From imposter to algorithmic attribution: A theoretical model of AI-induced professional self-doubt

Abstract

The entry of generative artificial intelligence (AI) into knowledge-intensive professional practice unsettles a question older than the technology itself: who is entitled to claim authorship of their own work. Drawing on attribution theory, self-efficacy theory, self-discrepancy theory, and professional identity research, this article develops a theoretical construct, AI-Induced Imposter Syndrome (AI-IIS), to describe a cognitive condition in which professional competence is misattributed to AI systems rather than consolidated into the self-concept. A three-stage process model is advanced, though each stage is offered as a hypothesis under stress rather than a finished description. The first proposed stage, algorithmic attribution confusion, captures the uncertainty that arises when professionals cannot clearly parse the human-versus-AI origin of performance outcomes. The second, interrupted mastery consolidation, specifies a mechanism by which self-efficacy fails to develop when successes are attributed externally. The third, professional identity destabilisation, concerns the erosion of authorship, agency, and intentionality as pillars of occupational selfhood. AI-IIS is differentiated from six adjacent technology-psychology constructs – technostress, automation anxiety, algorithm aversion, skill atrophy concerns, general AI anxiety, and AI-induced shame and guilt – though the boundaries remain permeable, and empirical clarification of overlap is itself part of the research agenda. Seven formal propositions are advanced with operationalisation anchors, alongside a measurement development roadmap for a proposed AI-IIS Scale (AI-ISS). Multilevel intervention pathways spanning cognitive attribution retraining, the Companion Model of human–AI collaboration, and institutional AI Declaration Frameworks are framed as testable experimental hypotheses, not solutions.