University of Trento

First, it should establish a shared set of baseline principles—such as transparency, accountability, human oversight, and respect for fundamental rights—that can serve as a common reference across jurisdictions, without undermining regional specificities. Convergence at the level of principles is a necessary precondition for any meaningful coordination. Second, the Dialogue should deliver a roadmap for interoperability between existing and emerging regulatory frameworks. Rather than pursuing full harmonisation, which may be unrealistic, success would lie in identifying mechanisms for mutual recognition, alignment of standards, and cooperation among regulators, including in cross-border contexts.

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

First, ensuring safe, secure and trustworthy AI is a foundational requirement for the adoption and scalability of AI systems. Without adequate guarantees of technical robustness, cybersecurity, and reliability, AI systems risk generating harm, eroding public confidence, and undermining their own long-term viability. Trustworthiness, in particular, acts as a bridge between technical performance and societal acceptance. Second, the protection and promotion of human rights must remain at the core of any AI governance approach. AI systems increasingly affect fundamental rights, including privacy, non-discrimination, freedom of expression, and due process. Embedding human rights safeguards from the design phase is therefore essential to prevent misuse, mitigate risks of bias and discrimination, and ensure that technological innovation remains aligned with democratic values and the rule of law. Third, transparency, accountability, and human oversight are key enabling principles that operationalise both safety and rights protection. Transparency allows stakeholders to understand how AI systems function and make decisions; accountability ensures that responsibilities are clearly allocated and that remedies are available in case of harm; and human oversight provides a necessary safeguard against fully autonomous decision-making in high-risk contexts.

n/a

In the European context, and particularly in sectors such as public administration and space-related data governance, existing gaps in AI governance are becoming increasingly evident despite significant regulatory advances. On the one hand, the European Union has taken a leading role through comprehensive frameworks such as the AI Act and the GDPR, which embed principles of safety, human rights protection, and accountability. However, a key challenge lies in implementation and interoperability. Public institutions and operators often face difficulties in translating high-level legal requirements into operational practices, particularly when AI systems are integrated into complex, data-intensive environments such as Earth observation or satellite-based services. This is compounded by fragmentation across Member States and varying levels of technical and administrative capacity. Another significant challenge concerns transparency and accountability in practice. While legal obligations exist, ensuring meaningful explainability of AI systems—especially those relying on complex models—remains difficult. This can affect trust, hinder oversight, and create uncertainty regarding liability and responsibility, particularly in cross-border or multi-actor settings.

The AI Dialogue can serve as a neutral and inclusive platform to foster convergence across diverse regulatory and policy approaches. Its primary role should be to facilitate structured exchanges among governments, international organisations, industry, academia, and civil society, enabling the identification of shared priorities and common principles. Beyond dialogue, it can support the development of practical cooperation tools, such as interoperability frameworks, voluntary standards, and best practices, particularly in areas like risk classification, auditing, and oversight mechanisms. By promoting mutual understanding of existing regulatory models, the Dialogue can help reduce fragmentation and encourage forms of alignment that respect different legal and cultural contexts.

The AI Dialogue should build upon existing international efforts, including those led by the United Nations system, the OECD AI Principles, UNESCO's Recommendation on the Ethics of AI, and regional regulatory frameworks such as the EU AI Act. It should also connect with technical standard-setting bodies and multi-stakeholder initiatives that are already developing operational tools and guidelines. Rather than duplicating existing work, the added value of the AI Dialogue would lie in its ability to act as a coordination and bridging mechanism. It could bring together fragmented initiatives, facilitate information-sharing, and promote coherence between normative frameworks and technical standards.

Different stakeholders can contribute to the AI Dialogue by bringing their distinct expertise, experiences, and perspectives to bear on the development of governance frameworks. Governments and regulatory authorities can provide insights into policy objectives, legislative priorities, and compliance challenges, while industry actors contribute practical knowledge on AI system design, deployment, and operational constraints. Academia and research institutions offer rigorous analyses, evidence-based evaluations, and ethical guidance. Civil society and human rights organisations ensure that societal interests, fairness, and accountability remain central.

Several voices remain underrepresented in global AI governance discussions. These include stakeholders from developing countries, whose regulatory frameworks and technological capacities differ from those in high-income regions, as well as smaller enterprises and startups, which often face challenges in aligning with complex governance standards. Additionally, marginalized groups, including women, indigenous communities, and individuals with disabilities, are rarely fully integrated into policy conversations, despite being disproportionately affected by AI-driven decisions.

To foster meaningful engagement, the AI Dialogue could adopt hybrid formats that combine in-person plenaries with virtual participation, ensuring accessibility and global reach. Interactive workshops, scenario exercises, and role-playing simulations could help participants explore real-world challenges and ethical dilemmas. Multi-stakeholder hackathons or design sprints could generate concrete solutions to governance gaps.

Examples of policies, practices, and platforms that promote effective AI governance span regulatory frameworks, industry-led standards, and multi-stakeholder initiatives. At the policy level, the European Union's AI Act provides a risk-based regulatory framework that classifies AI systems according to potential harms, imposing requirements on transparency, human oversight, and accountability. Similarly, national strategies such as the United Kingdom's AI Strategy or Singapore's Model AI Governance Framework offer guidance on responsible AI adoption while fostering innovation.