Blame and Cost: Understanding Responsibility Attribution and Financial Risk in AI Usage

ABSTRACT This study examines public perceptions of responsibility and potential damage associated with AI usage across various scenarios, and estimates the financial impact on firms. The survey‐based analyses of AI failures evaluating hypothetical risk scenarios and their economic consequences reveal that sociodemographic determinants shape the perception of financial risks and responsibility in AI‐related scenarios. Respondents strongly attribute responsibility to AI system developers. Stratified analysis based on AI familiarity and educational background reveals significant differences in responsibility attribution, with respondents who are more familiar with AI demonstrating more nuanced perspectives. Furthermore, financial damage estimates vary according to respondents' gender, AI familiarity, and educational background, with female respondents and those less familiar with AI perceiving higher financial risks. The consistency between revenue‐based damage estimates and Likert‐scale financial burden assessments reinforces the robustness of these findings.