If 2023 marked the transformative influence of artificial intelligence (AI), 2024 is poised to be the year that U.S. copyright law undergoes a profound shift, particularly in response to the surge in generative AI and the widespread adoption of products from industry giants like Microsoft-backed OpenAI, Meta Platforms, Midjourney, and others.
The Copyright Conundrum
The proliferation of AI has triggered a wave of copyright cases, with writers, artists, and other copyright holders alleging that AI’s success is built upon their creative contributions. While judges have exhibited skepticism towards infringement claims based on AI-generated content, a more intricate and potentially multi-billion-dollar question remains unaddressed — whether AI companies violate copyrights on a massive scale by training their systems with vast amounts of data scraped from the internet.
Lawsuits Galore
In a series of proposed class-action lawsuits, various groups of authors, including luminaries such as John Grisham, George R.R. Martin, comedian Sarah Silverman, and former Arkansas governor Mike Huckabee, have challenged tech companies over the use of their texts in AI training. Copyright holders have initiated similar legal actions in visual arts, music publishing, stock-photo provider Getty Images, and even the venerable New York Times.
Want to know if you’re earning what you deserve? Find out with LawCrossing’s salary surveys.
Allegations and Demands
The crux of these lawsuits revolves around the contention that tech companies infringe copyrights by utilizing and reproducing materials without permission for AI training. Plaintiffs seek monetary damages and court orders preventing the misuse of their intellectual property.
The Defense Argument
Aware of the potential ramifications, tech companies have assembled formidable legal teams to counter these claims. Their defense hinges on the assertion that AI training is comparable to how humans learn new concepts, qualifying their use of materials as “fair use” under copyright law. They argue that, like a child learning language through everyday speech, AI models ‘learn’ by being exposed to massive amounts of text during training.
The Stakes for the AI Industry
AI proponents emphasize the potentially disastrous consequences for the industry should these legal battles go against them. Silicon Valley venture capital firm Andreessen Horowitz warns that imposing copyright liability on AI creators could either stifle or significantly impede the development of AI models.
The Copyright Owners’ Perspective
On the other side of the debate, copyright owners point to the immense success of AI programs like OpenAI’s ChatGPT and argue that these tech companies have the financial means to compensate for the use of copyrighted materials. Writers’ trade group, The Authors Guild, contends that licensing copyrighted materials for AI training should be costly, considering the substantial value derived from professionally created texts in developing large language models.
The Thomson Reuters Lawsuit
A pivotal case involving Thomson Reuters, the parent company of Reuters News, serves as a potential bellwether for AI copyright issues. Accusing Ross Intelligence of illegally copying thousands of legal “headnotes” from Thomson Reuters’ Westlaw platform, the ongoing lawsuit could set a crucial precedent for fair use and other aspects of AI copyright litigation.
Don’t be a silent ninja! Let us know your thoughts in the comment section below.