datablinder: Synthetic Look-Alike Copies of Sensitive Data Files

Turns a data file into a blinded copy: the same columns, the same R classes, the same file format, values that look roughly similar, and not one real value in it. Share the copy, get analysis code back, run that code on the real file.

Details

Everything in the package is built around one promise: code written against the blinded copy runs unchanged on the real data. Integers stay integers, a 0/1 column stays 0/1, dates keep their format, a factor keeps its number of levels, and no column ever changes class.

The functions

blind_file() reads a data file, blinds every column of every sheet and writes the copy next to the input in the same format.
blind_data() does the same to a data frame already in memory.
run_app() does it in a small app in the browser, on 127.0.0.1.
install_cli() puts a datablinder command where the shell can find it.

The first three all produce a blind_summary: a short description of the blinded data, and of it only, to paste into a chat along with the file.

The options

Five, the same everywhere: blind_names, keep_labels, keep_real, rows and seed, described in blind_data(). Detection thresholds are internal constants, and there is no configuration file.

Limits

This is not a formal privacy guarantee. It reduces the risk of disclosing the real values. The column names, the classes, the row count, the share of missing values, the number of categories and the broad shape of every distribution all stay visible, on purpose, because that is what makes the code work.
Read the summary before sharing the file. It is the list of decisions the package made, and a column blinded as a category when it is really an identifier is the case to catch.
keep_real turns the promise off for the columns it names: those are shared exactly as they are. A column that is harmless by itself can still identify someone next to the others, so the summary names every column kept, and the leak check says it cannot vouch for them.
The blinded data is for writing code, not for analysis. Columns are blinded independently, so no correlation, model coefficient or cross-tabulation on the copy means anything about the real data.
No network access, and no AI or LLM calls, anywhere in the package. Sharing the blinded file with an AI is a thing you do afterwards, deliberately.

Author

Maintainer: Alex Gallinat baloo.142@gmail.com [copyright holder]

Other contributors:

Zobaer Ahmed (author of FakeDataR; see inst/COPYRIGHTS) [contributor, copyright holder]