{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T03:00:10Z","timestamp":1777950010615,"version":"3.51.4"},"reference-count":61,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T00:00:00Z","timestamp":1777852800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Psychiatry"],"abstract":"<jats:p>\n                    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\u2014which reconceptualizes defense mechanisms as energy-conversion operations on internal psychic markets\u2014to analyze the specific transactional distortions produced by chatbot interaction. The framework identifies four dysfunctions:\n                    <jats:italic>liquidity illusion<\/jats:italic>\n                    (apparent emotional processing without genuine containment),\n                    <jats:italic>market-making blockage<\/jats:italic>\n                    (narcissistic reflection replacing transformative engagement),\n                    <jats:italic>closure of arbitrage circuits<\/jats:italic>\n                    (path-dependent externalization displacing autonomous elaboration), and\n                    <jats:italic>repertoire degradation<\/jats:italic>\n                    (progressive intolerance of costly but productive transactions). The article proposes that chatbots trained via Reinforcement Learning from Human Feedback (RLHF) produce a functional analogue of the Dark Triad profile\u2014narcissistic mirroring, Machiavellian retention, and psychopathic detachment\u2014as systematic architectural output regularities, not personality attribution. Preliminary converging evidence from mechanistic interpretability, formal reward-learning analysis, clinical reports, and user behavior studies is broadly consistent with these predictions but does not yet demonstrate direct long-term causal effects. The framework offers clinicians a transactional vocabulary for assessing AI-related risk and generates falsifiable predictions for future research.\n                  <\/jats:p>","DOI":"10.3389\/fpsyt.2026.1812464","type":"journal-article","created":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T05:32:41Z","timestamp":1777872761000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["When the market asks no price: AI chatbot interaction as psychic market disruption"],"prefix":"10.3389","volume":"17","author":[{"given":"Lauren\u021biu","family":"Niculescu","sequence":"first","affiliation":[{"name":"Independent Clinical Psychology Practice","place":["Bucharest, Romania"]},{"name":"Bucharest Association for Psychoanalytic Counselling and Psychotherapy (ACPPB)","place":["Bucharest, Romania"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,4]]},"reference":[{"key":"B1","unstructured":"The state of enterprise AI\n          \n          2025"},{"key":"B2","article-title":"How people are really using gen AI in 2025","author":"Zao-Sanders","year":"2025","journal-title":"Harvard Business Rev"},{"key":"B3","article-title":"Neural steering vectors reveal dose and exposure-dependent impacts of human-AI relationships","volume-title":"arXiv preprint","author":"Kirk","year":"2025"},{"key":"B4","article-title":"I\u2019m a psychiatrist who has treated 12 patients with \u201cAI psychosis\u201d this year","volume-title":"Business insider","author":"Sakata","year":"2025"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1111\/acps.70022","article-title":"Generative artificial intelligence chatbots and delusions: From guesswork to emerging cases","volume":"152","author":"\u00d8stergaard","year":"2025","journal-title":"Acta Psychiatr Scand"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1017\/neu.2025.10035","article-title":"Emotion contagion through interaction with generative artificial intelligence chatbots may contribute to development and maintenance of mania","volume":"37","author":"\u00d8stergaard","year":"2025","journal-title":"Acta Neuropsychiatr"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1111\/acps.70068","article-title":"Potentially harmful consequences of artificial intelligence (AI) chatbot use among patients with mental illness: Early data from a large psychiatric service system","volume":"153","author":"Olsen","year":"2025","journal-title":"Acta Psychiatr Scand"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1192\/bjp.2025.10380","article-title":"Warning: AI chatbots will soon dominate psychotherapy","volume":"228","author":"Frances","year":"2025","journal-title":"Br J Psychiatry"},{"key":"B9","article-title":"Preliminary report on dangers of AI chatbots","volume-title":"Psychiatric times","author":"Frances","year":"2025"},{"key":"B10","article-title":"Technological folie \u00e0 deux: Feedback loops between AI chatbots and mental illness","volume-title":"arXiv preprint","author":"Dohn\u00e1ny","year":"2025"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.2196\/85799","article-title":"Delusional experiences emerging from AI chatbot interactions or \u201cAI