"AI systems are often trained on noisy, unverified web data. Combined with their tendency to "fill in the blanks" when sources are lacking, this leads to hallucinations. As a result, companies looking to fine-tune AI systems for specific use cases often need access to large, high-quality datasets. By publicly releasing the MCP Server for its Data Commons, Google aims to tackle both challenges."
"Data Commons' new MCP server bridges public datasets - from census figures to climate statistics - with AI systems that increasingly depend on accurate, structured context. By making this data accessible via natural language prompts, the release aims to ground AI in verifiable, real-world information. "The Model Context Protocol is letting us use the intelligence of the large language model to pick the right data at the right time, without having to understand how we model the data, how our API works,""
Google's Data Commons aggregates public datasets from government surveys, local administrative sources, and international bodies such as the United Nations. The Model Context Protocol (MCP) Server exposes that structured data via natural-language prompts, enabling developers, data scientists, and AI agents to query real-world statistics directly. The capability aims to reduce reliance on noisy web crawls and lower hallucinations by grounding models in verifiable data and by supplying high-quality context for fine-tuning. Data Commons includes census figures and climate statistics among its datasets. MCP originated at Anthropic and has seen adoption by major AI providers including OpenAI, Microsoft, and Google.
Read at TechCrunch
Unable to calculate read time
Collection
[
|
...
]