The United Nations system has launched the UN System Data Commons, an open-source platform that consolidates statistics from across UN agencies into a single, AI-ready knowledge graph. The project is built on Google’s Data Commons technology and supported by Google.org through a grant to the UN Foundation.
According to Google, UN agencies have historically maintained some of the highest-integrity global data available, but that data has lived in disconnected silos with conflicting formats. Analysts often spent months manually reconciling datasets before any substantive analysis could begin. The new platform automatically normalizes metrics, timelines, and geographic boundaries so disparate datasets can be queried together.
Natural language access and validation
The platform supports natural-language search, allowing researchers, journalists, and policy analysts to ask plain-language questions (for example, on the relationship between clean water access and school attendance, or life expectancy trends across regions) and receive interactive visualizations in response. An Explore tab also allows browsing by location or theme, and a blog section publishes analysis such as UNICEF-sourced reports on child poverty reduction.
Google states that every dataset on the platform is validated with UN system statisticians and technical experts to keep responses grounded in official figures, though it recommends users review underlying sources before citing critical numbers.
Agentic AI access via MCP
The launch also introduces AI assistant capabilities built on the Model Context Protocol (MCP), an open standard that lets AI agents autonomously query the Data Commons, pull authoritative figures across domains, and assemble the results into charts, infographics, or draft reports without manual data-gathering.
For security and data-governance teams, the MCP integration is notable: it exposes a structured, machine-queryable interface to a large, cross-agency dataset that AI agents can access programmatically, a pattern increasingly common as AI-ready data platforms adopt agentic access standards.
Rollout plans
The UN system says it will continue adding datasets from additional UN entities over the coming year, with a stated goal of incorporating 80% of UN system statistical datasets by 2027. The platform is publicly accessible at data.un.org.