UNRELIABILITY OF INFORMATION IN THE DIGITAL SPACE AS A CHALLENGE TO ARTIFICIAL INTELLIGENCE IN THE PROCESS OF RISK MANAGEMENT

Анотація

Schuchmann V.А. UNRELIABILITY OF INFORMATION IN THE DIGITAL SPACE AS A CHALLENGE TO ARTIFICIAL INTELLIGENCE IN THE PROCESS OF RISK MANAGEMENT Purpose. The aim of the article is to investigate the impact of unreliable information and disinformation flows in the digital environment on the efficiency of using artificial intelligence in the enterprise risk management system, as well as to substantiate multi-level approaches to data verification in order to improve the quality of strategic management decisions. Methodology of research. A combination of general scientific and special methods was applied in the research process. System analysis was used to structure information threats; analysis and synthesis ‒ to establish the relationship between data quality and algorithmic errors; graphic modelling ‒ to display the logic of risk escalation; methods of generalization and classification ‒ to organize tools for reducing information risks. Findings. Key threats of the digital information environment affecting the reliability of decisions formed using artificial intelligence are identified. It is proved that these threats have an external nature related to the manipulation of information flows, and an internal one conditioned by the peculiarities of generative models' operation. It is substantiated that data distortion triggers a chain reaction of risks, which manifests in assessment errors at the operational level and leads to strategic miscalculations in the long-term development of the enterprise. A conceptual model of risk minimization is proposed, which involves a combination of preliminary data verification, interpretation of algorithm results, and mandatory expert assessment of management decisions. Originality. The theoretical substantiation of ensuring information reliability in AI-based risk management systems has received further development. A multi-level classification of data verification methods according to their processing stages is proposed, and a model of the cascade transformation of information distortions into strategic losses of the enterprise is developed. Practical value. The research results can be used in the activities of enterprises to improve decision support systems, reduce financial losses from disinformation, and increase the resilience of strategic management in conditions of digital uncertainty. Key words: artificial intelligence, efficiency, digital economy, information unreliability, AI hallucinations, Explainable AI, strategic management, data verification.

Класифікація

Ідентифікатори

Схожі роботи