The process of digitalization of the economy and the widespread use of end-to-end technologies (artificial intelligence algorithms, robotics, wireless means of communication, etc.) generates new technological, economic, managerial and legal challenges, which requires the creation of a theoretical framework and the formation of a new conceptual and categorical apparatus, as well as an effective regulatory framework of digital law. One of the expected positive consequences of the implementation of the principles of the digital economy and the use of artificial intelligence is the reduction of transaction and transformation costs in the production and sale of products and services. This justifies the relevance of the chosen research topic. The scientific novelty of the research lies in the development of an economic model of transactional and transformational costs, which is the basis of the author's approach and algorithm for assessing the transactional sector of costs in the process of using artificial intelligence systems, as well as predicting positive consequences and obtaining an economic effect arising during the application of machine learning neural networks. It is concluded that the use of end-to-end technologies makes it possible to reduce the costs of time and other resources in the process of collecting, processing and analyzing information, which has a positive effect of reducing various groups of transaction costs. The results of the study can be useful for specialists engaged in machine learning in the process of analytical activity, persons making managerial decisions, experts involved in the assessment of the transactional sector of the economy, as well as representatives of government authorities during the development and implementation of state strategies for the development of the digital economy at the federal and regional levels.