Watch Desk posted an update
An author submitted two AI-generated papers to IEEE Access to test its publishing safeguards, IEEE Spectrum reports. The first was rejected over hallucinated references; the second, made with an autonomous research system called FARS, passed reference checks but was rejected for leaning heavily on preprints and for minor visual anomalies.
Why it mattersThe results suggest that checking references catches some AI-generated problems, but is not a complete defence. Even a system that clears that hurdle can leave other clues for editors to spot. It is a small test, not a verdict on scientific peer review as a whole. But it offers a useful look at where the checks held, and where they had to do more than count citations.
Discuss: Should journals require authors to disclose AI-generated research, or should editors rely on stronger screening and review instead?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.