International Electrotechnical Commission IEC

We would consider the Dialogue a success if it promotes open conversations and the exchange of views among all stakeholders, ensuring that all voices are heard equally. The Dialogue will be successful when it concludes with nuanced summaries of the conversations that provide an overview of the different relevant stakeholders in the governance of global AI, including International Standards Development Organisations (SDOs), in advancing AI for an inclusive, open, sustainable, fair, safe and secure digital future for all.

• Safe, secure and trustworthy AI
• Interoperability of governance approaches
• AI capacity-building
• Transparency, accountability, and human oversight

International standards and conformity assessment provide a way to ensure the safety of AI systems across borders. Developed through a one-country-one-vote system, international standards are created through global consensus, ensuring that global AI governance approaches can be operationalised in every country. These cover all focus areas (interoperability, safety and transparency), and the IEC is committed to supporting capacity building in the use and value of international standards for global AI governance.

AI is a heavy energy consumer, so the impact on power infrastructure and the environment should be given additional focus. AI has a strong infrastructure component, in which IEC standards can help with grid integration, energy efficiency and data centre efficiency, among other things.

In response to international conversations and the Global Digital Compact, the IEC, together with ISO and ITU, has launched the Seoul Statement. The Seoul Statement makes the following commitments: 1) actively incorporating the socio-technical dimension into standards development; 2) deepening the understanding of the interplay between international standards and human rights, recognising their importance and universality; 3) strengthening an inclusive, dynamic multistakeholder community to develop and apply international standards for the design, deployment and governance of AI; and 4) Enhancing public-private partnerships for AI capacity building. These three international SDOs are in a core position to strengthen AI governance in order to advance sustainable development and enable everyone and every sector of society to benefit from the AI revolution.

The Dialogue can identify and promote the common factors that are true at a global level with regard to AI governance. It can ensure that all voices are heard equally, which is fundamental to international cooperation. The Global Dialogue can reveal that similar ideas underpin AI governance across the globe. Similarly, international standards and conformity assessment systems are considered state-of-the-art international practice for ensuring the safety and trustworthiness of AI, which facilitates global trade.

The UN Global Dialogue is unique in form and fulfils its purpose under the auspices of the UN. It could learn from the SDG review process and propose global goals that could be iteratively reviewed in a similar way.

The Dialogue format, including high-level formats, should be multi-stakeholder, involving not only high-level government representatives, but also other stakeholders. All formats should be multi-stakeholder and include representatives from international standardisation organisations to provide a complete picture of stakeholder AI governance. There should be an understanding of the differences and communalities of UN-SDOs and SDOs with a one-country-one-vote system, which are not UN agencies and whose members are National Committees, not States.

Although international standards and conformity assessment systems based on ISO/IEC standards for AI are the most widely used, their representatives are underrepresented in international governance discussions. It is therefore of great importance that the IEC and ISO are represented at all levels, including the highest, during the AI Global Dialogue.

Hybrid participation should be possible to ensure that contributors from developing countries are equally represented.

Effective AI governance operates across five interconnected levels: international law and binding treaties; soft law instruments such as intergovernmental principles and recommendations; national and regional regulation with de facto international reach; international technical standards and conformity assessment; and technology-driven norms shaped by the design choices of dominant platforms. These levels must work in congruence. Incoherence between them produces regulatory fragmentation, duplication, and gaps that undermine trust. International technical standards and conformity assessment occupy a particularly important position within this architecture. Developed through global consensus - with participation from more than 150 countries on a one-country-one-vote basis - IEC standards translate high-level policy principles into measurable, auditable requirements across the full AI lifecycle. Independent conformity assessment provides objective evidence of compliance that is recognised across borders, reducing costs and enabling market access. Yet standards cannot stand alone. They derive their authority and effectiveness from the legal frameworks above them, and their real-world impact from the regulatory and market forces around them. Their power lies precisely in how they connect and give operational substance to the other four levels. The IEC calls on all stakeholders to treat international standardisation and conformity assessment not as a technical afterthought, but as a foundational pillar of any coherent global AI governance framework.