Avery Orrall

Science flags paper that found AI chatbots help debunk conspiracy theories

Science has issued an expression of concern for a highly publicized study looking into whether conversations with AI chatbots could convince conspiracy theorists to abandon their beliefs. The move came after the authors of the paper found inconsistencies in their dataset, but a reanalysis shows the findings still stand, they say.

The September 2024 article found conversing with an AI chatbot called DebunkBot reduced people’s belief in a particular conspiracy theory by an average of 20%. The research was featured in news stories in The New York Times, Washington Post and The Atlantic.

This February, the authors — Thomas Costello of Carnegie Mellon University in Pittsburgh, psychologist Gordon Pennycook of Cornell University in New York and cognitive scientist David Rand at the Massachusetts Institute of Technology — won the Newcomb Cleveland Prize from the American Association for the Advancement of Science, which publishes Science, for the work. It has been cited 192 times, according to Clarivate’s Web of Science.

Editors of semantics journal resign, launch new journal after publisher ‘ultimatum’

The new journal, *Semantics of Natural Languages*, launched in May.

The editors of a semantics journal owned by Springer Nature have resigned to launch a new one, citing pressure from the company to increase their annual publication volume by 25%.

The editor-in-chief and the two associate editors Natural Language Semantics resigned from the journal in early April, editor-in-chief Amy Rose Deal, a linguistics professor at the University of California, Berkeley, told Retraction Watch. They resigned in consultation with the 25-member editorial board, which gave “a very high level of support for the move,” Deal said.

In a May 19 open letter, the three resigning editors and the founding co-editors of Natural Language Semantics announced the launch of their new journal, Semantics of Natural Languages. The new journal is published by the Open Library of Humanities, an open-access publisher whose goal “is to liberate university research from commercial control,” according to its website.

Journal retracts study linking hepatitis vaccine to autism that was included in CDC review

A toxicology journal has retracted a 16-year-old study linking hepatitis B vaccines to autism in children following an independent statistical review that found a half-dozen concerns with the study’s methodology.

Using data from the National Health Interview Survey, the authors claimed boys vaccinated in their first month of life had “threefold greater odds for autism diagnosis” than those vaccinated later or not at all.

The study was included in a rapid systematic review of hep B vaccine studies presented by John Su, director of the Immunization Safety Office for the Centers for Disease Control and Prevention, at the Sept. 18 meeting of the Advisory Committee on Immunization Practices. It resurfaced in a presentation at the Dec. 4 ACIP meeting, just before the committee’s decision to no longer recommend that infants receive the hep B vaccine at birth if the mother tests negative for the virus.

In what EIC calls an ‘honest mistake,’ journal approves paper without peer reviewing it

For most researchers, having an article accepted comes with constructive feedback from editors and reviewers. But when a sociology researcher learned his article was accepted at a Taylor & Francis journal, he was surprised to find the journal had skipped the peer review process altogether.

Martino C. submitted his article on the effects of economic instability on political ideology in Slovakia to the journal Democracy and Security on October 15. (We’ve withheld the author’s last name at his request for digital privacy reasons.) He told Retraction Watch he was hoping peer reviews would help him improve his argument.

But on January 13, the paper was marked “Accepted” in the journal’s submission portal without feedback.

Critics of birdsong study fight to be named in Nature’s retraction

A zebra finch in New South Wales, Australia. *Source: JJ Harrison/Wikimedia Commons (CC BY-SA 4.0)*

Researchers who flagged methodological issues in a paper on birdsong a year and a half before Nature retracted it say they should be credited in the editorial notice. But the editors have refused, with one telling the critics the paper was retracted for unrelated reasons.

The March 2024 study at the center of the dispute looked at how sexual selection may drive song patterns in male zebra finches. Nature retracted the paper last month because two of the synthetic song pairs used in the study were found to be unreliable, according to the notice. All three authors agreed to the retraction.

Todd Roberts, the paper’s corresponding author, told Retraction Watch the critics now asking for credit “prompted us to check the synthetic song pairs used in our paper.” He said his team did not do the reliability analysis of the pairs until after publication.

Widely criticized keto diet study retracted

A 2025 paper claiming the keto diet does not promote the formation of arterial plaques has been retracted after widespread criticism of the study’s methods and claims. The journal found “the identified errors are too great to be corrected with a corrigendum,” according to the March 11 retraction notice.

