It is shown that it is expedient to interpret post-industrial local conflicts as “wars of the values” due to the fact that the result of the collision, as shown, in particular, by the conflict in Nagorno-Karabakh in October 2020, is largely determined by the technical and economic indicators of the robotic means of the warfare, used by the warring parties. The widespread use of the robotic systems, in particular, leads to the fact that it is advisable to significantly transform the approaches to their development, the production and subsequent combat use. In particular, it is advisable to use the physical implementations of the neural networks (more broadly, the artificial intelligence systems), each element of which (analogue of the neuron) is a relatively independent element (for example, the cheapest unmanned aerial vehicle). This approach uses one of the basic properties of a neural network its ability to perform basic functions even when a significant part of the elements is lost. The survivability of such a system will certainly be higher in comparison with any apparatus made as a physical whole. Moreover, the inclusion of a large number of the cheapest false (or partially false) targets in the combat neural network makes its suppression economically difficult due to the high consumption of the ammunition (in any case, existing types). It is significant that in the “war of values” a significant part of the costs falls on the development and implementation of the innovations in production. The transition to the use of physically distributed systems (combat neural networks) makes it possible to drastically reduce the costs of the prototyping (for example, cheap UAVs based on mini-airships), which makes it possible to maximize the intellectual and creative potential of the technical specialties students directly within the educational process. The feasibility of this approach has been proven in our works on the business educational ecosystems.