Georgetown Law
Technology Review

The intersection of artificial intelligence (AI) and legal analysis can revolutionize how we address foundational questions in legal philosophy. This Technology Explainer discusses how artificial intelligence might address the biggest problems in legal analysis on an unprecedented scale. Experimental jurisprudence is an emerging field that combines legal theory and empirical methods to make progress on fundamental questions of legal philosophy.1See Kevin Tobia, Experimental Jurisprudence, 89 U. Chi. L. Rev. 735 (2022). Drawing from cognitive and behavioral science2See Roseanna Sommers, Experimental Jurisprudence, 373 Sci. 394 (2021). among other disciplines,3Tobia, supra note 1, at 779 (explaining how experimental jurisprudence might apply corpus linguistics, the processing of large bodies of text. Surveying laypeople is the most common method in experimental jurisprudence, but the responses collected can also be used as a “corpus” for linguistic analysis in addition to the common statistical analyses). it focuses on the interpretation of data to make progress on legal philosophy.4See id. (explaining that while experimental jurisprudence often relies on the creation of data, a rich body of research in the field has turned to the processing and interpretation of existing data).

Machine learning (ML), an application of artificial intelligence (AI) capabilities to process large amounts of data, has arisen as a potential tool to help. Applied to experimental jurisprudence, ML treats judicial opinions as data5Big data is the collection of increasingly massive quantities of information, often enumerated in file size rather than number. Francis X. Diebold, On the Origin(s) and Development of “Big Data”: The Phenomenon, the Term, and the Discipline 7 (PIER Working Paper No. 12-037, 2012) (“[I]n a landscape littered with failed attempts at interdisciplinary collaboration, Big Data is emerging as a major interdisciplinary triumph”). to identify trends, such as in the use of interpretive tools. For example, ML can be used to understand the rise of textualism, an interpretive tool that prioritizes the semantic language of a statute, ostensibly by considering how a reasonable person would understand it, and how it is applied.6Anita S. Krishnakumar, Statutory Interpretation in the Robert’s Court’s First Era: An Empirical and Doctrinal Analysis, 62 Hastings L.J. 221, 223 n.2 (2010) (“Textualism is an interpretive philosophy that prioritizes the statute’s text above all else.”); John F. Manning, What Divides Textualists from Purposivists?, 106 Colum. L. Rev. 70, 76 (2006). By the same token, ML can also be used to analyze legal texts for things other than interpretive tools. For example, intellectual property scholars might be interested in how various jurisdictions analyze the likelihood of confusion test in trademark cases. ML can then be used to discover patterns over the weight judges accord to each element.7See, e.g., Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492 (2d Cir. 1961).

Continue Reading