There is evidence that peer reviewers are using bullshit generators (chatbots) to generate evaluations of submitted papers.

Since those programs don't really understand the articles that are being reviewed, or the subject that those are about, this is asking for error.

Please don't call those programs "AI"!