The use of artificial intelligence (AI) in peer review is increasingly necessary due to the growing number of scientific manuscripts, the shortage of peer reviewers, and concerns about the effectiveness, fairness, and efficiency of traditional peer review. Large-language models, a form of AI, have the potential to assist in the triaging of manuscripts for peer review. While peer review is essential for ensuring the quality of scientific publications, it faces challenges such as detecting research misconduct, bias, and inefficiency. AI could be more effective at detecting research misconduct and assessing adherence to reporting guidelines than human peer reviewers. Peer review systems vary, including open, single-blind, and double-blind reviews, each with its own biases. Studies have shown that peer reviewers can be biased based on the author's reputation or language proficiency. AI could potentially reduce such biases. The number of scientific publications is rising rapidly, and peer reviewers are increasingly overwhelmed. AI could help alleviate this burden by assisting in the initial screening of manuscripts. Recent studies suggest that AI, such as GPT-4, can provide useful feedback on research papers, with high agreement between AI and human reviewers. However, AI is not yet sufficient to replace peer review entirely. Concerns about AI include data privacy and the potential for AI to be biased. Nevertheless, AI could be developed to be less biased and more effective in peer review. The authors envision a future where AI is used to initially screen manuscripts, providing a summary of their quality, which would then be reviewed by editors before peer review is requested. This approach is likely to become more common in the near future. Editors should embrace AI to improve the peer review process, ensuring that manuscripts are fairly and appropriately evaluated.The use of artificial intelligence (AI) in peer review is increasingly necessary due to the growing number of scientific manuscripts, the shortage of peer reviewers, and concerns about the effectiveness, fairness, and efficiency of traditional peer review. Large-language models, a form of AI, have the potential to assist in the triaging of manuscripts for peer review. While peer review is essential for ensuring the quality of scientific publications, it faces challenges such as detecting research misconduct, bias, and inefficiency. AI could be more effective at detecting research misconduct and assessing adherence to reporting guidelines than human peer reviewers. Peer review systems vary, including open, single-blind, and double-blind reviews, each with its own biases. Studies have shown that peer reviewers can be biased based on the author's reputation or language proficiency. AI could potentially reduce such biases. The number of scientific publications is rising rapidly, and peer reviewers are increasingly overwhelmed. AI could help alleviate this burden by assisting in the initial screening of manuscripts. Recent studies suggest that AI, such as GPT-4, can provide useful feedback on research papers, with high agreement between AI and human reviewers. However, AI is not yet sufficient to replace peer review entirely. Concerns about AI include data privacy and the potential for AI to be biased. Nevertheless, AI could be developed to be less biased and more effective in peer review. The authors envision a future where AI is used to initially screen manuscripts, providing a summary of their quality, which would then be reviewed by editors before peer review is requested. This approach is likely to become more common in the near future. Editors should embrace AI to improve the peer review process, ensuring that manuscripts are fairly and appropriately evaluated.