Performance of 3 Conversational Generative Artificial Intelligence Models for Computing Maximum Safe Doses of Local Anesthetics: Comparative Analysis
Generative AI models like Gemini, ChatGPT, and Copilot currently lack the accuracy and reliability needed for safe LA dose calculation. Their poor performance suggests that they should not be used as decision-making tools for this purpose. Until more reliable AI-driven solutions are developed and validated, clinicians should rely on their expertise, experience, and a careful assessment of individual patient factors to guide LA dosing and ensure patient safety.
Authors
- Caroline Samer (ORCID: https://orcid.org/0000-0001-8178-0616)
- Pietro Elias Fubini (ORCID: https://orcid.org/0000-0003-1371-6465)
- Georges L. Savoldelli (ORCID: https://orcid.org/0000-0002-8968-6920)
- Mia Gisselbaek (ORCID: https://orcid.org/0000-0002-2296-8106)
- Mélanie Suppan (ORCID: https://orcid.org/0000-0002-8807-9619)
- Alexandra Stefani (ORCID: https://orcid.org/0000-0002-7483-2075)
Institutions
- University of Geneva (CH)
- University Hospital of Geneva (CH)
Publication Details
- Journal
- JMIR AI
- Published
- 2025-04-01
- DOI
- https://doi.org/10.2196/66796
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00