Multi-agent systems as a technological basis for the implementation of zero trust concept

The paper considers the problem of applying multi-agent systems in the implementation of the concept of Zero Trust. Possible architectures of multi-agent systems used to solve the problems of collecting information about the state of the system, adapting the rules of the schemes for placing points of application of security policies are discussed. The features of the LSTM neural network application in the network traffic analysis system at the coordination level of a multi-agent system are considered.