They should flag the AI-generated content as “inaccurate” or “too verbose”, assuming it actually is of course GPT detectors are inaccurate, full-stop. Although

They should flag the AI-generated content as “inaccurate” or “too verbose”, assuming it actually is of course GPT detectors are inaccurate, full-stop. Although “strict standard of evidence” is going to rule out most real AI-generated content, because there’s not many objective measures.