Internal validation of the French Attitudes toward AI in Defense (AAID) scale: Evidence from an open dataset

Artificial Intelligence (AI) is becoming a central strategic concern for military powers. Understanding public attitudes toward its use in defense may be critical for predicting acceptance or rejection, underscoring the need for validated assessment tools in empirical research. In the present research, we aimed to validate the French version of a newly developed scale in English to assess Attitudes toward AI in Defense (the AAID scale; Hadlington et al., 2023). The original AAID scale captures positive (anticipated benefits) and negative (anticipated harms) attitudes toward using artificial intelligence in defense. After translating the scale into French with a back-translation procedure, we administered the scale to an online sample of 844 French-speaking participants, along with sociodemographic questions. Confirmatory factor analysis supported the hypothesized two-factor structure (positive and negative outcome dimensions) and showed satisfactory internal reliability. Sociodemographic variables such as gender and age were associated with attitudes. These findings suggest that the French version of the AAID scale is a valid tool to assess both positive and negative attitudes toward uses of AI in defense. The publicly available dataset we used, obtained prior to recent shifts in public and political discourse on military rearmament, can additionally serve as a reference point for future research documenting changes in these attitudes over time. More broadly, the availability of the AAID scale in French may facilitate the study of attitudes toward AI in defense within French-speaking populations and across cultural contexts and support the development of socially acceptable AI policies and applications.