructure provides practical mechanisms for implementing trustworthy AI systems beyond policy commitments alone. Open-source ecosystems and collaborative standards initiatives also play an important role by encouraging transparency, interoperability, security review, and broader participation in AI development. Multi-stakeholder standards processes involving academia, industry, civil society, and technical communities can help create more globally interoperable governance foundations. Another promising area is the development of governance-aware infrastructure approaches where permissions, delegation rules, auditability, policy enforcement, and verification mechanisms are embedded directly into system architecture. This may become increasingly important as AI systems gain greater autonomy and operate across financial systems, enterprise workflows, digital services, and public infrastructure. Finally, international regulatory sandboxes and cross-border experimentation environments may help policymakers and technical builders collaboratively evaluate emerging governance challenges while supporting responsible innovation and interoperability across jurisdictions.