Naomi Orseks: How Monsanto fabricated and infiltrated a fake scientific article that led to its deregulation of a toxic substance

In a remote lecture at the annual Science and Environment Conference in Jerusalem, the Harvard University historian of science argued that the data does not indicate a widespread crisis of public trust in science. She said that articles written under the influence of corporations could shape regulation, influence the courts and undermine the credibility of science from within.

Harvard University professor Naomi Orseks warned that the political attack on science in the United States is serious, but it is not the only threat to public trust. She said the scientific community must also confront a problem that originates within the research system itself: corporate interference in science, articles written by commercial companies without proper disclosure, and the continued use of problematic studies years after their flaws were exposed.

Orseks delivered her lecture remotely on Wednesday, July 8, 2026, as part of the annual Science and Environment Conference of the Israeli Society for Ecology and Environmental Sciences, which took place on July 8 and 9 at the Binyamin Ha'ama Congress Center in Jerusalem. She apologized to the participants for not being able to attend and lecture on site.

Orseks is a professor of the history of science and adjunct professor of Earth and planetary sciences at Harvard University. She is particularly known for her research on climate denial, the economics of scientific knowledge, and the ways in which corporations and interest groups spread doubt about scientific findings. Among her notable books is Vendors, which she wrote with Eric Conway.

An all-out attack on American science

At the beginning of her lecture, Orseks described the US government’s actions as “an all-out attack on American science.” She cited cuts to federal research budgets and scientific partnerships, pressure on universities, attempts to interfere with the admission of employees and students, curricula and the language used by researchers, as well as accusations of anti-Semitism that she said were sometimes exaggerated or unfounded.

Many of the actors promoting these measures, she said, are familiar to her and Conway from their research on climate crisis denial. The actors, language, and claims of “climate panic” are not new, but the intensity of the measures being deployed today is new and alarming.

But Orseks wanted to examine whether the political attack actually reflects a collapse in public trust in science. She argued that despite the widespread discourse about a “crisis of trust,” the data does not support the conclusion that there is such a general collapse.

An international study based on nearly 72 respondents in 68 countries found that the level of trust in scientists is medium-high in most countries. In the United States, Orseks added, between 70% and 80% of the public expressed a great or fair degree of trust in science over the past decade — more than the trust given to the government, the media, or the banking system.

“Sometimes science cannot be trusted”

From here, Orseks moved on to the more complex argument in her lecture. According to her, the question is not just how to convince the public to trust science, because not everything published under the title “scientific research” is automatically worthy of trust.

“Sometimes science cannot be trusted,” she said. In recent years, she explained, cases have been revealed in which commercial companies influenced the design of studies, the selection of data, the writing of articles, and the way their conclusions were presented.

As a test case, Orseks presented an article published in 2000 in the journal Regulatory Toxicology and PharmacologyThe article was signed by Gary Williams, Robert Cross, and Ian Munro, and was presented as an independent and comprehensive review of the safety of the herbicide Roundup and its active ingredient, glyphosate.

The article concluded firmly that under current and anticipated conditions of use, Roundup poses no risk to human health. For years, it was cited in studies, policy documents, and regulatory discussions, becoming one of the leading sources used by those who claim that glyphosate is safe.

The workers wrote, the academics signed

In 2017, internal Monsanto documents, known as the “Monsanto Papers,” were revealed in legal proceedings. The documents indicated that company employees played a central role in planning and writing the paper, even though their names did not appear on the list of authors.

In one internal correspondence, concerning the preparation of another paper after the International Agency for Research on Cancer classified glyphosate as a “probable carcinogen,” a Monsanto executive suggested that the company do the writing while outside academics “edit and sign off.” He mentioned that the company had also done the same with a 2000 paper.

This behavior is called Scientific shadow writing ghostwritingA commercial entity co-authors or writes most of the article, but the article is presented as if it were written independently by external researchers. In this way, the commercial message is given the appearance of independent scientific authority.

Orseks and Alexander Kaurov then examined how the article continued to influence the corporate involvement after it was exposed. They found that its citation count did not decline significantly. It continued to appear in academic literature, policy documents, and Wikipedia entries, in most cases without citing its problematic source. The article is in the top five cited articles on glyphosate, and appears in sources that also influence the information fed into artificial intelligence systems.

According to Orseks, the paper achieved exactly what it was intended to do: it influenced academic debate, regulatory decisions, and public perception. Monsanto employees even described it in internal documents as a source that would become their central authority on the safety of Roundup and glyphosate.

The article continues only 25 years later.

Despite the revelation of the shadow writing in 2017, the article was not immediately retracted from the scientific literature. It was not until late 2025, some 25 years after its publication, that the journal decided to retract it. The retraction was published in the February 2026 issue of the journal.

