UNESCO unveils AI readiness framework as countries move from principles to implementation

The 4th UNESCO Global Forum on the Ethics of Artificial Intelligence shifted from principles to implementation on its second day, with the organisation unveiling an updated AI readiness methodology and a global analysis of how countries are putting ethical AI governance into practice.

The UNESCO Readiness Assessment Methodology (RAM) 2.0) was presented alongside a new meta-analysis drawing on assessments from 55 countries. The tools are intended to help governments identify gaps in areas including regulation, institutions, skills, infrastructure and accountability.

UNESCO said the RAM has now been used in more than 75 Member States, with the Asia-Pacific region accounting for 12 completed assessments and three in progress as of August.

Sri Lanka was among seven jurisdictions highlighted at the forum as moving forward with RAM 2.0, alongside Uganda, Somalia, Comoros, Suriname, Zambia and Quebec. Sri Lanka has been working with UNESCO on the readiness methodology as part of a wider effort to strengthen government capacity around ethical and inclusive AI. Earlier this year, UNESCO also supported AI literacy training for almost 70 Sri Lankan civil servants from more than 30 government departments and agencies.

The new meta-analysis highlighted a significant gap between countries’ AI ambitions and their institutional capacity to implement them. Among the findings presented at the forum, 82% of countries reported policies integrating AI into education, but only 13% reported policies for training educators in AI ethics. Thirty per cent reported policies addressing women’s participation in the AI workforce.

RAM 2.0 is designed to make the assessment more closely aligned with UNESCO’s 2021 Recommendation on the Ethics of AI, while giving countries greater flexibility depending on their stage of AI readiness. It also expands attention to environmental impacts, gender, Indigenous peoples, people with disabilities and emerging AI risks.

Health was among the policy areas covered by the broader readiness framework. UNESCO’s analysis found that 69% of assessed countries have a digital health policy, but only 47% explicitly refer to AI, while 42% address both physical and mental health. Saudi Arabia was highlighted as a case study for integrating AI into health and social well-being.

The discussions also focused on the capacity of governments to turn ethical principles into everyday decisions. Speakers argued that ethical AI governance needs to be embedded in procurement, risk assessment, evaluation and institutional decision-making rather than treated as a one-off compliance exercise.

The forum, taking place in Riyadh from September 14–17, is co-hosted by UNESCO, the Saudi Data and AI Authority and the International Center for AI Research and Ethics. UNESCO says the event is bringing together governments, academia, civil society and industry to examine how ethical AI governance can be translated into practical action.