Innovations of Contemporary Artificial Intelligence: Value-Based Approach

Introduction . In recent years, the role of artificial intelligence (AI) in social life has increased. Innovations in the field of AI create not only new opportunities for a person and society as a whole, but also risks, problems and threats, which leads to the actualization of a risk-oriented approach towards the regulation of AI development on both the international and national levels. The problem of AI risk management has several interrelated areas, one of the central ones is the problem of identification of values on which the ethics of AI is built. The article considers issues related to the value-oriented direction of development of contemporary AI. Methodology and sources . In the article there were used the methodology of cultural philosophical, axiological and interdisciplinary approaches is used. The sources used in the article are scientific research of domestic and foreign authors, documents, publications and websites devoted to the current state of AI and its problems. Results and discussion . The topic of contemporary AI is developing as a special area of scientific and disciplinary knowledge, as well as a scalable AI industry. One of the latest trends in the current AI is emotional AI and its capabilities in establishing effective communication with person. However, despite the revolutionary nature of the new AI technology and the special significance of this innovation in AI communication with person, the problems of ethics and the value-oriented development of contemporary AI are defined by experts as key problems of our time. Conclusion . Contemporary AI is analyzed today from the standpoint of various classifications, the central place is occupied by the classification based on the comparison of human intelligence and AI. In this regard, the formalization of moral values and ethical principles in the process of developing and operating AI algorithms is important for the value-oriented human-AI interaction. However, the system of universal human values and AI values coincide only partially. As a result, new approaches emerge that do not compare AI and human intelligence, in particular, the interdisciplinary approach «4E Cognition», which is considered by experts as the most productive in all respects.