By Gerald L. Maatman, Jr. and Jennifer A. Riley
Duane Morris Takeaway: Generative AI hit mainstream in 2023 and quickly become one of the most talked-about and debated subjects among corporate legal counsel across the country, as numerous companies jumped to incorporate AI while attempting to manage its risks. In 2023, we saw the tip of the iceberg relative to the ways that generative AI is poised to transform class action litigation.
In the video below, Duane Morris partner Jennifer Riley discusses the latest AI class action rulings, and what companies can expect to see in AI litigation in 2024.
DMCAR Trend #8 – Generative AI Began Transforming Class Action Litigation
- Opportunities For Enhanced Efficiency
As the COVID 19 pandemic brought video-conferencing tools into the mainstream, such tools enabled more litigants to conduct and to attend more hearings, more depositions, and more mediations in less time. While the debate continues as to their effectiveness, generative AI is poised to enable lawyers to far surpass those gains in efficiency, potentially enabling the plaintiffs’ class action bar to do “more with less” like never before, leading to more lawsuits that can be handled by fewer lawyers in less time and a potential surge of class actions on the horizon.
Less than a year into the generative AI movement, we have seen the technology influence various aspects of the legal process, including by assisting legal professionals in analyzing vast amounts of data; automating the review of documents, contracts, and communications; increasing the speed and potentially enhancing the accuracy of e-discovery; and automating and enhancing the dissemination of information in the class action settlement administration process.
Legal research, for example, traditionally required a time-consuming undertaking that involved sifting through dozens of decisions and secondary authorities. AI tools are enhancing this process through natural language search capabilities and machine learning algorithms that streamline the process and enhance the results. Document review similarly traditionally required a time-consuming and painstaking process. AI tools are using machine learning and text analytics, for example, to sort and categorize large datasets with increasing accuracy. By quickly analyzing extensive document sets, AI tools can expedite the discovery process, making litigation more efficient and cost-effective.
Likewise, AI has the potential to revolutionize the process of administering class action settlements. The participation in claims-made settlements, for instance, often falls within the range of 15% to 35%, depending upon various factors such as the type and method of notice. AI can be used in a variety of ways, including to find potential class members and thereby raise claim rates, while reducing administrative costs, increasing the amount available for distribution as well as the ultimate settlement payout.
In sum, the legal industry is poised to leverage this transformative technology to make leaps in enhancing the efficiency and effectiveness of the class action litigation process.
- Risk Of Class Claims
While improving the efficiency with which the plaintiffs’ class action bar can litigate class actions, generative AI is providing an ocean of raw material for potential claims. Upon hitting the mainstream, AI promptly became the subject of class claims, which span multiple theories and areas of law.
While generative AI might improve the speed of interactions, for instance, users have the ability to exploit AI to generate massive amounts of false information or to simply inadvertently rely upon errors in AI-generated communications, giving rise to claims. Similarly, the SEC has warned businesses against “AI washing,” or making false claims regarding their AI capabilities, likening it to the greenwashing phenomenon that has been the target of an agency crackdown. The plaintiffs’ class action bar is using such representations about AI to fuel class claims for consumer fraud based on allegedly misleading or deceptive representations about the efficacy of AI technology. In Matsko, et al. v. Tesla, Case No. 22-CV-5240 (N.D. Cal. Sept. 14, 2022), for instance, a plaintiff filed a class action alleging that Tesla exaggerated the capabilities of its software and asserting various causes of action for breach of warranty and violation of California consumer protection laws, among others.
Companies that incorporate AI to streamline their decision-making processes likewise face the prospect of class action suits. Plaintiffs have filed suits against insurers that used algorithms to adjudicate claims, for example, as well as against agencies that used programs to deny or reduce government benefits. In Kisting-Leung, et al. v. Cigna Corp., Case No. 23-CV-01477 (E.D. Cal. 2023), for instance, a group of California consumers filed a class action complaint against a national health insurance company alleging that its use of an algorithm to deny certain medical claims constituted breach of the implied covenant of good faith and fair dealing, unjust enrichment, intentionally interfered with contractual relations, and violated California’s Unfair Competition Law.
The developers of generative AI products have not remained immune. Such companies have faced a slew of class action lawsuits alleging privacy violations. In a series of lawsuits beginning in June and July 2023, the plaintiffs’ class action bar has alleged that, by collecting publicly-available data to develop and train their software, developers of generative AI products stole private and personal information from millions of individuals. In P.M., et al. v. OpenAI LP, No. 3:2023-CV-03199 (N.D. Cal. 2023), a group of plaintiffs filed a class action suit against OpenAI LP and Microsoft, Inc. alleging that by collecting publicly-available information from the internet to develop and train its generative AI tools, including ChatGPT, Dall-E, and Vall-E, OpenAI stole private information from millions of people, violating their privacy and property rights, among other claims. In J.L., et al. v. Alphabet Inc., No. 3:23-CV-03440 (N.D. Cal. 2023), the same plaintiffs’ firm filed a class action lawsuit against Google, similarly alleging that, by collecting internet data to train its tools like Bard, Imagen and Gemini, Google infringed privacy rights and violated the Copyright Act.
Developers of generative AI tools similarly have faced claims. Plaintiffs have filed class action lawsuits claiming that, by collecting and using internet data to train generative AI models, developers violated copyright laws. In Andersen, et al. v. Stability AI, Ltd., Case No. 23-CV-00201 (N.D. Cal. Oct. 20, 2023), for example, plaintiffs filed a class action on behalf of artists alleging that Stability AI, Ltd. and Stability AI, Inc. “scraped” billions of copyrighted images from online sources, without permission, to train their models to generate new images without ascribing credit to the original artists. In Doe v. GitHub, Inc., 22-CV-06823 (N.D. Cal. May 11, 2023), the plaintiffs, a group of developers who allegedly published licensed code on GitHub’s website, filed a class action lawsuit against GitHub, the online code repository, as well as Microsoft and OpenAI claiming that GitHub improperly used that code to train its AI-powered coding assistant, Copilot, without appropriate attribution in violation of copyright management laws.
As technology continues to grow and change, and the plaintiffs’ class action bar continues to flex its creativity, the number and types of claims are likely to expand and evolve during the upcoming year.