Rethinking criminal profiling through cognitive artificial intelligence.

Area of Science:

• Forensic Science
• Computer Science
• Criminology

Background:

• Traditional criminal investigations face challenges in integrating dispersed and heterogeneous information, hindering the detection of serial or escalating patterns.
• Advances in artificial intelligence (AI) and cognitive computing offer data-driven solutions for cross-source correlation and temporal anomaly detection in investigations.

Purpose of the Study:

• To conduct a narrative review of AI applications in forensic analysis, pattern detection, and investigative support.
• To synthesize findings on AI's role in evidence correlation and anomaly detection.
• To examine AI's utility through retrospective case studies.

Main Methods:

• A focused narrative review of peer-reviewed literature on AI in forensic analysis.
• Synthesis of selected studies into evidence correlation and anomaly detection dimensions.
• Retrospective case-based illustration of AI applications.

Main Results:

• AI approaches link low-level traces to high-level events, enabling structured reconstruction and pattern identification.
• Machine learning models achieve high predictive performance (often >90% accuracy) by integrating heterogeneous data.
• AI methods detect temporal anomalies and cross-source patterns often missed in manual analysis, revealing longitudinal regularities in historical cases.

Conclusions:

• AI significantly enhances investigative methodologies by enabling continuous signal integration and data-informed profiling.
• A gap exists between AI's performance in controlled environments and its validation in real-world operational contexts.
• Future research should focus on empirical benchmarking, explainable AI, and robust governance frameworks for transparent and accountable deployment.