Traditional methods of crime prevention—the punishment of the culpable and the preventive restraint of the dangerous—are slowly being supplemented and supplanted by technologies that seek to perfectly prevent crime. For instance, the federal government is developing in-car technology that would prevent vehicle operation when a driver has a blood alcohol level in excess of the legal limit. Less directly, the anti-circumvention provisions of the Digital Millennium Copyright Act of 1998 try to prevent copyright infringement by eliminating technologies that enable such infringement. Such structural regulation of private conduct is not new, but few scholars have focused on its use to prevent crime, and fewer still have examined how structural methods to fight crime fit within legal theory. This Article begins that discussion with three aims. First, I argue that perfect prevention—the use of technology by the State to make criminal conduct practically impossible—is a novel approach to crime prevention that requires separate scrutiny from punishment and prevention. Second, I identify concerns with the use of perfect prevention and propose limitations on the perfect preventive state that are responsive to those concerns. Specifically, I address the impact of perfect prevention on individual autonomy, concerns raised by the blanket application of perfect prevention on all people, and the question of whether and when perfect prevention should be the preferred approach for preventing certain criminal conduct. Third, I highlight areas for future discussion of perfect prevention by scholars.