Contemporary regulation of artificial intelligence is dominated by a powerful intuition: that the risks posed by opaque machine learning systems can be mitigated through transparency. This Article challenges that intuition. Drawing on both legal and computer science literature on explainable AI, it argues that transparency is not only technologically unrealistic for most modern AI systems but also conceptually mismatched with the kinds of accountability that law actually demands. What law requires is not wholesale visibility into complex models, but targeted, causal explanations that connect specific processes to specific outcomes in legally salient ways. The Article begins by disaggregating the "black box" problem of AI into distinct forms of opacity-interactive, semantic, and procedural-and demonstrates that only a narrow subset of these implicates law: when potentially harmful processes are causally linked to harmful outcomes, a problem that law has confronted for centuries. Against this backdrop, the Article offers a sustained critique of the transparency-centered framework underlying much AI regulation, including the European Union's Artificial Intelligence Act and many state laws, including several in California alone. It shows that extensive documentation and disclosure mandates reflect an undifferentiated conception of transparency that will fail to provide either meaningful explanations or effective accountability. By contrast, the Article develops an explanation-centered regulatory model grounded in three features common to both legal reasoning and human explanatory practice: selectivity, comparison, and interactivity. Finally, the Article bridges legal theory and computer science by examining the capabilities and limits of contemporary explainable AI techniques and suggesting specific ways that computer science and law can inform each other for more effective AI regulation. The Article concludes that effective AI regulation must abandon transparency as a governing principle and instead embrace explanation as a regulatory paradigm.