Philosoph-AI

Success would mean moving AI governance from principles to practice: recognizing that governance only works when accountability is clear, AI literacy exists across organizations, and decision-makers have usable structures to guide real choices, not just policies on paper.

• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches
• Open-source software, open data and open AI models
• Transparency, accountability, and human oversight

My priorities reflect a core finding of my research: AI governance only works when it becomes operational. Clear accountability, transparency, and human oversight must guide real decisions. Interoperable frameworks and responsible openness can help organizations move from principles to usable governance.

Yes. A major cross-cutting issue is organizational readiness for AI governance. Many institutions lack AI literacy, clear accountability structures, and practical decision processes. Without operational capacity, governance frameworks risk remaining symbolic rather than shaping real AI use and decisions.

Governance gaps are increasingly visible across organizations. In my ongoing research on operational AI governance, I have conducted more than 100 interviews with leaders, employees, and decision-makers across sectors. One of the most significant challenges emerging from these conversations is organizational readiness. Many institutions lack sufficient AI literacy, clear accountability structures, and practical processes to guide responsible AI use. As a result, governance frameworks often exist at the policy level but remain difficult to operationalize in everyday decision-making. This gap creates several risks. Organizations face uncertainty about when and how AI should be used, concerns about intellectual property and data protection, and potential reputational exposure if systems are deployed without adequate oversight. In practice, responsibility for AI-related decisions is often diffuse, which slows adoption and increases hesitation among leaders and employees. Governance becomes symbolic until accountability becomes personal and clearly assigned. At the same time, these challenges also present important opportunities. Organizations that invest in AI literacy, clear decision rights, and operational governance practices are better positioned to adopt AI responsibly and strategically. My research also shows that governance becomes effective when it is visible and usable in practice, embedded in workflows, leadership processes, and everyday decisions rather than remaining in high-level policies. Interoperable governance approaches and shared standards can further support organizations by reducing fragmentation between frameworks and helping translate principles into practical implementation.

The AI Dialogue can help bridge the gap between global principles and practical AI governance. While many frameworks define responsible AI at a high level, organizations often struggle to implement these principles in real decision-making. The Dialogue can support international cooperation by sharing practical governance approaches, aligning standards across countries, and reducing fragmentation between frameworks. It can also strengthen AI literacy and institutional capacity, helping organizations develop the structures needed for responsible AI use. Ultimately, its value lies in moving the conversation from principles to operational governance, enabling countries and institutions to learn from one another and implement AI governance more effectively.

Several important initiatives already shape the global conversation on AI governance, including the NIST AI Risk Management Framework, the EU AI Act, and international standards such as ISO/IEC 42001. These frameworks provide valuable principles, risk management structures, and emerging regulatory guidance for responsible AI. However, one persistent gap lies between high-level governance frameworks and how organizations implement them in practice. The AI Dialogue could add value by connecting these initiatives with insights from real-world organizational experience. In my research on operational AI governance, I conduct interviews with leaders, employees, and decision-makers to understand how governance frameworks are interpreted and applied inside organizations. A consistent finding is that governance often remains symbolic until accountability is clearly assigned and decision-makers have usable structures to guide their choices. The AI Dialogue could therefore help bridge academia, policy, and industry practice. By integrating research insights with practitioner perspectives, the Dialogue could surface practical lessons on what works, what fails, and why governance often struggles to move from principle to implementation. This would complement existing initiatives by focusing not only on framework design, but also on organizational readiness, AI literacy, and operational governance mechanisms. In this way, the Dialogue could create added value by strengthening the translation layer between global governance principles and real-world adoption, helping organizations implement responsible AI in ways that are both practical and accountable.

Different stakeholders should contribute their complementary roles: governments provide regulatory direction, industry shares implementation experience, researchers contribute evidence, and civil society ensures societal and human rights perspectives. The Dialogue should combine plenary discussions with smaller working groups focused on practical governance challenges. It should also include evidence from research and real organizational experience. Finally, the AI Dialogue should be an ongoing process, with follow-up collaboration and knowledge sharing to help translate governance principles into practice.

A key gap in global AI governance discussions is the voice of organizations that actually implement AI in daily operations. Much of the conversation is dominated by policymakers, large technology companies, and technical experts, while the perspectives of managers, employees, and practitioners responsible for using AI systems inside organizations are often missing. In my research interviews across sectors, these actors frequently describe governance as unclear or difficult to operationalize. Their experience is critical because they are the ones translating policies and frameworks into real decisions about AI use. Another underrepresented perspective is smaller organizations and institutions, including SMEs, public-sector teams, and organizations outside major technology hubs. These actors often lack the resources and expertise assumed in many governance frameworks. To address this gap, AI governance dialogues should include practitioner perspectives, structured consultations with organizations using AI, and mechanisms to capture insights from diverse sectors and regions. Integrating these voices would help ensure that governance frameworks are not only principled, but also practical and implementable.

Meaningful engagement requires active participation rather than passive listening. The AI Dialogue should prioritize in-person, small-group discussions where participants can exchange experiences and work through concrete governance challenges together. Instead of panels or surveys, sessions could be structured as facilitated working groups focused on specific questions (e.g., accountability, implementation of governance frameworks, AI literacy). Each group should have a clear objective and expected outcome, such as identifying practical governance mechanisms or implementation barriers. Participants from different sectors, government, industry, research, and civil society, should be intentionally mixed within these groups to encourage cross-perspective dialogue. Each session should end with a brief synthesis of key insights and actionable recommendations that can feed into the broader Dialogue. This format would create more dynamic, honest, and practical exchanges, helping move the conversation from general principles to real governance solutions.

Examples of initiatives and approaches that support effective AI governance include: OECD AI Principles (2019) → Example: Governments using them to shape national AI strategies and responsible AI guidelines. UNESCO Recommendation on the Ethics of AI (2022) → Example: Countries integrating human rights, inclusion, and ethical safeguards into national AI governance policies. EU AI Act (2024) → Example: Organizations classifying AI systems by risk level and implementing compliance measures such as documentation, risk assessment, and human oversight for high-risk AI systems. NIST AI Risk Management Framework (2023) → Example: Companies embedding AI risk identification and mitigation into product development and governance workflows. ISO/IEC 42001 - AI Management System Standard → Example: Organizations implementing structured AI governance systems, including documentation, accountability mechanisms, and continuous monitoring of AI systems. Montreal Declaration for Responsible AI (2017) → Example: Multi-stakeholder collaboration between academia, policymakers, and civil society to define shared ethical principles guiding AI governance. Organizational governance practices → Examples include internal AI governance committees, defined decision rights for AI deployment, and AI literacy programs to help employees understand capabilities, limitations, and risks.