EnglishEspañol

AI Model Licensing: Legal Rules for Open-Source Attribution

Independently fact-checked against primary sources (last audited August 3, 2026). · 5 primary sources cited on this page. How we verify our legal content

AI Model Licensing: Legal Rules for Open-Source Attribution

Frequently Asked Questions

Can companies rebrand open-source AI models as their own?

Companies can modify and commercialize open-source AI models, but they cannot remove the original license notices or present the model as entirely their own creation. The specific requirements depend on the license. MIT requires retaining the copyright notice. Apache 2.0 requires preserving the NOTICE file and stating changes. Custom licenses like Kimi K2.5's Modified MIT require displaying the original model name above certain commercial thresholds, and licenses like the DeepSeek Model License require passing use restrictions to downstream users. Deleting license files does not discharge those obligations.

Are AI model weights copyrightable?

This is unresolved, and two separate questions are often confused. On whether weights can infringe someone else's copyright, the U.S. Copyright Office concluded in May 2025 that where a model can reproduce a memorized training example there is a 'strong argument' that copying the weights implicates the reproduction right for that example. On whether a developer owns an enforceable copyright in its own weights that it can assert against a downstream licensee, the Office did not address the question and no court has definitively ruled. That is why license enforcement currently proceeds on contract theories.

What happens if you violate an open-source AI license?

Potential consequences include breach-of-contract claims, copyright infringement claims where the copyright question is resolvable, injunctive relief, and statutory damages up to $150,000 per work for willful copyright infringement. In Doe v. GitHub, the two claims left in the district court are both breach-of-contract claims, though it is worth noting the defendants never moved to dismiss them rather than the court upholding them on the merits. That case is currently stayed: the plaintiffs' DMCA Section 1202(b) claims were dismissed on the ground that the statute reaches only identical copies, and the Ninth Circuit heard argument on that question on February 11, 2026, with a decision pending. Companies including Cursor and Rakuten have also faced reputational consequences and, in Cursor's case, congressional scrutiny.

Do the code and the model weights use the same license?

Often they do not, and assuming they do is a common and serious error. DeepSeek-V3 published its repository code under the MIT License in a LICENSE-CODE file while the weights were governed by a separate LICENSE-MODEL file containing the DeepSeek License Agreement, which adds use-based restrictions that must be passed to downstream users. The later DeepSeek-V3-0324 release states that the repository and the model weights are both licensed under the MIT License. Check both files, and record which release you used.

Does fine-tuning an open-source model create a derivative work?

Courts have not definitively ruled on whether fine-tuning creates a derivative work under copyright law, so treat the question as open. The practical point is that it may not matter for compliance: an attribution obligation in an open-source license is a contract term, and a contract term can bind you whether or not the derivative-work question is ever resolved in your favor. The conservative approach is to carry the original license terms forward with the fine-tuned weights.

What is copyright laundering in AI?

Copyright laundering describes feeding copyleft or attribution-required code or model weights into an AI system, where the provenance information falls away during processing, so the output appears free of license obligations. Developers including Jamie Tanna have used the term to describe it. The concept is central to the Doe v. GitHub Copilot lawsuit, where plaintiffs allege that Copilot emits code without the license notices attached to the code it was trained on.

Which open-source license is safest for commercial AI products?

MIT and Apache 2.0 are the most permissive and commercially friendly options. MIT requires only that you retain the copyright notice. Apache 2.0 adds a patent grant and requires stating modifications. GPL-licensed components are the hardest to reconcile with a proprietary product because derivative works must be released under the same license. Always read for custom clauses, such as Kimi K2.5's display requirement or the DeepSeek Model License's use restrictions, which add obligations beyond the base license.

Was Cursor sued for using Moonshot AI's Kimi K2.5 model?

No licensing lawsuit against Anysphere has been reported, and the company has since acknowledged that Composer 2 was built on Kimi K2.5. What did follow was a joint congressional investigation: on April 29, 2026, the chairmen of the House Committee on Homeland Security and the House Select Committee on the Chinese Communist Party sent letters to Anysphere and Airbnb seeking details about their use of Chinese-developed AI models, with the Anysphere letter focused on Composer 2 and the Moonshot AI model it was built on. That inquiry is a national security matter rather than a copyright enforcement action.

