From Probable Cause to Predictive Cause: AI, Policing, and the Erosion of Reasonable Suspicion

Abstract

The progressive entanglement of Artificial Intelligence (AI) with contemporary policing represents one of the most profound transformations in criminal procedure since the institutionalization of probable cause. Predictive policing systems—trained on large historical datasets—now operate as algorithmic arbiters of suspicion, displacing human judgment with computational projections of criminality. This paper interrogates how AI-driven systems redefine the constitutional standards of reasonable suspicion and probable cause under U.S. law, while simultaneously unsettling European conceptions of proportionality and necessity under the Charter of Fundamental Rights of the European Union. Drawing from interdisciplinary research in law, computer science, and philosophy, it analyzes the emergence of “predictive cause” as a new epistemic standard, rooted in algorithmic immutability and opaque reasoning. The article exposes how immutable algorithmic classifications perpetuate systemic discrimination, undermining the procedural fairness and transparency central to due process. Comparative inquiry reveals that while the U.S. leans toward operational efficiency, the E.U. foregrounds human dignity and data protection, yet both face difficulties ensuring accountability. The study concludes with doctrinal and policy recommendations aimed at reconciling algorithmic accuracy with constitutional integrity, advancing a framework for algorithmic accountability that restores the balance between technological progress and civil rights.