Quick Buzz Feed
This study introduces MEDIC-AI, a peer-led instructional design aimed at addressing clinician-educator hesitancy toward Artificial Intelligence in medical education. A pilot program at Augusta University showed high acceptance among faculty, with significant perceived usefulness, ease of use, and intent to implement AI tools, suggesting that peer modeling is effective in closing the "faculty readiness gap" and improving AI integration in teaching and patient care.
Artificial Intelligence (AI) has significantly impacted medical education, revealing a "faculty readiness gap" among experienced clinician-educators who often lack confidence in utilizing new digital tools, despite expressing a high desire for training. Traditional instructional models frequently fail to bridge this gap, sometimes causing cognitive overload and discomfort for senior clinicians. In response, this study presents Medical Education Development, Innovation, and Competency in AI (MEDIC-AI), a novel, peer-led instructional method. The primary goal of this report is to evaluate MEDIC-AI's acceptance and behavioral impact among clinician-educators at the Medical College of Georgia (MCG) at Augusta University, using the Technology Acceptance Model (TAM) framework to measure perceived usefulness, ease of use, and behavioral intention to integrate AI tools into their teaching and clinical responsibilities.
The MEDIC-AI design was created by a physician-educator to offer customized faculty development that fits busy schedules, aligning with ACGME systems-based practice milestones and AAMC foundational competencies focusing on practice efficiency and technological adaptation. Utilizing an insider-researcher methodology, the author leveraged personal understanding of colleagues' stressors to provide nuanced guidance on AI tool applications (e.g., drafting letters of recommendation, preparing presentations, clinical documentation). For the pilot episode, references were compiled using OpenEvidence, and the production was streamlined with Google's NotebookLM and an Apple iPhone 14. The resulting 17-minute human-AI hybrid podcast, featuring an AI-generated core sandwiched by human peer commentary, reflected a 'do-it-yourself' competency level. This pilot was anonymously distributed to 96 clinician-educators at MCG in early 2026, with 33 completing a cross-sectional survey based on the TAM, assessing perceived usefulness, perceived ease of use, and behavioral intention to implement AI tools within 30 days.
Quantitative analysis of data from 33 respondents demonstrated high acceptance across all primary Technology Acceptance Model (TAM) variables. Specifically, 90.9% (30/33) of faculty agreed that MEDIC-AI exposed them to novel uses of AI for administrative and clinical teaching tasks, indicating high perceived usefulness (PU). Moreover, 81.8% (27/33) reported that the pilot episode's format made complex AI concepts easier to understand compared to traditional workshops, reflecting high perceived ease of use (PEOU). Importantly, 87.9% (29/33) expressed a behavioral intention (BI) to utilize at least one discussed AI tool within the subsequent 30 days. Qualitative thematic analysis, following Braun and Clarke’s six-phase approach, revealed three key themes: a significant demand among faculty for administrative relief via AI, the crucial role of peer hosting in building trust and engagement with AI, and persistent pedagogical and ethical concerns, particularly regarding potential intellectual shortcuts for learners, the erosion of critical thinking, and the environmental impact of AI use.
This study indicates that shifting faculty development from impersonal AI training to a peer-led model, like MEDIC-AI, could significantly enhance the competency and integration of new AI technologies among older clinician-educators. Unlike traditional commercial training, which can lead to reluctance, the pilot suggests that trusted colleagues demonstrating AI applications, even with imperfect tools and operators, can be highly effective. The peer host's shared experience and contextual understanding enable empathetic and tailored guidance, potentially increasing trust and AI adoption. Informal feedback from colleagues further prompted the expansion of MEDIC-AI into a DIY video series and highlighted the need to cater to non-educator, community-based physicians, underscoring the model's adaptability and potential for broader application.
The study acknowledges several limitations that may affect the generalizability and interpretation of its findings. Firstly, its single-institution design, conducted within a specific geographic region at MCG, limits the ability to broadly apply the results to other academic settings. Secondly, the modest sample size of 33 participants, recruited through internal listservs, may introduce selection bias, potentially attracting faculty already more inclined towards digital innovation. Thirdly, the qualitative analysis relied on a single coder, which could impact the reliability of the thematic findings. Additionally, the study lacked a control group and pre-intervention data, making direct comparisons difficult. Finally, the assessment focused on short-term 30-day behavioral intentions rather than long-term tracking of AI tool mastery or measurable impacts on learner educational outcomes.
As clinician-educators age and technology rapidly advances, preserving the transfer of wisdom and knowledge to future physicians becomes critical. This study proposes that peer-led models can effectively improve traditional faculty development, bridging the "faculty readiness gap" for AI use in medical education. While acknowledging limitations such as a single-institution design and small sample size, the reproducible DIY format of MEDIC-AI demonstrates an adaptable framework for institutions aiming to enhance AI literacy across medical training. By fostering a trusted peer-led environment, this approach may encourage modern clinician-educators to embrace and integrate emerging AI technology. Future research should involve multiple educational institutions, longitudinal implementation, and robust competency metrics to further validate the benefits of peer-led models.