GSMA

The Dialogue should not act in isolation from existing AI convenings, e.g. the previous AI Summits; or other multilateral institutions. GSMA proposes that there should be representation and input from all sectors. There are already fora for inter-government discussion, but the Global Dialogue can add value by ensuring that there is sharing of lessons and knowledge between sectors. Private sector can be active partners, not only consultees. Success means meaningful inclusion of the Global South not just in participation, but in influence. This includes concrete commitments to fund, support, and institutionalise representation across regions, ensuring that governance frameworks reflect diverse socio-economic, cultural, and technological contexts. It should point to practical pathways for operationalising existing global principles. High-level frameworks must be translated into adaptable, grounded, real-world playbooks that countries and organizations at varying levels of digital maturity can implement without being excluded from progress. It is not recommended that the Dialogue intends to propose new regulation, as regulation is outpaced by AI technological development. However. the dialogue can encourage a shared vision and baseline for "minimum viable governance" (risk identification, accountability, human rights considerations, impact to environment/ society) that enable safe AI adoption while allowing flexibility for innovation and scaling. There should be alignment on shared risks, even if regulatory approaches differ. Issues like bias, misuse, exclusion, and harm are universal, and global consensus on these risks can serve as a foundation for coordinated action. Success would include mechanisms to facilitate continuous, cross-border collaboration, such as regulatory sandboxes, knowledge-sharing platforms, and real-time learning exchanges. Governance must evolve alongside AI, and this requires sustained cooperation not one-time dialogue. The dialogue must clarify the role of the private sector as active co-shapers of governance, not just implementers. Given their role in building and deploying AI systems, their engagement is critical to ensuring that policies are both practical and effective.

• Safe, secure and trustworthy AI
• AI capacity-building
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches

Safe, secure and trustworthy AI -Trust and confidence are prerequisites for widespread AI adoption, particularly in high-stakes sectors such as healthcare or justice for example. This is true for enterprises as well as individuals, especially as increasingly powerful models present greater risks, as well as benefits. Without trust, adoption will always be limited, regardless of how advanced the technology becomes. AI capacity building - It is critical to invest in capacity building to ensure policymakers' and regulators' actions are guided by best practice. This is also true for the work of international institutions. Capacity building is important for the Global South. Governance cannot be effective if entire regions are unable to meaningfully participate due to gaps in skills, infrastructure, or access. Inclusion starts with capability. Social, economic, ethical, cultural, linguistic and technical implications of AI - Reflecting, representing and understanding the importance of local culture, context and language on AI is critical for ensuring the technology benefits all people. It is particularly critical to ensure that AI bridges socioeconomic, regional and other divides, and doesn't deepen them. Interoperability of governance approaches - AI operates across borders, but governance remains fragmented. Harmonisation is fundamental to support cross-border and cross-sector cooperation, and avoid unnecessary compliance burden. Aligning on common principles, standards and enabling systems to work across jurisdictions is essential to avoid regulatory silos and inefficiencies.

While these priorities are critical, one area that remains underrepresented is sustainability, ensuring that AI systems are not only safe and inclusive today, but viable and responsible in the long term, keeping in mind their impact and reliance on energy resources.

The AI Dialogue can act as a neutral convening platform bringing governments, industry and civil society together to build trust and reduce fragmentation. It can drive alignment on shared risks and principles, even if countries take different governance approaches. Ensuring a shared governance framework based on the work of the OECD and UNESCO will allow for a shared basis of understanding. There should be an objective and means to translate global intent into local action/support.

The AI Dialogue should seek continuity of work with the OECD/GPAI. It should build upon, not recreate the OECD principles for ethical AI, and the UNESCO recommendation on the ethics of AI - both of which the GSMA has used to shape its responsible AI work and adoption by members. The added value of the AI Dialogue is to connect, align, and operationalise what already exists.

Some regions, countries or large tech companies, are very well represented, where others are not, in global discussions on AI governance. It is critical to create a forum which is globally representative, being conscious to include traditionally underrepresented groups and regions, including private sector companies across the AI infrastructure stack. Representatives from small and developing countries must be enabled to move from consultation to active participation and leadership.

The GSMA developed a Responsible AI Maturity Roadmap, ensuring the whole mobile industry aligns to a common approach to AI governance. It is a model of how industry can lead on responsible AI, and can be a reference for others to replicate. The roadmap provides a structured framework to establish, monitor and enhance responsible AI practices, allowing organisations to assess their current maturity, identify areas for improvement, and align their strategies with their ambitions. Developed through extensive industry consultation and grounded in international standards including the OECD guidelines and the UNESCO Recommendations on the Ethics of AI, it translates high level principles, fairness, transparency, accountability, human oversight, into measurable, evidence-based indicators organisations can act on. The roadmap also position governance as an enabler of growth, not a constraint on it.