Legal Tech • February 2024

Systemic Theft on a Massive Scale

AI continues to spark new legal controversies.

Written by Mandi Phillips

The Development of artificial intelligence chatbots has ushered in a tidal wave of legal woes (and concerns) ranging from plagiarism in college classrooms to false legal citations and other issues arising out of the use of AI in briefings submitted to courts. Now, those legal woes include numerous lawsuits filed in relatively quick succession—at least three in California and one in New York—a number of them involving high-profile writers and authors.

CURRENT AI LAWSUITS
In the first suit, filed in June 2023, writers Paul Tremblay and Mona Awad brought claims against OpenAI in San Francisco federal court maintaining that OpenAI mined data copied from their books without permission, infringing the authors’ copyrights. In response, OpenAI claims “fair use” of the copyrighted content.

In July 2023, again in California, comedian Sarah Silverman and two writers, Christopher Golden and Richard Kadrey, brought a proposed class-action lawsuit against Meta Platforms and OpenAI, alleging both defendants used the authors’ copyrighted content (without permission) to train their chatbots (Meta AI and ChatGPT, respectively) to respond to human prompts.

Two more lawsuits followed soon after in September 2023. Writer Michael Chabon, author of The Amazing Adventures of Kavalier & Clay, joined forces with fellow writers David Henry Hwang, Rachel Louise Snyder, and Ayelet Waldman asserting similar claims against OpenAI, once again in California federal court. And in a separate suit filed alongside the Authors Guild in the Southern District of New York, high-profile authors John Grisham, George R.R. Martin, Jodi Picoult, Jonathan Franzen, Elin Hilderbrand, and 12 other writers brought similar claims against OpenAI. According to their complaint in Authors Guild v. OpenAI, having used the authors’ copyrighted content to train ChatGPT, OpenAI’s chatbot can now produce “derivative works” that can copy and summarize the authors’ books without permission, potentially harming the market for the authors’ work. As noted by these authors, “[a]t the heart of [OpenAI’s ChatGPT] algorithms is systemic theft on a massive scale.”

AI LAWSUITS FOCUS ON TRAINING OF BOTS
AI programs (or chatbots) such as ChatGPT respond to user prompts by application of an algorithm that selects words based on “training” the chatbots received from scanning billions of pieces of text across the internet. The primary argument in each of these lawsuits involved the alleged illegal use of copyrighted content to train the AI programs, an act each set of plaintiffs maintains is an unauthorized reproduction of the copyrighted works in violation of the Copyright Act of 1976 (17 U.S.C. § 501, et seq.). Thus, the plaintiffs allege, OpenAI and others creating and “training” chatbots scan and make use of the copyrighted material, which creates or fosters written work that publishers would otherwise pay authors to generate.

In response, OpenAI, Meta Platforms, and others maintain their use does not violate the provisions of the Copyright Act but constitutes “fair use” in an effort to create wholly separate original content.

It’s clear that OpenAI and others continue to stack legal woes on top of legal woes as the use of AI grows. What remains unclear is exactly what types of information AI companies are using— and how—in the “training” of chatbots, and whether the protections afforded under the Copyright Act prohibit such use. How the courts react to these claims could significantly alter the  industry—and the law. Decisions in favor of the authors could ultimately require OpenAI and others to obtain prior permission and paid licensing before using copyrighted content. On the other hand, decisions in favor of the AI companies could open all online content to AI developers worldwide, without consequence. Either way, these cases are positioned to alter the legal landscape and the application of existing copyright law to recent technological advances.

This article was originally published on the Klemchuk Ideate blog and has been edited and reprinted with permission.

MANDI PHILLIPS is a senior attorney with Klemchuk, where her practice focuses on commercial and intellectual property litigation and enforcement and trademark prosecution.