Litigation partners Dan Toal and Geoff Chepiga, and deputy chair and counsel, e-discovery, Ross Gotler’s latest federal e-discovery column, “Court Evaluates Generative AI Review Under Established Discovery Principles,” appeared in the August 4 issue of the New York Law Journal. The authors discuss a recent decision from the U.S. District Court for the Northern District of California in Schulte v. LinkedIn Corp., in which the court declined to apply heightened scrutiny to LinkedIn’s use of a generative AI document review tool during discovery, instead evaluating the process under established standards of relevance and proportionality. The dispute centered on LinkedIn’s use of Relativity aiR to review documents for responsiveness after first narrowing its data set through agreed-upon search terms. The plaintiffs, a putative class of LinkedIn Premium subscribers, sought to bar the pre-review culling with search strings and to compel disclosure of additional metrics regarding the AI review process. The court rejected these requests, emphasizing that responding parties generally retain discretion to design reasonable discovery methodologies and that “discovery on discovery”—an attempt to probe a party’s discovery process—requires concrete proof of a specific deficiency. The decision signals that a well-designed and validated generative AI review process stands on the same settled footing as the review methods that came before it.
E-discovery attorney Lidia Kekis assisted in the preparation of this article.
» read the article
Related Insights
