The idea
Field Station AI is an open-source, local-first AI workspace that runs in a web browser for health, behavioral, and digital-research workflows. It lets researchers use browser-based language and speech models without sending chats or study files to a cloud AI service by default.
What it does
From a single HTML page, users can chat with a local assistant, ask questions about attachments, transcribe audio, classify text, summarize documents, combine spreadsheets, and search an optional compendium, a bundle of knowledge from many sources indexed for AI search. Models and compendiums download on first use and can run from the browser cache afterward. The whole application is one dependency-free HTML file served over HTTP.
Context
Field Station AI is designed for workflows that may involve sensitive data or protected health information, but institutional cybersecurity and IRB review may still be required before it is used with regulated data. The compendium it ships with is built by Extractium.
My contribution
I wrote Field Station AI as part of my work at the Depression Center's Mobile Technologies Core, which develops and maintains it. The project is university property and open source.