AI File Renamer
Privacy Policy
The short version. AI File Renamer does not collect any personal data. It has no accounts, no telemetry and no analytics. Your documents, photos, the text extracted from them and everything derived from that text stay on your own computer. The application contacts the internet for exactly one purpose: downloading recognition models when you explicitly ask it to.
1. Who this policy is from
AI File Renamer is an independently developed desktop application. Because the app collects no personal data and operates no server, there is no user database, no profile and no account associated with your use of it.
2. What the application processes, and where
Everything the application does with your files happens locally, on the device where it is installed. Nothing is uploaded, and there is no remote service involved in the analysis.
- File contents. Documents and images you point the app at are opened, rendered and read (optical character recognition, or the PDF's own text layer) in memory on your machine.
- Derived information. Document type, dates, person names, photo descriptions and confidence scores are computed locally by models bundled with or installed into the application.
- Changes to your files. When — and only when — you press Apply, the app renames the files you accepted and writes metadata (such as title, dates and the original file name) into the files themselves or into a sidecar file next to them. File contents are never re-encoded, and no file is ever moved outside its folder or deleted.
3. What is stored on your computer
The application keeps its working data in your user profile, under
~/.filerenamer (on Windows, %USERPROFILE%\.filerenamer):
- Analysis cache (
index.db) — results of previous analyses, including extracted text, so that re-scanning a folder is fast. It can be cleared at any time from Settings. - Rename journal — the record that makes Undo possible: original and new file names for applied batches.
- Settings (
settings.properties) and optional custom naming keywords (lexicon.json). - Diagnostic log files — a rotating local log used for troubleshooting. It is never transmitted anywhere; if you want to send it to support, you attach it yourself.
- Downloaded models — the recognition and optional language models.
All of it is ordinary data in your own user profile. Deleting that folder removes it; removing the application from your machine ends the processing entirely.
4. Network access
The application makes no network requests while analyzing, renaming or describing your files. There is no background connection, no update ping, and the application itself performs no licence check.
To be precise about one thing, because it would otherwise be misleading: when you obtain the app from the Microsoft Store, Windows verifies your licence and manages any trial period. That check belongs to the Store and happens outside the application — we are not told who you are, and no such call is made by our code. Nothing about your files is involved in it.
The single exception is model installation, which you start deliberately by pressing a Download button in the application (or by running the setup command). Those downloads fetch files over HTTPS from:
| Host | What is downloaded |
|---|---|
| modelscope.cn | Text-detection and per-script text-recognition models (OCR) |
| huggingface.co | Optional local language and vision-language models |
| github.com | Optional CPU and GPU inference runtimes |
These requests download files only. No information about you, your files or your usage is sent with them beyond what any HTTPS download unavoidably reveals to the server you are downloading from — your IP address, the time of the request and the standard request headers your system sends. Those hosts are operated by third parties under their own privacy policies. Every download is verified against a pinned checksum before it is used.
5. What is never collected
- No telemetry, usage statistics or feature analytics.
- No crash or error reports sent off the device.
- No accounts, sign-in, email address or licence activation.
- No file names, file contents, extracted text or metadata transmitted anywhere.
- No advertising identifiers, no tracking, no profiling.
- No sale or sharing of personal information — there is none to sell or share.
6. Optional local language models
The application can use local language and vision models to improve difficult names and to describe photographs. These run as a process on your own computer, talking to the application over a connection that never leaves the machine (localhost). They are optional, can be disabled, and are not a cloud service — no prompt or document text is sent to any external provider.
7. This website
aifilerenamer.app is a static site. It sets no cookies, uses no analytics or tracking scripts, embeds no third-party fonts, images or widgets, and asks for no personal information. There is no form to fill in and nothing to sign in to.
Like any web server, the server that hosts these pages may write standard access logs — IP address, timestamp, requested page and browser user-agent string — for security and operational purposes. Those logs are not used to build any profile of you.
These pages are served from a single server in the EU, behind Cloudflare. There is no analytics, no tracking pixel and no cookie set by this site. Access logs rotate on the server's default schedule and are not exported anywhere.
8. Children
The application is a general-purpose utility and is not directed at children. It collects no personal data from anyone, of any age.
9. Your rights and how to reach us
Rights such as access, correction, export or erasure apply to personal data held by a service operator. No personal data about you is held anywhere outside your own computer, so there is nothing held here to request, correct or delete. The data described in section 3 is yours, stored in your user profile, under your control.
Support contact: support@aifilerenamer.app.
10. Changes to this policy
If this policy changes, the "last updated" date at the top of the page changes with it. Any change that would affect what the application does with your data will ship together with the release that makes the change, not before or after it.