Is AI replacing faculty? Rethinking faculty roles in medical education

Abstract

Medical education is undergoing a profound transformation as artificial intelligence (AI) becomes increasingly embedded in both clinical practice and learning environments. A recent study by Zainal et al. revealed that early-career physicians are already facing ethical challenges in AI-mediated care for which their training has not adequately prepared them.1 Participants described uncertainty related to opaque algorithms, biased datasets, hallucinated outputs, unclear accountability, and the risk of over-reliance on AI, all of which are issues that reflect deeper tensions in clinical judgement, responsibility and professional identity. At the same time, AI is transforming medical education by reshaping how learners engage with content and are evaluated through tools such as AI-generated explanations, adaptive practice questions, automated feedback and performance analytics.2 Together, these shifts prompt a broader question: As both clinical practice and learning environments become increasingly AI-mediated, what remains the distinct and essential role of faculty in contemporary medical education? This commentary examines how the integration of AI across clinical practice and medical education is reshaping faculty roles and argues that, rather than diminishing their importance, AI clarifies the distinctly human work that defines educational expertise. To appreciate the significance of this shift in faculty's role, it may be helpful to consider how faculty roles have historically been defined. For much of modern medical education, faculty have taught primarily through the transmission of knowledge, with their authority grounded in privileged access to specialised information and the ability to interpret it for learners.3 This model was necessary in an era when knowledge was scarce and difficult to access. That era, however, has ended. Digital resources, online platforms and generative AI have made knowledge abundant, accessible and increasingly interactive. Students can now generate explanations, simulate clinical reasoning and receive feedback almost instantly, and often without direct faculty involvement.4, 5 This expanded access does not render faculty obsolete; rather, it signals a shift away from their traditional role as primary transmitters of knowledge. Zainal et al.'s findings bring this shift into sharper focus. Their participants did not describe any knowledge gaps. Instead, they described students' difficulty interpreting AI outputs, explaining them to patients, recognising their limitations and acting responsibly when errors occur.1 These interpretive and ethical challenges signal a gap not in access to knowledge, but in the formation of judgement. This reframing of faculty's roles aligns closely with the goals of competency-based medical education (CBME), which emphasises developmental progression, contextualised assessment and the integration of knowledge, skills and professional behaviours in practice.6, 7 However, implementation has thus far been constrained by the demands of observation, data synthesis and longitudinal assessment.8 AI may help mitigate some of these challenges by organising data, identifying patterns and generating immediate feedback. However, as Zainal et al. suggest, the domains that matter most such as epistemic awareness, relational integrity, reflexive accountability and adaptive professionalism are not easily automated.1 These domains thus highlight where faculty are becoming more, not less, essential. Faculty can guide learners in discerning when to trust AI and when to question it, model how to communicate AI-informed decisions while preserving patient trust and help navigate situations in which responsibility is distributed across clinicians, institutions, and technologies. In doing so, they teach students to refine their professional judgement in environments characterised by uncertainty and continual change. In this sense, AI does not replace faculty; it redistributes their workload. Tasks related to information delivery and assessment may increasingly be supported by technology. As these time-intensive responsibilities shift, faculty can devote greater attention to the relational dimensions of teaching, such as serving as coaches who guide learners through ambiguity, integrate competing inputs, and support the development of professional identity.9 This evolving role aligns closely with CBME, which emphasises longitudinal coaching, formative feedback and the integration of knowledge, skills and professional behaviours in authentic clinical contexts. In this model, faculty are not primarily content experts, but developmental coaches who interpret performance data, support individualised learning trajectories, and facilitate progression toward competence.6, 7 This shift in faculty's roles carries important structural implications. Institutions that continue to prioritise lecture-based teaching may become increasingly misaligned with how learning actually occurs. At the same time, the type of faculty work that is becoming more important, such as coaching, mentorship, and ethical guidance, may be less prioritised or rewarded. Without changes to faculty development, workload models, and promotion criteria, institutions risk undervaluing the very faculty work that AI makes most necessary. Zainal et al. further highlight that preparing learners for AI-mediated care requires more than technical training.1 Their framework emphasises ethical formation through practice, using strategies such as simulation, reflective work, and case-based reasoning. This strategic shift reinforces that faculty must be prepared not only to use AI, but to teach learners how to think critically about it, question it, and act responsibly in relation to it. The challenge for medical education, then, extends beyond integrating AI into curricula to rethinking what constitutes educational expertise. The true concern is not that AI will replace faculty, but that medical education may fail to recognise the transformation already underway. Rather than asking whether AI should replace faculty, the more important question is: What forms of faculty work become indispensable when knowledge is no longer scarce? Zainal et al.'s study offers a clear answer: As AI becomes embedded in both learning and clinical care, educators' value lies in cultivating judgement, sustaining relationships and supporting moral agency.1 At this critical moment of transformation within medical education, the role of faculty is not being diminished, but being brought into sharper focus. As AI reshapes how knowledge is accessed and applied, the most essential contributions of educators become clearer: guiding judgement, modelling integrity, and shaping the development of thoughtful, trustworthy physicians. Our task ahead is not to hold tight to traditional faculty roles, but to elevate and invest in this distinctly human work, ensuring that faculty are empowered to cultivate the insight, responsibility, and professional identity formation that will remain essential to medicine as technology evolves. The author was solely responsible for the conception, drafting, and revision of the commentary and approved the final version for submission. During the preparation of this manuscript, the author used generative AI tools (ChatGPT, OpenAI; version 2026) to support organisation, language refinement and synthesis of previously drafted sections. These tools were used to improve clarity, coherence and readability of author-generated text and to assist with restructuring content for flow. No unpublished data, confidential information or copyrighted material from prior editions was uploaded into any AI tool. All AI-generated suggestions were independently reviewed, fact-checked, edited and integrated by the author to ensure accuracy, completeness and alignment with the authors' original analysis and interpretation. The author verified all citations. The author maintains full responsibility for the content of the manuscript, including the accuracy of all information and the integrity of the arguments presented. The opinions and assertions expressed herein are those of the author and do not reflect the official policy or position of the Uniformed Services University of the Health Sciences or the Department of War. The author declares no conflicts of interest related to this work. This manuscript is a conceptual commentary and does not involve human participants, patient data or identifiable information. As such, institutional review board approval was not required. No new data were generated or analysed for this commentary. Data sharing is not applicable to this commentary.

Ähnliche Arbeiten