In April 2025, JACC: Advances published the study, which looked at plaque build-up in 100 otherwise generally healthy people who had experienced an increase in their cholesterol levels while being on a keto diet. The study claimed scans performed one year apart by the company Cleerly showed the diet was not associated with the development of arterial plaques.

This finding went against what previous studies had found, and it led to what Wired called “a new war in the nutrition world.”

Nine years after journalist raised concerns, BMJ Group journal retracts stent paper

A BMJ Group journal has retracted a paper nearly nine years after a journalist raised concerns about undisclosed conflicts of interest and the study’s details contradicting those of its trial registration. The researchers also excluded a patient’s death from the study, the retraction notice says.

The study, published in Open Heart in May 2017, described the results of a clinical trial that tested commercially available stents with microengineered grooves produced by Abbott Vascular.

But four months later, veteran cardiology journalist Larry Husten pointed out the clinical trial registration described a plan to employ two stents produced by a different device maker – Palmaz Scientific, a company that funded the work and was owned by one of the authors. He also wrote that records “indicate that one patient in the trial died as a result of pancreatitis. It seems unlikely that this was related to the stent but shouldn’t this information have been reported in the Open Heart paper?”

Elsevier journal removes two 42-year-old papers on cesium as a cancer treatment

An Elsevier journal has removed two papers on a discredited alternative treatment for cancer nearly half a century after they were published, after researchers found a quarter of patients in case reports of the therapy, cesium chloride, died from taking the substance.

Some alternative medicine advocates marketed cesium chloride as a cancer treatment in the 1980s and 1990s, although the risks and ineffectiveness of the therapy have been known for decades. In 2018, the U.S. Food and Drug Administration issued a warning about “significant safety risks” associated with the salt.

Marcel van der Heyden, a professor at the University Medical Center Utrecht in The Netherlands, told Retraction Watch he and his students came across the articles while writing a review of case reports on the use of cesium. Although the therapy was supported online and in health books, he said, all pointed to two 1984 papers in Pharmacology Biochemistry and Behavior: “Cesium therapy in cancer patients” by Hellfried Sartori and “The high pH therapy for cancer tests on mice and humans” by Aubrey Keith Brewer.

One in 277 PubMed-indexed papers in 2026 shows fabricated references, says analysis

*Figure from correspondence to* The Lancet by Maxim Topaz and colleagues.

Fabricated citations in the biomedical literature have increased 12-fold in two years, according to an audit of nearly 2.5 million papers published as a letter to The Lancet today.

The analysis of articles indexed in PubMed found that about one in 277 papers published in the first seven weeks of 2026 referenced a paper that didn’t exist. That was a jump from 2025’s rate of one in 458 and 2023’s one in 2,828. The researchers, led by Maxim Topaz of Columbia University’s Data Science Institute, used AI to “distinguish genuine fabrications from formatting discrepancies such as informally abbreviated titles.”

Topaz’s group located the sharpest increase in hallucinated references in mid-2024, which they note coincided with the rise of AI writing tools. The findings come as Nature reported last month that tens of thousands of publications from 2025 “might include invalid references generated by AI.” Retraction Watch has seen its fair share of reports of hallucinated citations generated by LLMs like ChatGPT.

Black marks on published papers don’t change citation rates, new study finds

Among the data analyzed were mean monthly citations per article for 151 papers that were retracted or issued some other editorial notice, and for a set of control articles. The solid vertical lines are median time to the peak citation month, and the dashed line is median time to the editorial notice.
*H. Studd et al/medRxiv 2026*

Neither retractions, expressions of concern, nor other editorial notices seem to keep authors from continuing to cite problematic papers, according to a look at what happened to more than 170 articles by one author.

“After the public notification of integrity concerns about an article, it would be expected that other authors would no longer cite the article because it is unreliable,” write the authors of a new preprint. But that’s not what they found in a limited comparative study. Whether the study is generalizable has yet to be seen, says one other expert.

Four sleuths – the University of Aberdeen’s Hugo Studd and Alison Avenell and the University of Auckland’s Andrew Grey and Mark J. Bolland – charted citation data for 172 papers on clinical trials from Zatollah Asemi, a nutrition researcher at Kashan University of Medical Sciences in Iran, whose work has come under scrutiny.

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this: