# Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Copyright 2026 Mediar, Inc.

This repository distributes a derivative work — fine-tuned weights and
training-pipeline artifacts — built on top of the OpenAI Privacy Filter,
which is itself licensed under the Apache License, Version 2.0. See
[`NOTICE`](NOTICE) for the dual-license arrangement and the upstream
attribution, and [`LICENSE.upstream-apache2.txt`](LICENSE.upstream-apache2.txt)
for the full Apache 2.0 text.

The Derivative Work in this repository — the fine-tuned weights, the
README, the example code, and the model card — is licensed under the
Creative Commons Attribution-NonCommercial 4.0 International License
(CC BY-NC 4.0).

The Apache 2.0 obligations on the underlying base weights are preserved.
You comply with both licenses if you (a) keep the upstream attribution
and Apache 2.0 license text intact when redistributing this Derivative
Work, and (b) honor CC BY-NC 4.0 on this Derivative Work itself
(non-commercial use only, with attribution).

You are free to:

  * **Share** — copy and redistribute the material in any medium or format
  * **Adapt** — remix, transform, and build upon the material

Under the following terms:

  * **Attribution** — You must give appropriate credit, provide a link
    to the license, and indicate if changes were made. You may do so in
    any reasonable manner, but not in any way that suggests the licensor
    endorses you or your use.
  * **NonCommercial** — You may not use the material for commercial
    purposes.
  * **No additional restrictions** — You may not apply legal terms or
    technological measures that legally restrict others from doing
    anything the license permits.

Full license text: https://creativecommons.org/licenses/by-nc/4.0/legalcode

## What "non-commercial" means here

You MAY use this model for:
  - Personal use
  - Academic research and publication
  - Teaching
  - Internal evaluation at any organization (running it, comparing it
    to alternatives, integrating it into a research prototype)
  - Open-source projects that are themselves non-commercial

You may NOT use this model for:
  - Production redaction in a commercial product or paid service
  - Embedding in software you sell or charge for
  - Powering a SaaS / API offering
  - Any use where the model output contributes to revenue

## Commercial licensing

If you want to use this model commercially, contact:

    louis@screenpi.pe

Commercial licenses are available and reasonably priced. Including:
production deployment, redistribution rights, support, SLA-backed
inference, and custom fine-tunes for your domain.

## Citation

If you use this model in research, please cite:

    @misc{screenpipe-pii-redactor-2026,
      title  = {screenpipe-pii-redactor: a PII redactor for accessibility
                trees, OCR'd screen text, and computer-use traces},
      author = {{screenpipe}},
      year   = {2026},
      url    = {https://huggingface.co/screenpipe/pii-redactor}
    }

## Disclaimer

This model is provided "as is" without warranty of any kind, express or
implied, including but not limited to the warranties of merchantability,
fitness for a particular purpose and noninfringement. In no event shall
the authors or copyright holders be liable for any claim, damages or
other liability, whether in an action of contract, tort or otherwise,
arising from, out of or in connection with the model or the use or
other dealings in the model.

PII redaction is a hard problem with documented failure modes (see
`README.md` "Limitations" section). Do NOT rely on this model alone
for compliance with privacy regulations (GDPR, HIPAA, CCPA, etc.).
Validate on your own data before deploying.