CIR v. OpenAI and Microsoft
Journalism · AI · Intellectual property · Litigation · USA
Artificial intelligence has advanced significantly with the development of generative AI technologies, such as GPT-3 and GPT-4. These innovations have led to an increasing number of intellectual property (IP) litigation cases, which are crucial in influencing the future of AI development and IP law.
The DMCA
The US Digital Millennium Copyright Act (DMCA) of 1998 (Pub. L. No. 105-304, 112 Stat. 2860 (Oct. 28, 1998)) implements the WIPO Copyright Treaty and the WIPO Performances and Phonograms Treaty. The DMCA criminalizes the creation and distribution of technology, devices, or services designed to bypass measures controlling access to copyrighted works, also known as digital rights management (DRM). Additionally, it criminalizes the act of circumventing access controls, regardless of whether copyright infringement occurs. The DMCA also increases penalties for copyright infringement conducted over the Internet. It forbids removing author, title, copyright, and terms of use information from protected works when it is known that doing so could lead to, enable, facilitate, or hide copyright infringement. Unlike copyright infringement claims, which necessitate significant and often prohibitive registration costs for copyright owners to enforce their rights, a DMCA claim does not require such registration.
The complaint
On 27 June 2024, the CIR filed a complaint against OpenAI and Microsoft in the US District Court for the Southern District of New York.
The Center for Investigative Reporting, Inc. (CIR) v OpenAI, Inc., OpenAI GP, LLC, OpenAI, LLC, OpenAI Opco LLC, OpenAI Global LLC, OAI Corporation, LLC, OpenAI Holdings, LLC (OpenAI) and Microsoft Corporation (Microsoft), 1:24-cv-04872
The CIR alleges that the defendants are “responsible for the creation and development of the highly lucrative ChatGPT and Copilot artificial intelligence (AI) products, which are built on uncompensated and unauthorized use of the creative works of humans.” The CIR complained that the systems and the large language models (LLMs) that power them have been trained via regurgitations of their copyrighted works, jeopardising the CIR’s continued ability to fund its nonprofit public interest journalism. Further, the defendants allegedly removed the CIR’s author, title, copyright notice, and terms of use information. The CIR’s claims concern direct and contributory copyright infringement, unlawful public display, and abridgements of copies (or derivatives) of its works.
The CIR seeks actual damages, including actual and statutory damages, of at least US$750 per infringed work, and US$2,500 per DMCA violation. The CIR also requested injunctive relief and legal fees.