Artificial intelligence and weapons of mass destruction

There is considerable evidence in the modern history of warfare that when tensions between states are high, decision makers are more likely to seek detours or shortcuts to war to avoid great losses and costs. Accidental wars and unintentional escalations are actually relatively rare, despite fears and some signals of rising tensions. It is this type of warfare that is a major source of concern for many states, especially when combined with high-risk weapons systems or systems with the potential for mass destruction. The purpose of the article is to determine the role of artificial intelligence in the policy of nonproliferation of weapons of mass destruction. Among the most dramatic risks of escalation through the use of artificial intelligence is nuclear conflict. This topic has received significant and increasing political attention and research in recent years, driven by two main factors. The first is the tremendous progress that has been made in the field of artificial intelligence, especially machine learning. Such advances may provide an opportunity to improve early warning and decision support systems, or may contribute to improved targeting data. A second related reason is that competitive pressures to implement artificial intelligence in a different area amid the discourse of the "global race" have the potential to accelerate the adoption of artificial intelligence in the nuclear architecture. With its implementation, new risks of accidents, unforeseen escalation and vulnerability may arise. Artificial intelligence also expands the range of attack options that an attacker can take advantage of, including using cyber-attacks and information operations. The objective of the article is todetermine with the development of artificial intelligence, what role it will play in the non-proliferation of weapons of mass destruction or vice versa.