Summary
Research indicates that providing AI models with incomplete legal context—such as a statute without its controlling case law—can actually degrade reasoning performance through a phenomenon called "Statutory Myopia." This occurs because the explicit text in the context window tends to overwrite the model's pre-trained knowledge of higher-level constitutional precedents. To ensure reliability, the author argues that legal AI must move toward hierarchical retrieval systems that link statutes directly to the authoritative precedents that govern them.