The withdrawal notice raised concerns regarding the attribution of the authors, conflicts of interest, partial selection of studies, and the validity of the conclusions presented.

Orseks said that when she and her colleague asked why the article hadn't been pulled earlier, the editor's answer was simple: Until they contacted him, no one had asked to check it.

She says the answer points to a systemic failure. Journals, scientific publishers and universities do not always act proactively even when evidence of problematic conduct is published. The researchers whose names appeared on the Monsanto paper were investigated by their institutions, but not punished. Of the 14 papers linked to Monsanto's shadow-writing of other research, only one went as far as that.

The EPA decision became a legal shield

Orsax linked the research's impact on regulation to a ruling issued by the United States Supreme Court on June 25, 2026, in the case Monsanto Company v. Durnell — A ruling that we reported on last week in Hidan.

John Dornell, who used Roundup for about 20 years and developed non-Hodgkin lymphoma, sued Monsanto, claiming that the company should have included a cancer warning on the product's label. A Missouri jury found in favor of the claim and awarded him more than $1 million.

The Supreme Court reversed the decision. It ruled that federal pesticide regulation law prevents states from requiring a manufacturer to add a warning to a label that differs from the one approved by the U.S. Environmental Protection Agency, the EPA. Because the EPA did not require a cancer warning on the Roundup label, the court ruled that Monsanto could not be held liable under state law for the lack of such a warning. Monsanto is now a subsidiary of Bayer.

The ruling did not itself determine whether glyphosate causes cancer, nor did it grant Monsanto complete immunity from any potential lawsuit. It dealt with a particular type of lawsuit—claims for failure to warn—and the question of whether a state law could require a warning that was not required at the federal level.

But for Orseks, the result illustrates the connection between science, regulation, and law: The EPA’s position that a cancer warning is not necessary has become a legal shield for the company. If the scientific literature and regulatory debate have been influenced for years by articles written or promoted by the company itself, their influence does not remain in the confines of the laboratory and the journal—it can shape policy, product labels, and court decisions.

A problem broader than one article

Orseks stressed that it is difficult to know how widespread shadow writing is because everyone involved is usually interested in hiding it. Most known cases have been uncovered not through scientific oversight mechanisms, but through lawsuits in which companies were forced to hand over internal correspondence and documents.

The available evidence indicates that shadow writing, financial conflicts of interest, funder control of data, and giving a company the right to review results before publication are not uncommon, especially in the fields of biomedicine and toxicology.

Orseks also mentioned the author of another Monsanto-related paper who published more than 1,100 scientific papers over his career—an average of about two papers a week. She said such figures require journals and research institutions to ask who actually wrote the papers, who paid for them, and who controlled the data.

“The scientific community must put its house in order”

Orseks' conclusion was not that science should not be trusted, but that such trust cannot be blind. It must be based on full disclosure of funding sources, transparency about each author's contribution, access to data, post-publication review, and a willingness to quickly retract articles when serious flaws are discovered.

Peer review before publication is not enough. When new evidence suggests shadow writing, bias, or hidden conflicts of interest, journals and universities should actively investigate rather than wait years for someone to file a formal complaint.

“We need to do more to put our house in order,” Orseks concluded, “so that the public can continue to trust the work we do.”

Questions and Answers

Has the public lost faith in science?
The data presented by Orseks does not indicate a general crisis of trust. A survey conducted in 68 countries found a medium-high level of trust in scientists, although there are differences between countries, political groups, and research fields.

What is scientific shadow writing?
This is a situation where a company, commercial entity, or individual who is not listed as an author participates in writing an article, while the article is presented as the independent work of other researchers.

Why was the article on glyphosate so important?
It has been cited more than a thousand times and has influenced the scientific literature, policy documents, Wikipedia, and the public debate on Roundup safety.

Did the Supreme Court rule that glyphosate is safe?
No. The court did not decide the scientific question. It held that federal law precluded a state lawsuit based on a requirement to add a cancer warning that the EPA was not required to add.

How can corporate interference in science be reduced?
Through full disclosure of funding and involvement in writing, publication of data, independent review, prompt handling of complaints, and holding authors, universities, journals, and publishers accountable.

More on the subject on the science website

• America in Test Tube 5: The Country That Built Silicon Valley – and the Entrepreneurs Who Prefer to Forget It
• An artificial intelligence system wrote scientific software that outperformed human code
• Nearly a billion shekels invested in bioconvergence: Israel maps 186 companies and establishes a new growth engine for high-tech
• Asimov's Laws Are Returning to the Discussion Table, But Military Artificial Intelligence Is Already Much More Complicated
• First study directly links metallic contamination in the atmosphere to rocket return from space