Emotional Responses to AI Use: Development of the SAGAT Scale for Shame and Guilt

The rise of generative AI tools has introduced new ethical dilemmas and emotional responses, yet little research has examined users’ social emotions such as shame, guilt, and impostor syndrome. This paper presents the SAGAT (Shame and Guilt towards AI Tools) scale, a 9-item psychometric instrument developed to measure these emotions in AI tool usage. Two cross-sectional studies confirmed a three-factor structure: AI Shame, AI Guilt, and AI Impostor syndrome, with strong internal reliability (α ≥ 0.75). The scale demonstrates measurement invariance across genders and predicts AI tool use frequency: guilt and impostor feelings are negatively associated, while shame is positively associated. Convergent and divergent validity were also supported. Emotional responses to AI appear to shape user experience and adoption, particularly in education and knowledge work. The SAGAT scale offers a novel way to assess affective dimensions of AI-human interaction, informing ethical design, AI literacy, and responsible integration of AI in society.