Judges and legislators now openly acknowledge using artificial intelligence (AI) to draft opinions and shape statutes. This Essay names this emerging phenomenon Artificial Legal Intelligence (ALI)—AI-created law—and argues that ALI presents a “hard” problem: whether law produced or substantively shaped by artificial systems can maintain the normative legitimacy that attaches to human-generated law. In contrast to the “easy” problems of AI in law (e.g., transparency, accuracy, accountability) whose solutions, however difficult, are conceptually imaginable, the hard problem probes a deeper incompatibility. Even if AI replicates human legal reasoning with perfect fidelity, should its outputs still be regarded as law in the full and proper sense? This Essay groups and examines the principal arguments supporting the proposition that there is a hard problem, including those emphasizing law’s essential deliberative character, its reliance on first-person moral judgment, and empirical evidence that law’s perceived authority diminishes when authored by machines. The Essay concludes by proposing that ALI be recognized as a distinct field of legal scholarship with its own research agenda.
LawReview
