Anonymize IFC

Scrub authors, organizations, contacts and file provenance so the model is safe to share.

Scrub the identifying metadata out of an IFC model so it can be shared under an NDA, published, or handed to a party who shouldn't learn who produced it. Upload the model, choose how far to go, and download a file with the authorship, contact details, project identity and file provenance removed.

An IFC carries more about its authors than most people realise. Every element's ownership history names a person, an organisation, the software and often the machine. The STEP header records who exported it, with what, and when. Addresses and contact details ride along in the project and actor records. None of it shows in a viewer, and all of it travels with the file.

When to reach for it: Before sharing a model outside the project: strips authors, organizations, contacts and file provenance that RemovePsets can't reach.

What you get: A functionally identical IFC with persons, organizations, applications, addresses, project identity and the STEP header scrubbed; optionally with author psets deleted and GUIDs regenerated.

Example: Strip identifying data from a model before sharing it outside the project under NDA.

On by default: everything that can go without changing the model

The defaults are set for an NDA hand-off — identity-bearing data that can be blanked without altering what the model means is removed unless you turn it off.

Authorship — the person, organisation and application recorded in every element's ownership history.
Timestamps — creation and modification times zeroed, since a timeline is itself identifying.
Addresses — postal and telecom details attached to organisations and people.
Actors — the named participants recorded in the model.
Project identity — the project's own naming and identification.
The STEP file header — which no other tool here touches, and which records the exporting application, the author line and the originating system.

Opt-in: the ones that break things

Three further passes are available and OFF by default, because each destroys something a downstream consumer may still need. Document-reference deletion removes links to external documents rather than just anonymising them. Property-set deletion removes sensitive property sets wholesale. GlobalId regeneration replaces every element identifier with a fresh one.

That last one deserves care. New GUIDs make the model unlinkable to any previous issue of it — which is exactly the point when the concern is correlation between two releases, and a disaster when anyone holds BCF issues, clash results or a revision history keyed to the old identifiers. Every one of those becomes unresolvable. Decide deliberately.

It stays valid IFC — and it is honest about its limits

Anonymising by writing nulls into mandatory attributes produces a file that fails schema validation, which helps nobody. So mandatory fields get a neutral placeholder instead: the organisation name, the application strings, and the IFC2x3 ownership chain that the schema requires are kept structurally intact while carrying no information. The result opens everywhere the original did. The operation is idempotent, so running it twice is harmless, and element identifiers are stable unless you asked for regeneration.

The honest caveat: this removes the identity data IFC has schema slots FOR. It cannot know that a custom property called "Notes" contains a designer's name, or that a free-text field carries a phone number, because those are unstructured text under labels no engine can interpret. Anonymise, then check what's left — Extract IFC Data exports the property values to a spreadsheet where you can read them, and Delete IFC Properties removes whatever you find.

How to anonymize an IFC model

Upload your IFC file
Choose what to scrub (defaults cover the NDA-handoff case)
Click Run to anonymize
Download the scrubbed IFC

Under the hood

BIMCamel parses IFC files in the browser using the open-source web-ifc engine and renders models with @thatopen/components + three.js. Heavy operations (clean / optimise / validate / convert) run on disposable server-side workers using the same web-ifc stack plus our own .NET pipeline; results stream back to your browser as soon as they're ready. Uploads sit on our servers only as long as your tier's retention window allows and are never used to train AI.

Frequently asked questions

How do I remove the author name from an IFC file?

Run this tool — authorship scrubbing is on by default and clears the person, organisation and application recorded in every element's ownership history, plus the STEP file header that records who exported it and with what.

What does it remove by default?

Authorship, timestamps, postal and telecom addresses, named actors, project identity and the STEP file header — everything identity-bearing that can be blanked without changing what the model means. Document-reference deletion, property-set deletion and GUID regeneration are opt-in.

Should I regenerate the GlobalIds?

Only deliberately. New identifiers make the model unlinkable to any previous issue, which is the point if you're worried about correlating two releases — and a disaster if anyone holds BCF issues, clash results or a revision history keyed to the old ones. All of those become unresolvable.

Will the anonymised file still be valid IFC?

Yes. Mandatory attributes get a neutral placeholder rather than being nulled, and the IFC2x3 ownership chain the schema requires stays structurally intact. The file opens everywhere the original did, and running the tool twice is harmless.

Is anything guaranteed to be missed?

Free-text custom properties. The tool removes the identity data IFC has schema slots for; it can't know that a property called "Notes" contains a name or a phone number. After anonymising, export the properties with Extract IFC Data to read what's left, and remove anything sensitive with Delete IFC Properties.

Which IFC versions are supported?

IFC2x3 and IFC4 — the two schema versions exported by essentially every AEC authoring application. IFC4x3, the infrastructure schema, works too: it opens in the viewer, and validation, health reports, property-set editing and georeferencing all run on it. Tools that need geometry processing, such as clash detection and 3D export, aren't verified on IFC4x3 yet.

What's the maximum IFC file size?

Up to 25 MB as a guest, 100 MB with a free account, 500 MB on Pro, and 1 GB on Team. The caps are on upload size only — every tool itself is available on every tier.

What happens to my file after processing?

Your file is uploaded over HTTPS and processed on our servers. Guest uploads are stream-only — the working files are deleted as soon as your download finishes. Free accounts keep results for 7 days; Pro keeps them 90 days and Team 180 days. Files are never shared with third parties or used to train AI.

Is Anonymize IFC free?

Yes. Anonymize IFC is free to use — guests get 3 runs a day with 25 MB uploads, and a free account raises that to 10 runs a day with 100 MB uploads. No trial clock, no watermarks.