THE ARTIFICIAL INTELLIGENCE RISKS
WHEN USED IN SOCIOTECHNICAL SYSTEMS 

Mikhail Yu. Mikheev, Doctor of technical sciences, professor, head of the sub-department of informational technologies and systems, Penza State Technological University (1a/11 Baydukov passage/Gagarin street, Penza, Russia)
Oleg V. Prokofiev, Candidate of technical sciences, associate professor, associate professor of the sub-department of informational technologies and systems, Penza State Technological University (1a/11 Baydukov passage/Gagarin street, Penza, Russia)
Irina Yu. Semochkina, Candidate of technical sciences, associate professor, associate professor of the sub-department of informational technologies and systems, Penza State Technological University (1a/11 Baydukov passage/Gagarin street, Penza, Russia)

Background. The use of autonomous devices with artificial intelligence in socio-technical systems has led to the emergence of new problems, the solution of which is much more complex compared to the tasks of the previous stage of improving the human-machine interface. The authors of the work conducted a study to identify risk sources in socio-technical systems and ways to reduce them at the stages of the life cycle of autonomous devices. Materials and methods. The study used data from open reports on statistical surveys of developers and users of critical devices, reports on the study of accidents of autonomous transport Uber AV. Results. The formulations of risk sources arising from the use of autonomous devices with artificial intelligence in socio-technical systems are given. An example of the life cycle of a device is presented, providing for human control at the stages of development and operation. Conclusions. Risk management in cases of critical use of a device is possible if the actual stages of the life cycle correspond to a number of safety criteria listed in the article.

artificial intelligence risks, socio-technical systems, autonomous devices, critical device

Mikheev M.Yu., Prokofiev O.V., Semochkina I.Yu. The artificial intelligence risks when used in sociotechnical systems. Nadezhnost' i kachestvo slozhnykh sistem = Reliability and quality of complex systems. 2025;(1):12–19. (In Russ.). doi: 10.21685/2307-4205-2025-1-2