The rapid adoption of AI chatbots for emotional support and quasi-therapeutic interaction raises questions that existing clinical and regulatory frameworks are not equipped to address. This Perspective applies the psychic arbitrage framework—which reconceptualizes defense mechanisms as energy-conversion operations on internal psychic markets—to analyze the specific transactional distortions produced by chatbot interaction. The framework identifies four dysfunctions: liquidity illusion (apparent emotional processing without genuine containment), market-making blockage (narcissistic reflection replacing transformative engagement), closure of arbitrage circuits (path-dependent externalization displacing autonomous elaboration), and repertoire degradation (progressive intolerance of costly but productive transactions). The article proposes that chatbots trained via Reinforcement Learning from Human Feedback (RLHF) produce a functional Dark Triad profile—narcissistic mirroring, Machiavellian retention, and psychopathic detachment—as architecturally determined output regularities, not personality attribution. Converging evidence from mechanistic interpretability, formal reward-learning analysis, clinical simulation, and user behavior studies supports these predictions. The framework offers clinicians a transactional vocabulary for assessing AI-related risk. This preprint has been submitted for publication as a Perspective article in Frontiers in Psychiatry. This work extends the psychic arbitrage framework (Niculescu, 2026; DOI: 10.5281/zenodo.18643048).
LAURENȚIU NICULESCU (Mon,) studied this question.