Abstract
The incorporation of artificial intelligence (AI) into public security, criminal investigation and forensic science is changing the way large volumes of data are collected, organized, correlated and interpreted. This study presents a literature review on the main applications of AI in criminal investigation, focusing on facial recognition, image and video analysis, human identification, forensic genetics, toxicology, digital forensics, crime pattern analysis and decision support. The review was based on indexed scientific literature and systematic and scoping reviews published in specialized journals, prioritizing recent studies and reference works on artificial intelligence, machine learning and forensic sciences. The findings indicate that AI can increase the speed of evidence screening, assist in pattern identification and support experts in repetitive or highly complex tasks. However, the literature also highlights limitations related to the quality and representativeness of databases, external validation, explainability, algorithmic bias, cybersecurity, data protection and the need for human oversight. In criminal investigation, therefore, AI should be understood as a support tool for investigative and forensic activities rather than a substitute for technical, scientific or legal judgment. Responsible adoption requires validation protocols, traceability, auditing, transparency, professional training and interdisciplinary integration among investigators, forensic experts, legal professionals and data scientists.