Sycophantic AI, critical-feedback tolerance, and the case for productive friction in medical training

Concerns about generative AI in medical training often focus on eroded reasoning and deskilling. Such competence-based accounts are incomplete. We identify a neglected casualty: critical-feedback tolerance (CFT), the capacity to receive disconfirming evaluative feedback and remain productively engaged.

Affirmation-optimized systems may lower CFT undetectably, surfacing only under authentic critique. This conceptual Perspective defines CFT, differentiates it from six established constructs, specifies measurable facets, and states explicit refutation conditions. This work was funded by the National Natural Science Foundation of China (Grant No 82470070), the Noncommunicable Chronic Diseases-National Science and Technology Major Project (2024ZD0543500), and the Hospital-level Project of Huadong Hospital - Key Personnel Introduction Project (H2014).

The funders played no role in study design, data collection, analysis and interpretation of data, or the writing of this manuscript. These authors contributed equally: Zhi Chen, Bo Jia. Huadong Hospital, Fudan University, Shanghai, China Shanghai Medical College, Fudan University, Shanghai, China Huashan Hospital, Fudan University, Shanghai, China The First Affiliated Hospital of Soochow University, Suzhou, China Putuo District Maternal and Child Health Centre, Shanghai, China Shanghai Municipal Center for Disease Control and Prevention, Shanghai, China Correspondence to Mengyun Sui or Long Xue.

The authors declare no competing interests. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material.

You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. Chen, Z., Jia, B., Ji, K. et al. Sycophantic AI, critical-feedback tolerance, and the case for productive friction in medical training. npj Digit.