The United Nations and Google have launched the UN System Data Commons, an AI-ready platform built on Google's open-source Data Commons that replaces the legacy UNData portal. The new system supports natural language search and the Model Context Protocol (MCP), enabling AI agents to directly query authoritative statistics from multiple UN agencies while maintaining source traceability. Twenty-six UN entities have committed to the initiative, with data from nearly 20 organizations available at launch and a target to include 80% of UN statistical datasets by 2027. Google.org is providing $2 million in funding and technical support for the core infrastructure. The platform addresses critical reliability gaps identified during UNICEF testing, where large language models achieved only 21.2% accuracy across 133,000 responses on global development indicators and showed significant inconsistency when queried repeatedly.