The "Boriviter" Center for Advanced Technologies has initiated testing of the "ODCR Assistant" neural network, designed to automate the process of collecting and analyzing data from enemy open sources. This is reported by the publication "Militarnyi". The project is implemented by the center's analytical department and aims to automatically identify useful information from the enemy's media landscape.

This module works with unclassified data streams, particularly from Russian military social networks and technical blogs. It processes around a thousand messages daily. The algorithms sort them using dictionary markers — "green" and "red". If a message contains "green" keywords and lacks "red" ones, it moves forward for observation formation.

The next phase will include checking the novelty of the data and its systematization. Developers note that complete automation of analysis is not yet achievable — human involvement remains essential.

In addition to the "ODCR Assistant", the team is working on a language model for Ukrainian military personnel, which will operate based on neural networks and will have an interface similar to ChatGPT. The goal is to facilitate dialogue with experienced practitioners and simplify situation analysis.

The development is being carried out by the "Boriviter" team on powerful servers. The organization is open to collaboration but emphasizes the importance of non-commercial support focused solely on the Ukrainian army.