Sports medicine (Auckland, N.Z.)

Expert Agreement on Using GPT-4 for Athlete Sleep and Jet Lag Guidance

Updated

Abstract

Large language models (LLMs) like GPT-4 are increasingly utilized in generating health information.

  • Concerns remain regarding the accuracy of health information produced by LLMs for elite athletes.
  • The relevance of generated content may vary based on the specific needs of elite athletes.
  • Continued evaluation of LLMs is necessary to ensure they meet the standards of health information for specialized populations.

Simplified

Full Text

Full text is available at the source.

Funding

Competing interests

Declarations. Conflicts of interest: Shona Halson and Alan McCall are Editorial Board members of Sports Medicine. Neither were involved in the selection of peer reviewers for the manuscript nor in any of the subsequent editorial decisions. All other authors declare no conflict of interest or competing interests. Availability of data and material: The datasets generated and analysed during the current study are available from the corresponding author, Jacopo Vitale (jacopo.vitale@uniecampus.it), upon reasonable request. Ethics approval: As specified in a letter from the local Ethics Committee of the Principal Investigator (J.V., dated 15/12/2023), ethical approval was not required because the broader research project, including both the prior LLM benchmarking study and the present Delphi consensus study, did not fall within the scope of the Human Research Act. The same waiver applied to both phases because the studies formed part of the same overarching project evaluating AI-generated sleep and jet-lag guidance for athletes. The present Delphi phase involved expert evaluation of FAQ responses and did not involve patients, athletes, clinical interventions or the collection of personal health data. Consent to participate: Expert members of ATSIG were invited and agreed to participate in both rounds of the Delphi process. Consent for publication: Not applicable.
PubMed

What Lands in Your Inbox Each Week:

  • 📚7 fresh studies
  • 📝plain-language summaries
  • direct links to original studies
  • 🏅top journal indicators
  • 📅weekly delivery
  • 🧘‍♂️always free