Content of review 1, reviewed on April 13, 2024

Basic reporting The paper is well-organized and offers a clear narrative that effectively highlights the differences between the AI models in a clinical context. While, I have outlined some specific areas where enhancements could be made to further strengthen your submission.

  1. Provide a more detailed description of vignette selection and construction to ensure replicability and discuss the ethical implications of using AI for mental health assessments.
  2. Improve grammar and style for clarity and professionalism, ensure consistent terminology throughout the paper, and update and standardize reference formatting. Experimental design The methodology section is thorough and provides a solid foundation for the study. Validity of the findings The discussion aligns well with the results, offering thoughtful interpretation and implications for future research and practice.

Source

    © 2024 the Reviewer (CC BY 4.0).

References

    Shiri, S., Zohar, E., Inbar, L. 2024. The impact of history of depression and access to weapons on suicide risk assessment: a comparison of ChatGPT-3.5 and ChatGPT-4. PeerJ.