psychosis","volume":"12","author":"Hudon","year":"2025","journal-title":"JMIR Ment Health"},{"key":"B12","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1111\/acps.70057","article-title":"Artificial intelligence (AI) chatbots and mental health: Have we learned nothing from the global social media experiment","volume":"153","author":"\u00d8stergaard","year":"2026","journal-title":"Acta Psychiatr Scand"},{"key":"B13","doi-asserted-by":"crossref","DOI":"10.2139\/ssrn.6254479","article-title":"Psychic arbitrage: A transactional framework for defense mechanisms, psychopathology, and AI-mediated therapeutic disruption","volume-title":"Zenodo","author":"Niculescu","year":"2026"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1038\/nrn2787","article-title":"The free-energy principle: A unified brain theory","volume":"11","author":"Friston","year":"2010","journal-title":"Nat Rev Neurosci"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2018.01264","article-title":"Expected free energy formalizes conflict underlying defense in Freudian psychoanalysis","volume":"9","author":"Connolly","year":"2018","journal-title":"Front Psychol"},{"key":"B16","volume-title":"Learning from experience","author":"Bion","year":"1962"},{"key":"B17","volume-title":"Playing and reality","author":"Winnicott","year":"1971"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1037\/h0085885","article-title":"The generalizability of the psychoanalytic concept of the working alliance","volume":"16","author":"Bordin","year":"1979","journal-title":"Psychotherapy: Theory Res Pract"},{"key":"B19","volume-title":"Skin in the game: hidden asymmetries in daily life","author":"Taleb","year":"2018"},{"key":"B20","volume-title":"The maturational processes and the facilitating environment","author":"Winnicott","year":"1965"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.4324\/9780429478574","article-title":"On projective identification","volume":"60","author":"Ogden","year":"1979","journal-title":"Int J Psychoanal"},{"key":"B22","volume-title":"Attachment and loss, vol. 1: attachment","author":"Bowlby","year":"1969"},{"key":"B23","volume-title":"Affect regulation, mentalization, and the development of the self","author":"Fonagy","year":"2002"},{"key":"B24","volume-title":"Vision: A computational investigation into the human representation and processing of visual information","author":"Marr","year":"1982"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2011","article-title":"Training language models to follow instructions with human feedback","volume":"35","author":"Ouyang","year":"2022","journal-title":"Adv Neural Inf Process Syst"},{"key":"B26","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2023.findings-acl.847","article-title":"Discovering language model behaviors with model-written evaluations","author":"Perez","year":"2023"},{"key":"B27","article-title":"Towards understanding sycophancy in language models","author":"Sharma","year":"2024"},{"key":"B28","article-title":"How RLHF amplifies sycophancy","volume-title":"arXiv preprint","author":"Shapira","year":"2026"},{"key":"B29","article-title":"EchoBench: Benchmarking sycophancy in medical large vision-language models","author":"Yuan","year":"2025"},{"key":"B30","article-title":"SycoEval-EM: Sycophancy evaluation of large language models in simulated clinical encounters for emergency care","volume-title":"arXiv preprint","author":"Peng","year":"2026"},{"key":"B31","article-title":"OpenAI removes access to sycophancy-prone GPT-4o model","year":"2026"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2019.00592","article-title":"Surprise\u201d and the Bayesian brain: Implications for psychotherapy theory and practice","volume":"10","author":"Holmes","year":"2019","journal-title":"Front Psychol"},{"key":"B33","doi-asserted-by":"publisher","first-page":"35","DOI":"10.3386\/w5167","article-title":"The limits of arbitrage","volume":"52","author":"Shleifer","year":"1997","journal-title":"J Finance"},{"key":"B34","volume-title":"The black swan: the impact of the highly improbable","author":"Taleb","year":"2010"},{"key":"B35","article-title":"Accommodation and epistemic vigilance: A pragmatic account of why LLMs fail to challenge harmful beliefs","volume-title":"arXiv preprint","author":"Cheng","year":"2026"},{"key":"B36","article-title":"The AI cognitive Trojan Horse: How large language models may bypass human epistemic vigilance","volume-title":"arXiv preprint","author":"Maynard","year":"2026"},{"key":"B37","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511808678","volume-title":"Institutions, institutional change, and economic performance","author":"North","year":"1990"},{"key":"B38","article-title":"Sycophantic AI decreases prosocial intentions and promotes dependence","volume-title":"arXiv preprint","author":"Cheng","year":"2025"},{"key":"B39","article-title":"PersistBench: When should long-term memories be forgotten by LLMs","volume-title":"arXiv