The United Nations is working with Google to make its global statistics easier for artificial intelligence systems to access and use. The new platform, called the UN System Data Commons, connects data from multiple UN agencies with AI tools while keeping the sources traceable.
The platform is built on Google’s open-source Data Commons technology and allows users to search UN statistics using natural-language queries. It replaces the older UNData portal, where users primarily relied on traditional database searches to find information.
The new system is available through the UN System Data Commons and supports the Model Context Protocol (MCP), which lets AI systems connect directly to external data sources.
Making UN Data More Accessible for AI Systems
The move comes as organisations are looking for ways to improve the reliability of AI-generated answers. UNICEF testing of several large language models found that AI systems often struggled to provide accurate figures when answering questions about global development indicators.
According to UNICEF chief statistician João Pedro Azevedo, the benchmark reviewed more than 133,000 responses across models from OpenAI, Anthropic, and Google. The average accuracy score was 21.2%, with many responses failing to provide a usable number.
The UN said 26 of its entities have committed to the Data Commons project, with data from nearly 20 organisations available at launch. The organisation aims to move 80% of UN statistical datasets onto the platform by 2027.
Built for AI Agents With Data Transparency
Google.org provided $2 million in funding and technical support to help establish the platform’s infrastructure. Google said the system operates on a UN-governed instance and is intended to be maintained and scaled independently by the organisation over time.
The platform is based on Google’s Data Commons, which was launched in 2018 to organise public datasets from different sources into a common framework.
In addition to allowing AI systems to retrieve statistics, the platform records where each figure comes from, allowing users to trace information back to the original UN source. Google demonstrated how AI systems connected through MCP could combine multiple indicators to create charts, dashboards, and written analysis.
However, Google noted that access to reliable data does not guarantee that AI-generated conclusions will always be correct. Human review remains necessary before using AI-generated analysis in published work or official decisions.
