AI as a criminogenic factor: issues of qualification and investigation tactics

The article addresses the problem of the lack of a comprehensive, proactive legal response based on a deep understanding of technological risks, enabling law enforcement agencies to maintain effectiveness in the new digital reality and ensure national security protection. The subject of the research is the norms of domestic legislation regulating relations arising from the use of artificial intelligence (AI). The aim of the study is to identify criminal law and forensic gaps in combating crimes related to AI technologies and to develop specific measures to address them. Based on the analysis of departmental statistics from the Ministry of Internal Affairs of Russia, the Investigative Committee of the Russian Federation, and judicial practice, the growing gap between the technological capabilities of criminals and the adequacy of existing legal mechanisms is demonstrated. As a result of the systematization of threats, an author's classification of AI systems has been proposed based on the degree of autonomy, application area, and level of risk, serving as a basis for a differentiated legal approach. The methodology includes: 1) formal-legal and comparative legal analysis; 2) sociological-legal methods; 3) legal technique methods (construction of hypotheses, dispositions, sanctions); 4) forensic methods for modeling crime commission methods and investigation tactics. A set of general scientific and specialized scientific methods has been used: dialectical, formal-legal, comparative-legal, system-structural, and predictive methods. The scientific novelty lies in the development of a criminally relevant definition of AI, the identification of specific qualification issues (attribution, subject of responsibility, causal connection), and the justification of a set of tactical and forensic recommendations and legislative innovations aimed at forming adequate tools for investigative bodies. Key conclusions: 1) the current norms of the Criminal Code of the Russian Federation (Chapter 28, Article 159 of the Criminal Code) are insufficient for qualifying the autonomous actions of AI; 2) new offenses are necessary (for example, "Creation of a malicious AI system") and qualifying features; 3) legislative consolidation of procedures for the seizure and examination of algorithmic models as physical evidence is required; 4) a risk-oriented federal law on AI should become the basis for systemic counteraction, establishing requirements for the safety, transparency, and auditing of high-risk systems.