UNESCO-led AI supervisors' network grows to 42 authorities, shifts to practice

A hands-on workshop on AI audits

The 4th UNESCO Global Forum on the Ethics of AI took place in Riyadh, Saudi Arabia, from 14 to 17 September 2026. There, GNAIS held a workshop on AI audit and evaluation led by Dr Gemma Galdón-Clavel, Usman Gohar and Alex Maagard from Eticas.ai. It combined presentations on AI auditing with a practical exercise, so participants could experience the audit process directly.

The session introduced the Eticas AI Risk Taxonomy and looked at how broad AI risks can be turned into concrete elements that can be assessed. The discussion centred on moving from identifying a risk to defining how it can be tested, what evidence should be collected and how findings can be communicated. Participants were shown a structured approach that breaks risks down into more specific mechanisms that can be tested.

The workshop also discussed the limits of simple pass or fail assessments, which can make it hard to understand how severe a finding is or where a system needs further attention.

A practical case involved an AI-based hiring tool. It illustrated several aspects of an AI system within an audit: the data the system uses, how the model is coded, how outputs are produced and checked, and how responsibility is assigned when a system fails. According to UNESCO, participants learned that an audit should begin with a clearly defined risk and follow a documented path. That path includes identifying what is being tested, selecting a methodology, examining evidence and producing findings that can be explained and compared.

Why international cooperation matters

On 16 September, GNAIS took part in a panel on international collaboration in AI supervision, moderated by Huub Janssen. It brought together GNAIS representatives from Cuba, Luxembourg, Saudi Arabia and Thailand.

The panel noted that AI systems and their supply chains often cross national borders, and that incidents can affect several jurisdictions at once. Panellists identified the sharing of knowledge, methodologies and experiences as a way to help authorities respond to common challenges. They also stressed building common approaches and trust between supervisory authorities and market actors.

According to UNESCO, greater consistency can support cooperation between jurisdictions and help avoid unnecessary duplication of work, while authorities continue to apply their own national legal frameworks. Capacity building was another key theme, particularly for countries still developing their AI regulatory and supervisory frameworks. Participants also underlined the value of sharing both successes and lessons learned.

Priorities for the second year

The network is led by its co-chairs, the Saudi Data & AI Authority (SDAIA) and Luxembourg's National Commission for Data Protection (CNPD), together with UNESCO. As GNAIS enters its second year, the Riyadh meeting reviewed the first year of operation and discussed priorities for the next phase. Members highlighted information sharing, international cooperation, capacity building, and greater alignment and harmonisation of approaches.

SDAIA pointed to the diversity of the network and the chance to grow it. Raneem Alkattan, General Manager of AI Governance at SDAIA, said: "We now have more than 30 members, representing a diverse group of countries, and we invite interested UNESCO Member States to join the Network and take part in the discussions."

What GNAIS is and how to join

GNAIS was officially launched at the 3rd Global Forum on the Ethics of AI in Bangkok, Thailand, in 2025. The initiative is led by UNESCO's Sciences Sector and forms part of implementing the UNESCO Recommendation on the Ethics of Artificial Intelligence, adopted by 193 Member States in 2021.

The network connects public supervisory bodies worldwide to build institutional capacity, strengthen international cooperation and promote oversight of AI systems. It offers a platform where supervisors can exchange practical experiences, access shared regulatory tools and receive specialised training. UNESCO says this helps countries navigate different legislative landscapes while pursuing a shared vision of trustworthy, ethical and human-centred AI.

Authorities interested in joining can contact the UNESCO GNAIS Secretariat at [email protected].

What it means for regulators and the market

For supervisory authorities, especially those still building their AI frameworks, the network offers shared tools, training and examples of audit practice. The emphasis on documented, comparable findings and on common approaches suggests that regulators want methods that work across borders without replacing national law. For companies deploying AI, the panel's call for trust between authorities and market actors points to more consistent expectations over time, though the source does not describe any binding rules.