Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1571
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-16T10:42:12.879610Z
status: published
lens: trace
sector: health
headline: Comparative evaluation of ChatGPT and Gemini feedback in scenario-based clinical reasoning among medical interns
dek: Objective: Artificial intelligence tools are increasingly used to provide individualized feedback in medical education. The aim of this study was to compare the quality of feedback generated by ChatGPT and Gemini during scenario-based clinical reasoning activities.
Methods: A convergent mixed-methods study was conducted with 32 medical interns randomly assigned to ChatGPT (n=16) or Gemini (n=16). Following a virtual patient interview on iron deficiency anemia, artificial intelligence-generated feedback was evalu…
gain_title: Both chatbots produced feedback that interns and medical education specialists rated as clear, supportive and educationally useful in a virtual iron deficiency anemia case.
problem_title: Feedback from both chatbots lacked patient-centeredness and systematicity and sometimes included unrealistic statements, so it cannot replace expert supervision.
trace_subject: AI-generated feedback on clinical reasoning for medical interns
gain_reading: Both chatbots produced feedback that interns and medical education specialists rated as clear, supportive and educationally useful in a virtual iron deficiency anemia case.
gain_evidence: ChatGPT and Gemini generated educationally useful feedback for scenario-based clinical reasoning activities. | Both systems were perceived as clear, supportive, and constructive
problem_reading: Feedback from both chatbots lacked patient-centeredness and systematicity and sometimes included unrealistic statements, so it cannot replace expert supervision.
problem_evidence: Common limitations included insufficient patient-centeredness, lack of systematicity, and occasional unrealistic statements.
quick_read: Researchers randomly assigned 32 medical interns to receive feedback from either ChatGPT or Gemini after a virtual patient interview on iron deficiency anemia. Interns and medical education specialists then scored the feedback with an 11-item rubric.

The results suggest that general-purpose chatbots can supplement personalized feedback in medical training and may ease faculty workload, but only under human oversight. The small sample and a single clinical scenario leave open how well the findings apply to other cases, settings or tools.
limitation: The study compared the tools in a single virtual scenario with 32 interns and found no significant differences, while both tools shared limits such as weak patient-centeredness and occasional unrealistic statements.
tag: Dual reading
key_points: The study enrolled 32 medical interns, split evenly between ChatGPT (n=16) and Gemini (n=16). | Interns scored ChatGPT 1.80±0.20 and Gemini 1.75±0.26; specialists scored them 1.91±0.06 and 1.87±0.10. | No statistically significant differences were found between the two systems (p>0.05). | ChatGPT feedback was more structured and reflective, while Gemini offered more concrete examples.
rundown: The design was a convergent mixed-methods study. Quantitative scores were analyzed with descriptive statistics and Mann-Whitney U tests, and qualitative comments were analyzed through content analysis.

The rubric covered clarity, accuracy, structure, identification of strengths and weaknesses, contribution to clinical reasoning, and overall educational value. Raters included both the interns and medical education specialists.

The authors conclude that the tools may support personalized learning and reduce faculty workload when used under human oversight, but they cannot replace expert supervision.
sources:
- peer_reviewed | Revista da Associação Médica Brasileira | https://doi.org/10.1590/1806-9282.20260869 | 2026-10-12
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

69e7fabaa44694269beb6c93a12788be6472ee63add2408d72bb63840412aa30

previous

0000000000000000000000000000000000000000000000000000000000000000