UN turns to Google to make its global data ready for AI agents (techcrunch.com)

The United Nations has partnered with Google to launch the UN System Data Commons, a new platform designed to streamline access to global statistics for AI systems. Built on Google’s open-source Data Commons platform, this initiative allows users to perform natural language searches across various UN agencies, replacing the outdated UNData portal. The system supports the Model Context Protocol (MCP), which facilitates direct connections between AI systems and external data sources. The aim is to make official UN data more reliable and accessible as the demand for authoritative information grows in the AI landscape. This collaboration is significant for the AI/ML community, particularly as recent benchmarks from UNICEF revealed a staggering 78.8% inaccuracy rate among leading AI models when responding to questions about global development indicators. By ensuring 80% of UN statistical datasets are available on the new platform by 2027, the initiative aspires to enable AI to generate data-driven analyses, such as infographics, while maintaining traceability of original sources. Google's $2 million funding will enhance the platform’s infrastructure, emphasizing the importance of human oversight to review AI-generated information, given the potential for misinterpretation by models.