UN partners with Google to make its global statistics easier for AI

In short: The United Nations is working with Google on a new system that makes UN statistics easier to search and easier for AI tools to use.

What happened

The UN announced a new platform called the UN System Data Commons. It is meant to help people find official statistics from across UN agencies, like health, education, and poverty numbers.

The new platform is built on Google’s open source Data Commons project. Open source means the underlying software is publicly available to inspect and reuse. The UN says 26 UN entities have committed to the effort, with data from nearly 20 available at launch, and a goal of bringing 80% of the UN system’s statistical datasets onto the platform by 2027.

A key change is that it supports natural-language queries, which means people can type a question in normal everyday wording, instead of digging through database menus. It also supports the Model Context Protocol, or MCP, which is a way for AI systems to connect directly to outside data sources (like giving an assistant a direct line to an official reference book, instead of guessing from memory).

Google.org provided $2 million in funding and technical support to help set up the core system. Google says the platform is hosted on an instance governed by the UN and is intended to be maintained by the UN over time.

Why it matters

The UN points to a problem many people run into today, AI tools often do a poor job with official numbers. UNICEF tested six popular AI models and found an average accuracy score of 21.2% on questions about global development statistics, and about three in five answers did not provide a usable number. The UN’s new setup aims to make it easier to get a number and also trace where it came from, but Google warns humans should still review any AI-made charts or written conclusions.

Source: TechCrunch AI