Ninth Circuit affirms dismissal of DMCA attribution claims over AI-generated code

What happened

On 16 September 2026, the U.S. Court of Appeals for the Ninth Circuit issued a published opinion in Doe v. GitHub, Inc., No. 24-7700, affirming the district court's dismissal of programmers' Digital Millennium Copyright Act claims against GitHub (owned by Microsoft), Microsoft, and OpenAI entities. The plaintiffs — programmers who published open-source code on GitHub — alleged that GitHub Copilot and OpenAI's Codex reproduce their code without the attribution and license notices required by open-source licenses, in violation of 17 U.S.C. § 1202(b).

The panel held that the plaintiffs' "output" theory fails: Copilot and Codex do not "remove or alter" copyright management information from a copy of an existing protected work, but instead create new works that never contained that information. The court declined to consider the plaintiffs' "input" theory (removal of attribution at the training-data stage) because it was forfeited below. It did hold that the plaintiffs have Article III standing, and the breach-of-contract claims remain pending in the district court.

Why it may matter to you

If you publish code under open-source licenses, this is the first federal appellate ruling on whether AI-generated code outputs trigger DMCA attribution-removal liability — and the answer, for the output theory, is no. The forfeited input theory leaves open whether stripping license headers before training could still violate § 1202(b)(1) in a future case that properly pleads it.

What is not yet known

Whether the plaintiffs seek rehearing or Supreme Court review; how the pending breach-of-contract claims (open-source license terms) fare on remand; and whether another plaintiff can successfully litigate the input theory the panel declined to reach.