Bolts from the blue: How technological novelty impacts crisis decision-making

In crisis, technological threats often present as unfamiliar to non-specialists. Today, however, emerging AI capabilities introduce distinct uncertainties for the decision-maker that extend beyond mere unfamiliarity, including about the fundamental capability of tools, underlying command structures, and moral accountability. This paper investigates how technological novelty, which I operationalize as multi-dimensional uncertainty beyond a baseline of simple decision-maker experience with a known system, influences decision-makers’ reaction to crisis conditions. To capture this spectrum, two large-N experimental simulations were run with military and policy professionals, embedding AI as a tactical aid, an autonomous strategic actor, or not at all. I then measured participants’ escalation choices, self-reported uncertainty, and requests for additional intelligence. Under conditions of most significant novelty in which AI assumed decision-making authority across several dimensions, participants showed marked hesitation and elevated uncertainty. By contrast, where AI presented as a support, tool decision-makers behaved similarly to those with AI systems exposure. Both AI scenarios, however, led to more intensive information-seeking than control. These findings suggest that crisis behavior is sensitive to the cumulative uncertainties of novel technologies as distinct from mere unfamiliarity. Accounting for this nuance offers a more precise framework for understanding foreign policy decision-making under technological disruption.