Non-Compositional Scope of Component Permissions: Why Approving Each Step Does Not Approve What the Steps Produce Together

This conceptual paper introduces Non-Compositional Scope of Component Permissions, a structural constraint on what separate permissions can establish. Individually permitted operations may combine into a use or disclosure whose authorization is not settled by approval of the components alone. The paper assesses a combined operation by the actual scope of authorization, including explicit terms and clear implications. An unexpected result is not necessarily unauthorized, and existing permissions may cover combinations without fresh approval. Whether a composition is authorized and whether it causes harm are separate questions. The discussion connects the constraint to Solove’s account of aggregation and examines U.S. Department of Justice v. Reporters Committee for Freedom of the Press (1989). The Court distinguished disclosure of an FBI compilation from access to scattered public records. The paper uses this statutory disclosure case as a structural comparison, keeping public accessibility distinct from permissions granted by individuals. Part of the Power, Permission, and Obligation series within On Human Modeling, the paper examines who defines permission scope, who can authorize a composition, and who bears its effects. Its framework can inform human–AI research on whether permissions for individual tools, data sources, or actions cover the uses and disclosures they produce together.