Can medical students use artificial intelligence to learn transfusion? Evaluating <scp>ChatGPT</scp> responses to the American Society of Hematology medical student transfusion learning objectives
Bibliographic record
Abstract
BACKGROUND AND OBJECTIVES: Chat generative pretrained transformer (ChatGPT) is a large language model that is already in wide use among medical students as a means of learning. Many papers have evaluated ChatGPT as a presenter of medical knowledge for the general public and as a test-taking engine. For students who rely on ChatGPT to learn transfusion medicine, it is important to understand the limitations of the application. MATERIALS AND METHODS: Transfusion content from the American Society of Hematology 'medical student learning objectives' was edited into questions for the ChatGPT interface. The answers generated by ChatGPT were then marked by three experienced transfusion medicine physicians. RESULTS: ChatGPT scored on average 2.27 ± 0.21 on a 4-point scale. Two-thirds of its answers scored A, B or C, representing excellent, good or satisfactory achievement, respectively. One-third of ChatGPT's answers were assigned a failing grade. Simple questions of basic transfusion science performed the best; more complex questions as well as questions where clinical practice has evolved substantially over the last several years performed the worst. Some answers were assessed to be unsafe in clinical practice. CONCLUSION: As a resource for medical students learning transfusion medicine, ChatGPT has significant limitations. A considerable proportion of its answers to transfusion questions are unreliable, inaccurate and even unsafe. These incorrect answers are presented with the same authoritative tone as its correct answers, and an inexperienced learner would be challenged to differentiate between true and untrue responses. At the present time, it is not recommended for medical students to use ChatGPT to learn transfusion medicine.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".