The Triple Path of AI-Empowered Criminal Investigation Curriculum Construction

Abstract

Artificial intelligence (AI) is changing criminal investigation from a predominantly experience-led practice into a data-intensive activity in which forecasts, similarity scores, rankings, and generated content increasingly shape investigative attention. Yet police and criminal-investigation curricula often treat AI as a collection of tools rather than as a sociotechnical system whose outputs require statistical interpretation, legal justification, and accountable human judgment. This conceptual paper develops a Triple-Path framework for curriculum construction. Path I, Technological Reconstruction, builds investigative AI literacy through probabilistic reasoning, model evaluation, data provenance, and explainability. Path II, Ethical and Legal Governance, integrates fairness, privacy, due process, contestability, evidentiary reliability, and algorithmic impact assessment. Path III, Practice-Oriented Human-AI Collaboration, uses simulations and laboratory work to calibrate trust, counter automation bias, and preserve meaningful human control. The framework converts broad calls for responsible AI into teachable competencies, learning activities, and assessment evidence. It also emphasizes that investigators should neither reject AI categorically nor defer to it uncritically; they should be able to explain what a system does, identify when it may fail, document how it influenced a decision, and justify the final investigative action under law and professional ethics. The proposed framework offers a practical foundation for modernizing criminal-investigation education while protecting fairness, accountability, and public trust.