About CLAIR

What CLAIR is

CLAIR (Confidential Local Analytics Inference Runtime) is a desktop application for people who hold data they can't upload anywhere: grant data, IRB-protected data, student records, patient lists, internal finance. Upload a spreadsheet, ask questions in plain English, and get answers checked against your data, running on your own machine through a local AI model, not a cloud service.

It's built for non-technical researchers, analysts and administrators, not data engineers. No terminal, no query language, no cloud account.

Why it exists

Most tools that do this kind of analysis, AutoML, statistics, forecasting, dashboards, ask you to send your data somewhere first: a cloud API, a hosted platform, a vendor's servers. For a lot of real research data, that's not a workflow question, it's a hard no from an IRB, a data-use agreement, or an institution's own policy. CLAIR runs the whole stack locally instead: nothing about your spreadsheet, your questions, or your results has to leave the machine it's on.

Three things CLAIR is built around, in order:

• Provable local handling. The claim that your data stays on your machine isn't a policy promise, it's how the software is built, and it's meant to be checkable: disconnect the network and the analysis still works.
• Accretion over time. A file you import stays a file CLAIR tracks, not a one-time upload. Refresh it later and CLAIR tells you what changed; the work you've done on it compounds instead of starting over.
• Methodological correctness. A result that looks confident but is quietly wrong (a hidden total row, a leaked outcome, a search that never disclosed how many things it tried) is worse than no result. CLAIR is built to catch that class of mistake and say so, not just to produce an answer.

Who it's for

Researchers, analysts, and administrators who are not, and don't want to become, programmers, working with data they're responsible for keeping confidential. Their IT departments and their advisors are the other reader: people looking for a clear, honest answer to "where does the data go" before anyone signs off.

Where it is right now

CLAIR is in alpha, running with an invited university research cohort as its field test. It's free during the alpha. This period exists to find out, against real work and real data, whether the product holds up, not to hit a fixed release date.