There are significant differences among artificial intelligence large language models when answering scientific questions
The study highlights the need for structured frameworks for evaluating LLMs and ethical considerations essential for responsible AI integration in scientific research. These findings should be interpreted with caution, as the limited sample size and domain-specific focus of the exam questions restrict the generalizability of the results.
Authors
- Mario Aranda (ORCID: https://orcid.org/0000-0001-7846-9149)
- J.V. García‐Pérez (ORCID: https://orcid.org/0000-0002-9993-4246)
- Luis Miguel Pedrero Esteban (ORCID: https://orcid.org/0000-0003-4949-2360)
- Enrique Barrajón‐Catalán (ORCID: https://orcid.org/0000-0001-8113-0795)
- David Arraéz-Román (ORCID: https://orcid.org/0000-0003-1267-6676)
- Francisco Javier Álvarez‐Martínez (ORCID: https://orcid.org/0000-0002-5275-1282)
- Vicente Micol (ORCID: https://orcid.org/0000-0001-8089-0696)
- Fernando Borrás (ORCID: https://orcid.org/0000-0002-5519-4573)
- RuAngelie Edrada‐Ebel (ORCID: https://orcid.org/0000-0003-2420-1117)
- Andrew C. Ross (ORCID: https://orcid.org/0000-0003-3401-0306)
- María Herranz-López
- Pedro Mena
- Alessandro Zambon (ORCID: https://orcid.org/0009-0002-2025-9813)
- James Lyng
- Lucas Frungillo
- Estefanía Butassi
- Federica Pollastro
- Anne Sylvie Tixier
Institutions
- University College Dublin (IE)
- Universitat de Miguel Hernández d'Elx (ES)
- Università degli Studi del Piemonte Orientale “Amedeo Avogadro” (IT)
- University of Parma (IT)
- University of Leeds (GB)
- Pontificia Universidad Católica de Chile (CL)
- Universidad de Granada (ES)
- University of Strathclyde (GB)
- Université d'Avignon et des Pays de Vaucluse (FR)
- National University of Rosario (AR)
- Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement (FR)
- Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (ES)
- Universitat Politècnica de València (ES)
- University of Bologna (IT)
- University of Edinburgh (GB)
Publication Details
- Journal
- Frontiers in Artificial Intelligence
- Published
- 2025-10-09
- DOI
- https://doi.org/10.3389/frai.2025.1664303
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- European Commission
- Ministerio de Economía y Competitividad
- Generalitat Valenciana
- Instituto de Salud Carlos III