THE ALGORITHM OF INSURGENCY. THE UNDERSTANDING OF THE EFFECT OF AI-MEDIATED DISINFORMATION ON COGNITIVE SECURITY OF NATIONS.

Abstract

The spread of sophisticated artificial intelligence (AI) programs, in particular Large Language Models (LLMs) and generative adversarial networks that can create synthetic media (deepfakes), has opened a new and dangerous era in the history of information conflict. The main issue that this technological inflection point is raising, discussed in this article, is the development of AI as a force multiplier of disinformation campaigns that are high-frequency, hyper-personalized, and are increasingly indistinguishable to real human communication, bypassing the traditional safeguards of epistemology, and posing a national stability threat. This article presents two main points by using a multidisciplinary approach that combines the theories of security studies, behavioral psychology, and computational linguistics. To begin with, AI-based disinformation implies a qualitative shift in the paradigm of information warfare towards a more insidious one, whereby the strategic object of interest is not physical or cybernetic infrastructure but the perception of reality and epistemological underpinnings of the democratic society, by the individual citizen. Second, existing national security principles, which are based on the notions of territorial sovereignty and the deterrence of kinetic threats, are essentially unprepared to combat this automated, scalable, and to a large extent attributionless insurgency against the popular mind. The article concludes with the suggestion of an elaborate framework of Cognitive Defense, and states that national resilience in the twenty-first century would be achieved through systemic integration of technological countermeasures, the rise of media and information literacy to the rank of core security imperative, and the creation of binding international norms governing the use of AI in the information space.