THIS SOFTWARE IS UNDER ACTIVE DEVELOPMENT, USAGE IS CURRENTLY ONLY ALLOWED FOR TESTING PURPOSES.
Hosted service

Welcome to PII Checker Online

PII Checker Online is a hosted service for data editors to scan replication packages — already published, or about to be made publicly available — for personally identifiable information, with no infrastructure of your own required. Classification is performed by an open-source model we run ourselves; your data is never sent to a third-party AI company.

For usage on your own replication packages prior to publishing, see Run it yourself.

How it works
1

Upload a package

A zip archive or an individual data file.

2

Extraction and screening

Files are unpacked, duplicates identified, and every column screened against rule-based criteria to decide which are worth checking.

3

Classification

Each candidate column and its values are evaluated for the presence of PII by a large language model. Classification is performed by an open-source model (Gemma 4) that we run ourselves on servers hosted on Google Cloud infrastructure — the data is not sent to an AI company to be processed.

4

Report

A downloadable spreadsheet, one row per checked column containing the classification.

Full guide →

Data retention & privacy

Your data are stored only for as long as necessary.

Uploads & extracted files Deleted the moment a job reaches any final state — completed, failed, or cancelled. No delay.
Reports Kept up to one week and then deleted automatically. You can also delete the reports manually at any time.
No model training No uploaded data, tabulation, or reasoning text is used to train any model or shared with anyone else.

Full data, privacy & terms →

Learn more