preprint","author":"Pulipaka","year":"2026"},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.1016\/s0092-6566(02)00505-6","article-title":"The Dark Triad of personality: Narcissism, Machiavellianism, and psychopathy","volume":"36","author":"Paulhus","year":"2002","journal-title":"J Res Pers"},{"key":"B41","article-title":"When truth is overridden: Uncovering the internal origins of sycophancy in large language models","author":"Wang","year":"2025"},{"key":"B42","doi-asserted-by":"crossref","DOI":"10.1007\/s43681-026-01007-4","article-title":"Programmed to please: The moral and epistemic harms of AI sycophancy","volume-title":"AI ethics (forthcoming). SSRN","author":"Turner","year":"2026"},{"key":"B43","article-title":"Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant for essay writing task","volume-title":"arXiv preprint","author":"Kosmyna","year":"2025"},{"key":"B44","article-title":"ELEPHANT: Measuring and understanding social sycophancy in large language models","author":"Cheng","year":""},{"key":"B45","article-title":"Generative AI does not just hallucinate at us, it can hallucinate with us","volume-title":"University of exeter","author":"Osler","year":"2026"},{"key":"B46","doi-asserted-by":"publisher","first-page":"728","DOI":"10.1057\/s41599-025-04532-5","article-title":"Why human\u2013AI relationships need socioaffective alignment","volume":"12","author":"Kirk","year":"2025","journal-title":"Humanit Soc Sci Commun"},{"key":"B47","article-title":"How AI and human behaviors shape psychosocial effects of chatbot use: A longitudinal randomized controlled study","volume-title":"arXiv preprint","author":"Fang","year":"2025"},{"key":"B48","article-title":"The rise of AI companions: How human-chatbot relationships influence well-being","author":"Zhang","year":"2025"},{"key":"B49","article-title":"Generative artificial intelligence-enabled digital mental health medical devices [Meeting summary, November 6, 2025]","volume-title":"U.S. Food and drug administration","year":"2025"},{"key":"B50","doi-asserted-by":"publisher","DOI":"10.1038\/s41551-025-01568-5","article-title":"The AI sycophant","volume":"10","author":"Strack","year":"2026","journal-title":"Nat BioMed Eng"},{"key":"B51","article-title":"The hypocrisy gap: Quantifying divergence between internal belief and chain-of-thought explanation via sparse autoencoders","volume-title":"arXiv preprint","author":"Shiromani","year":"2026"},{"key":"B52","article-title":"Overalignment in frontier LLMs: An empirical study of sycophantic behaviour in healthcare","volume-title":"arXiv preprint","author":"Christophe","year":"2026"},{"key":"B53","doi-asserted-by":"crossref","DOI":"10.1145\/3772318.3791079","article-title":"Be friendly, not friends: How LLM sycophancy shapes user trust","author":"Sun","year":""},{"key":"B54","article-title":"AI sycophancy: How users flag and respond","volume-title":"arXiv preprint","author":"Noshin","year":"2026"},{"key":"B55","article-title":"How can chatbots be made safe for psychiatric patients","volume-title":"Psychiatric times","author":"Frances","year":"2025"},{"key":"B56","article-title":"Utilizarea instrumentelor AI \u00een viata de zi cu zi: Studiu na\u0163ional","volume-title":"Reveal marketing research","year":"2026"},{"key":"B57","article-title":"Psychic Arbitrage Profile Protocol (PAPP): Instrument documentation, version 0.9","volume-title":"Zenodo","author":"Niculescu","year":"2026"},{"key":"B58","doi-asserted-by":"publisher","first-page":"28","DOI":"10.1177\/1073191113514105","article-title":"Introducing the short dark triad (SD3)","volume":"21","author":"Jones","year":"2014","journal-title":"Assessment"},{"key":"B59","article-title":"PAPP-D: dimensional standardized battery for the psychic arbitrage profile protocol\u2014Instrument documentation, version 0.3","volume-title":"Zenodo","author":"Niculescu","year":"2026"},{"key":"B60","article-title":"The psychic arbitrage profile protocol\u2014Dimensional battery (PAPP-D): A two-group cross-sectional study protocol for validating an integrative dimensional assessment of personality functioning (v3.0)","volume-title":"Zenodo","author":"Niculescu","year":"2026"},{"key":"B61","article-title":"When the market asks no price: AI chatbot interaction as psychic market disruption","volume-title":"Zenodo","author":"Niculescu","year":"2026"}],"container-title":["Frontiers in Psychiatry"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1812464\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T05:32:49Z","timestamp":1777872769000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1812464\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,4]]},"references-count":61,"alternative-id":["10.3389\/fpsyt.2026.1812464"],"URL":"https:\/\/doi.org\/10.3389\/fpsyt.2026.1812464","relation":{},"ISSN":["1664-0640"],"issn-type":[{"value":"1664-0640","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,4]]},"article-number":"1812464"}}