Some approaches to the analysis of factors affecting the information security of artificial intelligence systems

The article deals with the problems of ensuring the necessary level of information security of artificial intelligence systems which is one of the key conditions for their introduction into state practice. It is noted that the main specific threat is posed by the training procedures of artificial intelligence systems during which not only the information base but also the algorithms of intelligent conclusions are configured and can change. The information technology of analysis of certain factors is proposed which makes it possible to identify the features of ensuring the information security of artificial intelligence systems and characterize the totality of existing threats. Specific issues of information protection of artificial intelligence systems and a number of measures to ensure the information security of these systems are discussed.