Researchers uncovered a preprint plagiarism network that they say uses large language models to rephrase legitimate work posted on preprint servers. The investigation, described under “Fake Authors, Real Citations,” details a citation mill designed to inflate counts by distributing derivative content under fake author identities. The finding matters for biotech publishing and clinical science because preprints and citation metrics can influence perceived novelty, visibility, and downstream investment or trial interest. A system that manipulates citation behavior can distort research signals. The report indicates investigators identified the network by detecting patterns consistent with AI-assisted rewriting and coordinated citation behaviors. This is the kind of threat that can be addressed through improved metadata checks, author identity verification, and algorithmic anomaly detection. For industry teams monitoring literature, heightened diligence may be required when assessing novelty claims that rely heavily on citation momentum rather than study quality and reproducibility.