Today has turned out to be Rule on Motions to Dismiss Day. On the heels of Judge Lee’s dismissal of some claims of Concord Music against Anthropic earlier this morning, Judge Stein just issued an order dismissing some, but not all of the claims OpenAI and Microsoft sought to be dismissed. (Other than the assertion of a statute of limitations defense, they did not seek dismissal of the direct infringement claims.)
Claims Dismissed With Prejudice
- The common law unfair competition by misappropriation claims and OpenAI’s motion to dismiss the “abridgment” claims in the CIR action
Claims Dismissed Without Prejudice (allowing amending the complaints)
- DMCA section 1202(b)(1) claims against it in all three actions
- DMCA section 1202(b)(1) claim against OpenAI in The New York Times Company v. Microsoft Corporation, et al., No. 23-cv-11195 (“the Times action”), and
- DMCA section 1202(b)(3) claims against them in all three actions
Claims Not Dismissed
- Direct infringement claims involving conduct occurring more than three years before the complaints were filed against OpenAI;
- Contributory copyright infringement claims;
- State and federal trademark dilution claims in the Daily News action, and
- DMCA 1202(b)(1) claims against OpenAI in the Daily News and CIR actions
For the 1202(b)(1) claim, the Center for Investigative Reporting alleged: “Based on the information above, including the OpenAI Defendants’ admission to using the Dragnet and Newspaper extraction methods, which remove author, title, copyright notice, and terms of use information from copyright-protected news articles published online, the OpenAI Defendants intentionally removed author, title, copyright notice, and terms of use information from Plaintiff’s copyrighted works in creating ChatGPT training sets.” Paragraph 103, First Amended Complaint.
Similarly, the Daily News alleged (at paragraphs 159-63): “For example, in order to construct the datasets used to train their GenAI products, the Defendants used content extractors that, by design, removed the Publishers’ CMI from the Publishers’ Works. For example, OpenAI used the Dragnet67 and Newspaper68 content extractors in creating the WebText dataset, which intentionally removed the Publishers’ CMI from the Publishers’ Works scraped from their website. The abstract of the paper describing the Dragnet content extractor describes that copyright notices are removed as part of the process of extracting the text content of a website: “The goal of content extraction or boilerplate detection is to separate the main content from navigation chrome, advertising blocks, copyright notices and the like in web pages.”
The New York Times’ 1202(b)(1) claim was dismissed apparently because it didn’t have this level of specificity in the removal of CMI. Presumably, that may be fixable in an amended complaint.
Download Judge Stein’s order: