Since ChatGPT was released to the public, many people have started using it for various tasks, including information seeking and summarization, language translation, essay and scientific paper writing, and communication in different situations. However, unlike the systems for information search that we have been familiar with for a long time, such as expert systems, databases, information retrieval systems, and search engines, in order to take advantage of ChatGPT and similar systems, one must be aware of both the technologies on which they are based and their drawbacks. The talk will begin with an overview of AI methods applied to modeling information systems from the mid-20th century to the present day and will outline their peculiarities and differences. Concepts of knowledge representation and management, machine learning, statistical analysis, large language models, transparency, and explainability will be discussed in relation to the various types of system models. Finally, the potentialities and limitations of ChatGPT will be illustrated with examples of interaction.