Using AI to Make Open Data Useful: Lessons from the City of Boston

Governments at all levels now publish vast amounts of open data, but too often it’s hard for staff and residents to know which datasets to use—or even where to start. In this session, we’ll walk through how the City of Boston is using AI assistants  to search across its entire open data catalog at once, allowing users to ask questions in plain language without needing to know dataset names, formats, or schemas in advance.

Behind the scenes, this is made possible by MCP servers, which act as a secure bridge between AI tools and city data systems, allowing AI assistants to safely query live municipal data rather than relying on static files or manual uploads.

Using a live demo, participants will see how AI tools can help analyze 311 requests, permits, crime data, and other city datasets, making it possible to quickly surface patterns and insights that would otherwise require technical expertise or significant time. We’ll focus on what it takes to set this up, what problems it actually helps solve, and what other cities can realistically replicate. Participants will leave with concrete ideas for turning their own open data from static spreadsheets into tools that support faster, smarter public decisions.