Extract from Robert Keening, Ray Mangum, Eli Nelson, & Kevin Reiss’s Article, “Better Than TAR. Nearly Expert: What a Major Study Shows About Gen AI and TAR in a Complex Document Review.”
The question for generative AI-based review tools is not whether they can equal TAR, but whether they can outperform it.
For more than a decade, technology-assisted review (TAR) has helped legal teams conduct large-scale document review more efficiently and defensibly than traditional search-term or linear-review approaches.
Courts, practitioners, and discovery professionals have developed familiar ways to validate results, measure recall and elusion, and explain why a review process was reasonable and proportional. The question for generative AI-based review tools is not whether they can equal TAR, but whether they can outperform it.
TAR has repeatedly shown that it can perform well in finding documents related to specific topics. The harder question for generative AI (gen AI) is whether it can go further: whether it can understand the contents of a document, apply a complex review protocol to its contents, and determine whether the document is responsive.
