As publishers work to clean up a large number of fake papers and combat peer review fraud, the number of retracted research papers has exceeded 10,000 in 2023, breaking the annual record.

According to "Nature", most of the retracted papers in 2023 came from journals owned by Hindawi. Hindawi is a subsidiary of publisher Wiley in London, England. Hindawi has retracted more than 8,000 papers so far this year, citing concerns about problems with the peer-review process and systematic manipulation of publications and the peer-review process. Journal editors and research integrity experts have raised questions about incoherent text and irrelevant references in thousands of papers.

Most of Hindawi's retracted papers come from special issues - collections that are often run by guest editors and often publish low-quality or falsified papers quickly.

On December 6, Wiley announced that it would completely stop using the Hindawi brand name. Previously, Wiley had shut down four Hindawi magazines and suspended the publication of special issues at the end of 2022. Wiley's interim CEO Matthew Kissner revealed that the company expects to lose $35 million to $40 million in revenue this fiscal year due to these issues.

A Wiley spokesman said Wiley would retract more papers in the future, but he did not disclose a specific number. However, the company believes that "the special issue will continue to play a valuable role in serving the scientific research community." The spokesperson added that Wiley has implemented more rigorous procedures for identifying guest editors and overseeing publication, and has removed hundreds of bad actors from the system and expanded its research integrity oversight team. Wiley is also "pursuing legal options" to share lists of bad actors with other publishers, tool and database providers.

Guillaume Cabanac, a computer scientist at the University of Toulouse in France, said that most of the retracted Hindawi papers were probably fake, but they were still cited more than 35,000 times. The paper contained "strange phrases" used to evade plagiarism checks, as well as undisclosed signs of the use of artificial intelligence.

Today, the number of retractions is growing faster than the number of published papers, and this year's wave of retractions means that the total number of retracted papers published worldwide so far has exceeded 50,000. Although previous analysis has shown that most retractions are due to misconduct, some are retracted because the authors discovered errors in their research.

An analysis in Nature shows that retraction rates—the proportion of papers published in any given year that are retracted—have more than tripled in the past decade. In 2022, this proportion will exceed 0.2%.

Relevant experts believe that the approximately 50,000 retracted papers are just the tip of the iceberg. Essay mills alone are estimated to produce hundreds of thousands of papers, as well as papers with genuine scientific flaws. "The 'products' of paper mills are a problem because even if no one reads them, they are aggregated into review articles and 'laundered' into the mainstream literature," says New Zealand research integrity expert David Bimler.