In Schulte v. LinkedIn Corp., the U.S. District Court for the Northern District of California addressed several discovery disputes involving LinkedIn’s use of Relativity aiR, a generative AI tool used to assist with document review. The court rejected plaintiffs’ request to prohibit LinkedIn from using search strings to pre-cull documents before they were reviewed by Relativity aiR, finding that the practice was reasonable and proportional. The court noted that plaintiffs had not shown LinkedIn’s 25 search strings were too narrow or otherwise deficient, while applying the AI review process to all custodial files would impose significant processing, hosting, and review costs.

The court also declined to require LinkedIn to disclose additional metrics about its use of Relativity aiR, including elusion estimates, document error rates, and the number of human reviewers validating the tool’s predictions. LinkedIn had disclosed its use of the technology and provided additional information upon request, which the court found satisfied the parties’ Interim ESI Order. Without a specific showing that LinkedIn’s production was deficient, the court found that further “discovery on discovery” was not warranted.

The court similarly denied plaintiffs’ requests to add an in-house attorney as a custodian and to require LinkedIn to preserve and produce text messages from 19 custodians. For the attorney custodian, plaintiffs failed to establish that her files contained unique, non-privileged, non-duplicative information not already captured by the existing custodial framework. For text messages, the court found that plaintiffs had not shown good cause to override the Interim ESI Order’s exclusion of text messages from default preservation obligations, and that the requested collection would impose substantial burdens across personal and work devices.

For eDiscovery teams, Schulte reinforces the importance of proportionality when using AI-assisted review, developing defensible ESI workflows, and carefully, intentionally drafting ESI protocols. The decision recognizes that parties may use search terms to narrow large document populations before applying GenAI review tools, while also emphasizing that challenges to AI-assisted discovery should be supported by specific evidence of deficiencies rather than speculation. The ruling also highlights the need to carefully evaluate whether additional custodians or data sources are likely to provide unique, non-duplicative information before expanding discovery.

If your organization is seeking support with eDiscovery, our team has solutions to address all phases of the discovery process. At CODISCOVR, we deliver client-focused, defensible, and scalable solutions using advanced technology and intelligent review practices to meet eDiscovery, document review, and information governance needs in a manner that reduces the risks and costs associated with electronically stored information (ESI). Reach out to Nicole Gill, Chair, Managing Member CODISCOVR. With almost a decade of experience, she manages complex and high-profile eDiscovery projects and routinely navigates data and privacy protection laws across many domestic and foreign jurisdictions.