Commentary Research and Evidence Ethics, Regulation, and Responsible Use

Stop pretending AI peer review isn’t happening

Why every journal article should have an AI reviewer―and how banning AI from peer review could be making science worse.

April 21, 2026 By Editor-in-Chief: Johnson Thomas, MD, FACE, FEAA 8 min read
Share
Objective:

To discuss the challenges of the current peer review system, including submission overload, reviewer shortages, and biases, while advocating for the integration of AI in the peer review process.

Key Findings:
  • Global scientific output is increasing rapidly, outpacing the availability of qualified reviewers.
  • AI can significantly improve the quality of peer reviews by addressing systematic issues that human reviewers often miss.
  • The ICLR 2025 experiment showed that AI feedback led to improved review quality without affecting acceptance rates.
Interpretation:

Integrating AI into the peer review process is crucial to alleviate current challenges and significantly enhance the quality and efficiency of scientific evaluations.

Limitations:
  • Concerns about confidentiality and AI 'hallucinations' remain significant but can be addressed with secure implementations and robust oversight.
  • The current system's biases and inefficiencies may still persist even with AI integration, necessitating ongoing evaluation and adjustment.
Conclusion:

A collaborative model that combines AI and human reviewers is essential for advancing the peer review process, effectively addressing its inherent challenges and improving overall scientific integrity.

AACE Endocrine AI is published by Conexiant under a license arrangement with the American Association of Clinical Endocrinology, Inc. (AACE®). The ideas and opinions expressed in AACE Endocrine AI do not necessarily reflect those of Conexiant or AACE. For more information, see Policies.

Related Content