Objectives:
The primary objective of the research was to analyze the determinants that give rise to opportunities and risks associated with the deployment and utilization of artificial intelligence, specifically in the context of maintaining an adequate level of security for processed information. The cognitive goal of the article was to identify the factors influencing the secure implementation of machine learning-based tools. The author formulated the central research problem as follows: How should the deployment and operation of machine learning-based tools within the Armed Forces of the Republic of Poland (AFRP) be organized, while simultaneously ensuring the security of the information being processed? The implementation of machine learning-based tools is feasible within the existing instances of the "Polish Mission Network" ICT system, which is based on the concept of Federated Mission Networking.

Methods:
The research findings presented in this article are based on long-term observations and interviews conducted with experts. The empirical material was gathered during major military exercises held between 2015 and 2018.

Results:
If the results of the risk assessment process regarding the security of information stored and processed by machine learning engines are found to be acceptable, the author envisions the possibility of augmenting the existing instances of the Polish Armed Forces' ICT systems with AI services.

Conclusions:
Implementing the above measures can contribute to increasing the reliability of ICT infrastructure, reducing incident response times, and enhancing the overall level of security and operational efficiency of systems that utilize artificial intelligence.