Artificial Intelligence in Security Technologies: Levels of Intelligent Autonomy and Risk

Abstract

The development and use of Artificial Intelligence (AI) within security technologies presents a number of opportunities and risks for security professionals. The study was undertaken to investigate the use of AI in security technologies within the functional categories of Observing, Detecting, Controlling and Responding technologies, exploring the risks associated with enhanced AI across these technologies. In exploring these risks, the study developed a novel scale to define the level of intelligent autonomy a technology may possess during operation, in addition the degree of human involvement retained at each level across the various stages of the AI operational cycle. The Security Technology Intelligent Autonomy Scale may be used to consider the risks which may emerge at each level, with a weighting towards consequences of outcome.