The United Nations has partnered with Google to make its vast collection of global statistics more accessible to artificial intelligence systems, through a new platform called the UN System Data Commons. This new system, built on Google’s Data Commons platform, will allow users to search for statistics from across UN agencies using natural-language queries, replacing the existing UNData portal. The platform supports the Model Context Protocol (MCP), enabling AI systems to connect directly to external data sources.

However, recent tests by UNICEF on six large language models found that, while AI tools are becoming increasingly popular, they still struggle to provide accurate data. Only 21.2% of the responses to questions about global development indicators were accurate, and many models hedged their answers, providing unreliable numbers. Additionally, when the same questions were run again on the same models, about half the time the models provided the same number.

The UN’s platform keeps track of where each statistic comes from, allowing users to trace data back to the original UN source. Google demonstrated how an AI system connected to the UN data through MCP could pull together multiple indicators and use them to generate dashboards, charts, and written analysis. However, Ramaswami of Google warned that models can misinterpret nuance, and a human should always review the outputs before citing or publishing them.

Despite the challenges, the UN aims to bring 80% of the UN system’s statistical datasets onto the platform by 2027, connecting 26 of its entities, with data from nearly 20 available at launch. The platform is hosted on a UN-governed instance and is intended to be maintained, operated, and scaled independently by the UN.