TL;DR: The Ninth Circuit affirmed the district court’s dismissal of the Digital Millennium Copyright Act DMCA 1202(b) claims in DOE v. GitHub, Inc., holding that outputs generated by AI tools such as Copilot do not violate the statute when they are new works that do not contain the copyright management information (CMI) from the underlying source. The court emphasized that DMCA liability hinges on removal or alteration of CMI from an existing copy, and that generative outputs without CMI do not qualify. The decision narrows the path for DMCA based challenges to AI generated outputs, while leaving open the possibility of related contract claims and, importantly for trial teams, clarifying the evidentiary focus in DMCA cases. Practitioners should adjust discovery and trial strategies accordingly, particularly in cases involving AI assisted coding or drafting, and consider how prompt logs and outputs intersect with potential DMCA and contract theories. Objection Academy’s training resources on AI prompts and objections can help trial teams sharpen their approach to AI informed disputes.
What happened and why it matters
On September 16, 2026 the United States Court of Appeals for the Ninth Circuit issued its opinion in DOE v. GitHub, Inc., a putative class action involving programmers who contend that AI assisted tools trained on GitHub content violated DMCA 1202(b). The panel affirmed the district court’s dismissal of the DMCA claims, ruling that generating new works from prompts without copyright management information does not violate the statute. The court’s analysis clarifies that DMCA 1202(b) liability attaches to the removal or alteration of CMI from an existing copy, not to the mere reproduction of outputs that lack CMI. The decision is notable because it closes a potential avenue for DMCA liability when AI produced code or text does not carry the original CMI, even though the underlying content may have influenced the output.
The Ninth Circuit’s ruling also confirms standing to bring a DMCA claim in this context but finds that, under the court’s “output theory,” the Copilot outputs at issue did not remove or alter CMI from the source work and therefore did not violate 17 U.S.C. § 1202(b). The court did, however, acknowledge three important aspects: first, the district court’s dismissal of the DMCA claims; second, that the contract claims remain for consideration; and third, that the question of whether an “input theory” could proceed was forfeited and not addressed on the merits in this appeal. These points are important for practitioners because they shape how to frame and preserve DMCA arguments in AI related disputes moving forward. See the Ninth Circuit opinion and summary. (cases.justia.com)
The case also provides concrete doctrinal guidance on what constitutes “identicality” under DMCA 1202(b). The court stressed that the statute protects against removing or altering CMI from copies, not against the generation of new works that never contained CMI to begin with. This distinction matters in practice when clients use AI to generate outputs that might resemble existing code or texts but do not reproduce the same CMI format. In practical terms, this narrows the DMCA toolkit for litigants who rely on outputs rather than direct copies, and it encourages litigants to scrutinize whether CMI has actually been removed or altered in the exact sense contemplated by the statute. (cases.justia.com)
Practical implications for trial teams
Narrowing DMCA liability in AI disputes: This decision signals that DMCA 1202(b) liability is less likely to attach to AI outputs that lack CMI, even when those outputs reproduce some surface features of training data. For trial teams, this means that DMCA 1202(b) claims against software vendors or AI service providers that rely on generative outputs should be carefully evaluated against the “identicality” and CMI removal standards. If the output is a newly generated work that does not carry CMI, DMCA claims may face a high hurdle. The Ninth Circuit’s reasoning provides a clear line for focusing DMCA arguments on the presence or absence of CMI in the exact copies, not on the broader question of whether an AI produced a similar result. (cases.justia.com)
Discovery strategy in AI disputes: Given the emphasis on whether CMI was removed or altered from an existing copy, counsel should prioritize the documentary trail around CMI in the source materials and the nature of the outputs. In cases where AI tools are involved, it may be more productive to request evidence related to the presence or absence of CMI in the outputs themselves, rather than broad, open-ended prompts, unless an independent theory supports liability. The court’s ruling helps counsel frame discovery around CMI provenance and the exact copy status, which can streamline pretrial procedures and motion practice. (cases.justia.com)
Preserving and testing an “input theory”: The Ninth Circuit declined to reach the “input theory” because the issue was forfeited. This underscores the importance of preserving any alternative theories early in litigation if the theory could survive posturing in a later stage. Practitioners should consider whether an input-based theory could support a DMCA claim and, if so, preserve that issue for potential appeal or remand. This has practical implications for how early discovery requests are drafted and how the record is developed in AI disputes. (cases.justia.com)
Contract claims remain relevant: While DMCA claims were affirmatively resolved against the plaintiffs on the stated theory, the Ninth Circuit left open the possibility that contract claims could proceed in the lower court. This means that trial teams should not treat AI driven disputes as purely statutory fights; contract and licensing disputes around attribution, licensing terms, and usage rights may still be pivotal. Coordinate DMCA defenses with robust contract analysis to avoid letting a single statutory victory overshadow broader commercial disputes. (cases.justia.com)
Implications for trial-readiness and training: For litigation teams and trial readiness practices, this decision highlights the value of targeted training on AI prompts, outputs, and the legal frameworks surrounding CMI and 1202(b). Objection Academy provides scenario-based practice around AI prompts and objections, enabling counsel to sharpen cross examination related to AI outputs, CMI, and licensing terms. Integrating AI focused modules into trial prep helps teams anticipate DMCA and contract issues that arise when technology mediates evidence, discovery, or expert analysis. This is particularly relevant in complex IP and tech disputes where AI tools figure prominently in both evidence generation and expert testimony. (cases.justia.com)
Concrete steps for current cases
Map the DMCA theory to the facts: In any AI-driven dispute, assess whether the alleged DMCA violation involved removal or alteration of CMI from an existing copy. If the outputs are new works without CMI, the DMCA 1202(b) claim may be untenable under the Ninth Circuit’s framework.
Consider preserving an alternative theory: If a client might have an input-based or licensing-based theory against a defendant, preserve those arguments for potential appeal or remand, as the appellate court did not address an alternate theory in this opinion.
Plan for contract claims in parallel: Because contract claims may proceed, align discovery and trial readiness to address licensing terms, attribution obligations, and data usage rights alongside DMCA defenses.
Use AI-focused training for trial teams: Leverage practice resources to improve the handling of AI prompts, outputs, and related evidentiary issues. Objection Academy’s materials can help litigators rehearse objections and cross examinations around AI generated evidence and CMI issues, enhancing courtroom readiness.
Monitor related developments: This area remains rapidly evolving. While the Ninth Circuit’s decision provides a clear rule for DMCA 1202(b) in this context, other circuits and district courts may adopt different approaches to “input theory” or to the admissibility of AI generated prompts and outputs. Keeping an eye on parallel developments is essential for trial strategy.
Sources
DOE v. GitHub, Inc., 24-7700 (9th Cir. Sept. 16, 2026). Full opinion and summary available via Justia, including a downloadable PDF of the Ninth Circuit’s opinion. (cases.justia.com)
Summary and case details: DOE v. GitHub, Inc., 24-7700 (9th Cir. 2026). Justia Opinion Summary. (law.justia.com)
Ninth Circuit opinion (PDF): DOE v. GitHub, Inc., 24-7700, Sept. 16, 2026. Downloadable from Justia. (cases.justia.com)