Updates

Updated with the April 2026 joint congressional investigation into Anysphere (Cursor) and Airbnb over Chinese open-weight AI models, and its July 2026 expansion to DoorDash. Corrected the discussion of the U.S. Copyright Office's model-weights analysis, clarified that the surviving claims in Doe v. GitHub are breach-of-contract claims that the defendants never moved to dismiss, and added that DeepSeek-V3 licensed its code and its model weights differently. Also brought Doe v. GitHub up to date: the DMCA Section 1202(b) claims were dismissed on identicality grounds and certified for interlocutory appeal in September 2024, the district court case is stayed, and the Ninth Circuit heard argument in February 2026 with a decision pending. Removed an incorrect statement that Windsurf had disclosed its own base model (Cognition's announcement does not name it), added Anysphere's later acknowledgment that Composer 2 was built on Kimi K2.5, corrected the date of final approval of the Anthropic settlement to July 20, 2026, noted the express limitation Judge Chhabria wrote into the Kadrey v. Meta fair use ruling, and reported Rakuten's later addition of a NOTICE file.

Independently fact-checked against the cited primary sources

Sources and References

  1. 17 U.S.C. Section 504 - Remedies for Infringement: Damages and Profits(copyright.gov).gov
  2. U.S. Copyright Office - Copyright and Artificial Intelligence, Part 3: Generative AI Training (Pre-Publication Version, May 9, 2025)(copyright.gov).gov
  3. U.S. Copyright Office - Part 3 report, full text (model weights and the reproduction right, pp. 28-30)(copyright.gov).gov
  4. European Commission - AI Act (Regulation (EU) 2024/1689) regulatory framework and transparency obligations(digital-strategy.ec.europa.eu).gov
  5. House Committee on Homeland Security - Chairmen Garbarino and Moolenaar Announce Joint Investigation into National Security Risks Posed by PRC AI Models (April 29, 2026)(homeland.house.gov).gov
  6. Doe 1 v. GitHub, Inc., No. 4:22-cv-06823-JST (N.D. Cal.), Order Granting in Part and Denying in Part Motion to Dismiss (Tigar, J., January 2024, ECF No. 195)(skadden.com)
  7. Bartz v. Anthropic PBC, No. C 24-05417 WHA (N.D. Cal.), Order on Fair Use (Alsup, J., June 23, 2025, ECF No. 231)(copyrightalliance.org)
  8. MIT License Full Text - Open Source Initiative(opensource.org)
  9. Apache License 2.0 Full Text(apache.org)
  10. GNU General Public License v3.0(gnu.org)
  11. Moonshot AI - Kimi K2.5 Modified MIT License (100M monthly active user / $20M monthly revenue display threshold)(github.com)
  12. DeepSeek-V3 LICENSE-CODE - MIT License covering the repository code(github.com)
  13. DeepSeek-V3 LICENSE-MODEL - DeepSeek License Agreement v1.0 (23 October 2023) covering the model weights, with downstream use restrictions(github.com)
  14. DeepSeek-V3-0324 model card - repository and model weights licensed under the MIT License (March 2025)(huggingface.co)
  15. Nextgov/FCW - House panels probe Airbnb, Anysphere over use of Chinese AI models (April 2026)(nextgov.com)
  16. South China Morning Post - US lawmakers investigate DoorDash's use of Moonshot AI's Kimi K2.6 model (July 2026)(scmp.com)
  17. The Bookseller - Around 10,000 authors protest AI commercial research exception with publication of empty book (March 2026)(thebookseller.com)
  18. Venable LLP - District court certifies the DMCA Section 1202(b) identicality question for interlocutory appeal in Doe 1 v. GitHub, No. 22-cv-06823-JST (September 27, 2024)(venable.com)
  19. eWeek - Cursor acknowledges Composer 2 was built on Moonshot AI's Kimi K2.5 (March 2026), including Aman Sanger's statement and Lee Robinson's compute breakdown(eweek.com)
  20. Cognition - SWE-1.5 announcement, describing the base model only as "a strong open-source model" without naming it(cognition.com)
  21. Simon Willison - SWE-1.5 writeup noting Cognition did not share details of the base model, with a later update attributing it to GLM (October 29, 2025)(simonwillison.net)
  22. Authors Guild - Court grants final approval of the $1.5 billion Bartz v. Anthropic settlement (Judge Araceli Martinez-Olguin, N.D. Cal., July 20, 2026)(authorsguild.org)
  23. Goodwin - Analysis of Kadrey v. Meta (Chhabria, J., June 25, 2025), including the ruling's express limitation and the market-dilution theory(goodwinlaw.com)
  24. BigGo Finance - Reporting that Rakuten AI 3.0's initial release omitted the DeepSeek license file and that attribution was later added as a NOTICE file (March 2026)(finance.biggo.com)
  25. Jamie Tanna - Worries about Open Source in the age of LLMs, using the phrase "copyright laundering" (November 17, 2025)(jvt.me)
Share: