At the University of California, Davis, researchers are regularly invited to attend on-campus meet-and-greets with potential corporate funders to discuss possible sponsorship opportunities. Handshakes and business cards are routinely exchanged—so are nondisclosure agreements.

Jonathan Eisen, an evolutionary biologist at U.C. Davis, says such meetings and the attendant nondisclosure agreements are commonplace and that it’s university administrators—rather than the corporations themselves—who encourage their professors and researchers to attend. Eisen describes one meeting in which a company started out by passing around a document. “It was a 13-page agreement, and I refused to sign it,” Eisen says. “I said: ‘Look, there are 20 things in here I don’t understand and 15 things I completely disagree with. There’s no way I’m signing it.’”

But, unlike Eisen, many in the scientific community and academia do sign the NDAs—creating blind spots that make it impossible for the rest of the world to discern whether a corporation has had any undue influence on research. I spent a year poring over documents and talking to universities, companies, lawyers, and researchers to figure out what kind of role corporate funding plays in public-university studies across the United States. Nearly all of the people I spoke with talked about the increasing ease with which corporate representatives have access to researchers, although some were more comfortable with the arrangement than others.

Proponents of such arrangements—including all of the university officials I spoke with—say that corporate engagement in research is critical if universities are to continue their cutting-edge work. For many opponents, however, the mere mention that a corporation has sponsored research is enough to dismiss it as compromised. That’s because corporate backers can be given a great deal of power and latitude, selecting the specific kinds of studies, materials, and techniques to be used in exchange for their funding. Unsurprisingly, companies excel at creating the conditions most likely to give them the results they want. “It’s a problem, obviously,” says Ivan Oransky, a distinguished writer in residence at New York University’s Carter Journalism Institute, where he teaches medical journalism. “But if you tried to rid literature of every badly designed study, you’d be left with about four papers a year.”

But sometimes, a study is so controversial that even corporate backers cry foul. That’s what happened last December when a paper commissioned by the International Life Sciences Institute—composed of companies like Red Bull and Hershey among others—downplayed the importance of limiting one’s sugar intake. The university professors who had authored the paper quickly came under fire for conflicts of interest, and at least one has vowed to disclose all her funding from now on. Even Mars, one of the paper’s funders, slammed the study as based on weak evidence that only served to confuse consumers. According to the Associated Press, Mars said the paper “undermines the work of public-health officials and makes all industry-funded research look bad.”

Further complicating the issue is that there are no widely accepted or adopted industry guidelines for how these relationships should be conducted. Although the American Association of University Professors has issued recommendations for engagement, according to Laura Markwardt, a spokesperson for the association, “We don’t necessarily track which colleges and universities adopt which provisions.”* Among the sources I contacted—including representatives from the American Academy of Arts and Sciences, the American Association of State Colleges and Universities, the American Association of University Women, and others, as well as individual academics and officials at more than a dozen universities across the country—none knew of any universal standards governing such relationships, only the occasional rule or rules recommended by a given research department, university, or academic journal.

Defending science without universal engagement standards is difficult; but it is especially troubling given the contempt for science pervading government right now. For some scientists, it’s hard enough trying to convince many elected leaders that they should be informed by scientific research, rather than philosophical or religious beliefs, on politically charged issues, without having to also defend corporate-backed studies. Shortly after the presidential election last fall, thousands of scientists from around the country did call on the Trump administration to set high standards for integrity and transparency in supporting independent scientific research—and, importantly, to allow that science to guide state and federal policies. For many academics at public research universities, however, the question of who’s funding what, and how their research is used, is just politics as usual.

Academics and university officials, for example, are constantly under increased pressure from corporate funders to agree to conduct studies that would remain the property of the funder—meaning the researchers would be prohibited from publishing their own work. This allows a corporation to do its own vetting and to publish on its own time line—with the imprimatur of the university and without the peer review that any reputable academic journal would insist on. Plus, these lucrative work-for-hire arrangements often don’t count toward tenure reviews and can prevent academics from accepting better, more prestigious, work while they’re committed to the corporate projects. “We resist [work-for-hire agreements] because that’s not what we’re here to do,” says Thomas Coggins, who oversees sponsored awards and research compliance at the University of South Carolina. Still, resisting is not the same as banning. “Will we accept one of those? Yeah, we will, but everybody has to be formally educated about what this means.”

Meanwhile, the acute need for corporate funding at public universities is only growing—as is the opacity around whether funders are influencing research results. In an age of instant information, such deficiencies can undermine even the most valid studies and the most reputable scientists.

Unlike in the lab, when it comes to funding relationships, there is no scientific method.

* * *

At private universities, corporate donors and large endowments are part and parcel of day-to-day operations. But what about public universities, which are directly funded by taxpayers and which are subject to public-records laws? Is there an inherent conflict of mission between a public university and a private corporation? It’s hard to say—and that’s mostly because it’s impossible to get a clear handle on exactly how much corporate money is flowing into public-university research in the United States. That’s because much of that funding is simply not disclosed. Only one university, the University of Michigan, maintains a public database of how its research is funded. I sent public-records requests to more than a dozen public universities—many of them among the National Science Foundation’s top-10, tier-1 public research schools—for information on corporate-sponsored research. Only four of those requests yielded a full response. The rest netted documents with vital information redacted, were met with extended processing times of up to six months, required fees of several hundred dollars or more, or, in some cases, were ignored or just flat-out denied.

Research funding finds its way into universities in a number of ways. Sometimes the money goes into a school department dedicated to soliciting and working with corporate partners for research purposes. Other times, funds go into the university’s foundation. Donations to a university’s foundation are tax-deductible and, depending on the state, may or may not be subject to public-records laws. Research compiled by Alexa Capeloto, an assistant journalism professor at the John Jay College of Criminal Justice, found that 20 states have attempted to regulate public-university foundations through legislation or case law. Eleven states determined that university foundations serve as an extension of the university and are subject to public-records laws. Nine states exempted the foundations. Only eight states have passed clear laws on these foundations, and of those, three defined foundations as nonpublic entities. As a result, it can be near impossible to track the source of research dollars—however public the institution.

Often it is the companies themselves that provide the most information about educational outreach. The Dow Chemical Company website, for example, lists more than 30 universities it partners with “to advance scientific research and develop the world’s next generation of scientists and leaders.” Monsanto collaborates with universities in a number of ways, including through “peer-reviewed research in academic journals” and by providing “graduate-degree advisors and academic mentors.” And DuPont’s seeds division, Pioneer, sponsors symposia and workshops on university campuses throughout the United States.

Robert “Buck” Sanford Jr., a professor in the School of Earth Sciences and Environmental Sustainability at Northern Arizona University, finds the trend disconcerting. “You wake up one morning and you look around, and there are 10 people on campus that are getting research funding from Corporation X, Corporation Y, and Corporation Z, and some of it’s forest, some of it’s health, some of it’s food, some of it’s geology exploration,” he says. “It’s not at all regulated as far as I know. It’s really open-ended.”

But even though corporations may be more eager than their academic partners to announce their sponsored relationships, it doesn’t mean that universities are just passively accruing funding. In fact, universities are counting on that revenue and are actively courting industry dollars. One way they do so is through “industry-affiliates” programs, which—for a fixed fee—provide corporate partners with a wide range of benefits, like a guaranteed seat on an advisory board, early access to student résumés, private recruiting sessions, or full access to in-person meetings on campus. Industry-affiliates programs are usually organized by individual departments, even within the same university, and have varying levels of oversight and transparency.

Some schools openly advertise their programs. At Montana State University’s Center for Biofilm Engineering, for example, affiliates pay $24,000 a year to advise on research, specify deliverables, and have their products tested by university staff before going to market. “Advantages of directly funding project work at the [Center for Biofilm Engineering] include complete confidentiality and project direction by scientists and engineers at the top of biofilm investigation,” reads a brochure for the program. “We can offer confidential feedback on R&D direction, marketing ideas, or strategic decisions. Many of our members have found that this benefit alone is worth the annual membership fee.”

Virginia Tech has 30 industry-affiliates programs. One, for the Multifunctional Integrated Circuits and Systems Group, assures potential partners, “The University will, to the extent possible, take Members’ suggestions into account in selecting research topics, adopting research methodology, and directing research activities in an effort to maximize use of membership fees.” Affiliates of the university group can opt for a basic $10,000 membership level or for a $40,000 level, which allows for sponsorship of a graduate fellowship, during which the student will “explore a research area that is of interest to the sponsoring Principal Member, consistent with MICS’s strategic plans” and “collaborate on research/thesis directions.”

In the Department of Food Science at Purdue University, partners are invited to pay $5,000 a year for influence over “current and proposed curricula,” as well as an array of other benefits, from prepublication review to access to students and “potential consultants.” The offer is so attractive that on nearly every school day (and on some weekends) in September 2016 alone, one or more industry partners—like ConAgra, Nestle, Pepsi, and Dow Chemical—hit the Purdue campus to host an informational meeting, ice-cream social, or interview session.

Laurie Van Keppel, the career-services and external-relations coordinator at Purdue’s Department of Food Science, says, “Our department continually gets feedback from our industry partners about the skills that food-science graduates need to be successful professionals.” In return, she says, “Our faculty give short presentations at the industrial-associates meetings,” since industry partners want “to keep up to date with current research being conducted by our department.” Some professors even take a sabbatical to work on projects at a partner company; then they return the following semester to teach those case studies to their students.

Although schools create these prix-fixe research menus for their corporate members, Jeffrey Coriale, the director of external relations for the Institute for Systems Research at the University of Maryland, says the actual relationships are more à la carte. “Every partnership and partnership program is a little unique,” Coriale says, “and you have to structure your partnership to build that relationship so that it will provide the right benefits based on what you might want to do together.” That’s also why industry-affiliates programs don’t serve as disclosure clearinghouses for all the relationships between faculty and their corporate partners. Coriale says that faculty members develop individual consulting relationships with the corporate funders and are not always obliged to share the details with him or the university administration. Consequently, even the university may not be fully aware of a researcher’s involvement with a sponsor.

Eileen Buss, a former director of the industry-affiliates program in the entomology department at Purdue, says these programs are “controversial.” “It has an appearance, by paying a fee, of buying the researcher,” Buss says.

* * *

The U.S. Office of Research Integrity monitors allegations of research misconduct and fraud in publicly funded research. No such monitoring currently exists for corporate-funded research. That means corporate sponsors can often influence every part of the process—making calculated suggestions about everything from sampling size to specific survey questions.

“What tends to happen with industry-sponsored research is not fraud, it’s not misconduct, it’s nothing that would be retractable,” says NYU’s Oransky, who’s also the co-founder of Retraction Watch. “It’s that they pick questions, and they pick studies, and they pick study designs that are likely to give them the answer they want.” This is exactly why scholars like U.C. Davis’s Eisen, the professor who refuses to sign NDAs, argue that disclosing corporate involvement in research is critical.

Last year, a lack of disclosure derailed a multimillion-dollar industry-led initiative. The scientist-run Global Energy Balance Network was created to study the role of physical activity—as opposed to caloric intake—in obesity. But when the network’s results brushed aside a link between soft drinks and obesity, public-health officials were incredulous. Academics in the field looked into the matter and ultimately unmasked the network’s funder: Coca-Cola. Critics immediately called the initiative one-sided and claimed Coca-Cola had spent millions to hire academics throughout the country to conduct studies favorable to Coke’s products—without disclosing the corporation’s role. Even as scientists at the Global Energy Balance Network denied intervention from their parent company, an investigation by the Associated Press found that Coca-Cola had helped select the initiative’s leaders, drafted its mission statement, and designed its website. What’s more, the A.P. uncovered emails from Coca-Cola officials comparing their scientific project to a “political campaign.”

Coca-Cola has given millions to fund one researcher’s work in particular: Dr. Steven N. Blair, an obesity-research expert at the University of South Carolina who focuses primarily on exercise. “Most of the focus in the popular media and in the scientific press is, ‘Oh they’re eating too much, eating too much, eating too much’—blaming fast food, blaming sugary drinks, and so on,” Blair said in a promotional video for the Global Energy Balance Network. “And there’s really virtually no compelling evidence that that, in fact, is the cause.” The nutrition department at Harvard University disagreed and along with 36 co-signing scientists denounced the findings as “scientific nonsense.”

Blair insists Coca-Cola didn’t control his research—though he acknowledges that he may have taken their advice when crafting the study. “I don’t remember any of the details,” Blair says, “but I’m sure that in discussions with scientists—Ph.D.-level scientists at Coke, discussing the project—I’m sure occasionally we might have said: ‘Oh, that’s a good idea. We’ll put that measurement in, or we’ll do this, we’ll do that.’ So it’s not that we totally ignored them, but we decided what we wanted to do.” Despite this patina of compromise, Blair lauds Coca-Cola for funding a 430-participant exercise study he conducted—one of the most extensive of its kind.

According to a representative for Coca-Cola, the company “is committed to transparency” and regularly updates their website with their “funding of well-being-related scientific research, programs, and individuals.” But it was the iconic company’s lack of transparency that finally caught up with its Global Energy Balance Network: The group disbanded amid controversy just months after it launched.

Meanwhile, like thousands of other researchers across the world, Blair continues to welcome funding from all sources as he has for decades. “Don’t ask me to list all the corporations that I’ve received some consulting fee or honorarium from,” he says with a chuckle. “It would probably take me days to go back through mountains of computer files to find them.”

* * *

That lack of transparency about who’s funding what universities extends to off-campus engagements, too. Researchers regularly promote their work through interviews or speaking engagements, are panelists at conferences, endorse products, or even testify before Congress.

When Dr. Larry R. Faulkner, president emeritus of the University of Texas at Austin, testified at a hearing before the U.S. House Committee on Science, Space and Technology in September, he argued that the “continuing expansion of federal regulations and requirements is diminishing the effectiveness of the U.S. research enterprise.” But he did not—and he was not required to—disclose his seat on the board of governors of ExxonMobile, a company that would likely benefit from fewer government regulations on academic research. That’s because when faculty members testify before Congress, witness-disclosure forms require that public funding be disclosed—but corporate funding is exempt.

In an ideal world, public colleges would be able to thrive entirely on public dollars and modest tuition. But in the real world, higher education is getting less and less public funding. A recent report from the Center on Budget and Policy Priorities found that the average state is spending $1,598, or 18 percent, less per student at public two- and four-year institutions than it was before the 2008 recession. Overall, after adjusting for inflation, funding is almost $10 billion below what it was just prior to the recession. Not only is there less money, there are also far more colleges than there were a decade ago, according to the National Center for Education Statistics.

Even as the economy recovers, public funding isn’t necessarily returning to higher education. In California, for example, implementing the Affordable Care Act meant rerouting funds from public universities to state health and human services, and more money was diverted to criminal-justice groups after the laws governing the state’s corrections facilities changed. Because funding at public universities is discretionary, long-term planning can be extremely difficult. “Higher-ed is one of those things that’s an easy place to go if you need to make quick and sizeable reductions in your state budget,” says Ryan Storm, the assistant vice chancellor of budget at California State University. “Ten or 15 years ago, the mind-set was that the state of California was the principal investor of the university system. That’s no longer the case.”

The University of Alaska system was recently hit hard after plummeting oil prices affected state tax revenue. To keep the lights on, the campuses are getting creative. “The price of oil is going to rebound somewhat, but we’re never going to go back to the level of state funding we had in the past,” says Larry Hinzman, the vice chancellor of research at the University of Alaska, Fairbanks. “So we are looking to diversify our revenue sources.”

A lot of schools are. In many cases, state funding is earmarked to support university infrastructure, like lights, classrooms, and general maintenance, but there isn’t funding set aside for much else, like new lab equipment, sports teams, career services, and innovation hubs. That’s where corporate relationships prove vital—as noted by the names of many campus buildings or job-fair sponsors. Public records obtained through a Freedom of Information Act request show that at Iowa State University, for example, donations from Dow, Syngenta, and Monsanto helped fund the athletics department. At the University of Michigan, a consortium of automotive manufacturers, telecommunications companies, economic-development groups, and others have formed the Mobility Transformation Center, where they aim to build a smart grid of connected and automated vehicles. In a partnership with the University of Washington, Amazon launched Amazon Catalyst, which provides grant funding to any member of the university community, in any discipline, with an idea to tackle a pressing societal issue.

Often, these endeavors can be great academic experiences for university students. After all, research being conducted at universities at any given time could end up influencing nearly every aspect of the average consumer’s life—from the nutrients in food to car-safety mechanisms to life-saving drugs. Research leads to product development and policy decisions; it finds its way into medicine cabinets and hospital rooms and onto dinner tables and digital tablets. And when conducted at a university, research has the air of impartiality—something many corporations use to their advantage.

“Because public-education budgets are increasingly reliant on large private donations, the competition ends up being fierce for them,” says Michael Halpern, the deputy director for the Center for Science and Democracy with the Union of Concerned Scientists. “And that can be a significant threat to both academic freedom and public faith in the independence of higher education.”

* * *

In 2015, a group of researchers, including Dr. Stephen Smith of Iowa State University, argued in a paper that increased intellectual-property protections for genetically modified crops benefit farmers and consumers alike. An industry trade group, the American Seed Trade Association, funded the project, and Smith has served as chair of that group’s intellectual-property committee. Last spring, Smith published a second version of the paper, not as a result of duplicating the study, but to make the findings more accessible to plant breeders and policy-makers. This reissuance was timed to publish just as the first wave of patents on genetically modified organisms were expiring and new intellectual properties were entering the open market. The patent changes meant that seed dealers could now breed varietals based on previously protected intellectual property—and it would cost less and come with no restrictions on replanting. But Smith’s paper made the case that it’s in everyone’s best interest to continue planting crops with protected seeds.

Smith says that intellectual-property protections ensure that farmers don’t purchase seeds that won’t grow or buy already-harvested seeds, which produce a lower yield. “And that’s important to the farmer,” he says, as creating high yields is a critical part of agricultural development. “But the main benefit, the main beneficiaries of all of this, is actually you and me, as consumers. We are the end of the story, the end of the line in terms of getting the most benefits from improved quality and quantity of food production.” This statement—that protecting the intellectual property of seeds benefits farmers and consumers—wouldn’t be out of place on, say, the website of DuPont Pioneer, DuPont’s global seed company. That’s because, while encouraging farmers to continue using seeds with intellectual-property protections may or may not benefit both farmers and consumers, it definitely benefits companies like DuPont. Indeed, large agricultural companies like DuPont are arguably the biggest beneficiaries: DuPont owns thousands of patents and together with Monsanto and Syngenta controls more than half of the global market for seeds.

DuPont Pioneer is also where Smith spent 35 years of his career—developing molecular-marker technology, which is essentially DNA fingerprinting for seeds and goes to the core of intellectual-property protections for genetically modified crops. In fact, Smith was still at Pioneer when he began working on the American Seed Trade Association studies. In an email, Smith noted: “This rewrite for a broader audience was initiated and written by me when I was employed at DuPont Pioneer and completed by me after I retired from DuPont Pioneer ... on my own time.” But Smith still serves as a paid consultant for DuPont. And though his two papers acknowledge the American Seed Trade Association funding and list Smith’s affiliation with Iowa State University, they do not disclose his decades at DuPont. (DuPont did not return my request for comment.)

To be clear, there is absolutely no indication that Smith’s study isn’t valid or that he didn’t follow the letter of the rules for disclosure within the university, his department, or the journal that published his paper, Crop Science. The problem is that without universal guidelines requiring such transparency, consumers don’t actually have all the information they need. When a career-corporate scientist conducts a study, disclosing that career-corporate relationship seems particularly relevant. Many experts, like the Union of Concerned Scientists, argue that such affiliations are as critical a data point when assessing a study as sample size and that more transparency is preferable to less.

U.C. Davis’s Eisen is a proponent of the open-science movement, which advocates for access to all scientific research and data for anyone who wants it. The idea is that by making all research, methodologies, and results available to the public, everyone can participate in the scientific discussions affecting society and build upon each other’s work. It also makes studies easier to replicate and verify. The movement hinges on transparency—and disclosing funding is a fundamental component of that.

“It’s not that I don’t trust people,” Eisen says. “It’s not that I want to believe the worst in them, but I can easily imagine someone who got money from two different sources—one from the [National Institutes of Health] and one from a private enterprise—for 35 years, and amazingly, they never critiqued the private enterprise.” That doesn’t mean that anything was untoward, Eisen says, but “what’s the harm in disclosing that?”

In the absence of an institutional structure and support for disclosure, corporations are free to “game the system,” as the Union of Concerned Scientists put it. Researchers may be bound by the disclosure and conflict-of-interest policies of their universities or journals, but those policies can vary so widely, contain so many omissions, or be so riddled with caveats that loopholes remain—not to mention the labyrinthine matrices that any reader looking for the entire scope of funding would have to navigate. This “patchwork of disclosure requirements ... leaves scientists completely unprepared for the scrutiny that they’re going to face,” says Halpern, of the Union of Concerned Scientists. “Their institutions give them a minimum set of disclosure standards, and many scientists think that’s enough. And my view is that those kinds of guidelines can give scientists just enough rope to hang themselves.” That’s why the burden is currently on researchers to self-disclose. Eisen, for one, regularly publishes disclosures on his blog, listing everything from the corporate funding he has accepted to the stocks he and his wife own.

That’s something more scientists should probably be doing, because bias is very often completely unconscious. Few researchers will admit to being biased in their research, says Naomi Oreskes, a Harvard University researcher and the author of Merchants of Doubt, and they may not even be aware that their research is being influenced. Oreskes and her team called for the disclosure of all research funding, writing in Environmental Science and Technology, “Even if we think of ourselves as honest, objective, and independent, scientific evidence demonstrates that our research can be influenced by the sources of our funding.”

This unconscious bias expresses itself in any number of choices researchers make throughout the design, execution, and interpretation of their work. “One reason scientists may succumb to unconscious bias is that we think of ourselves as less susceptible to these effects than we really are,” writes Oreskes. Bias is usually subtle and yet “many researchers have a narrow conception of research integrity, restricting it in their minds to avoiding egregious misconduct such as fraud, fabrication, and plagiarism.”

And though any bias is hopefully caught during a peer review, that’s not always the case. Besides, in an age of instant communication, the idea that disclosure for funding could wait until a study has been completed is antiquated. The open-science movement argues that there are multiple opportunities throughout the process—from the first proposal to the final publication—for funding to be disclosed clearly and repeatedly. “People expect information about conflicts of interest and funding to be public and be immediately at their fingertips,” Halpern says. “It’s not really sufficient any longer for researchers, and specifically those who work in controversial fields, to disclose conflicts of interest when asked or simply say that they’ve given the information over to their university and that’s enough.”

* * *

In 1995, when the late nuclear physicist Joseph Rotblat accepted the Nobel Prize, he urged the scientific community to “remember your humanity.” Rotblat was referring to the destructive science behind nuclear weapons, but in that speech, he also called for the standardization of ethics in science: “The time has come to formulate guidelines for the ethical conduct of scientists, perhaps in the form of a voluntary Hippocratic Oath.”

“You are doing fundamental work, pushing forward the frontiers of knowledge, but often you do it without giving much thought to the impact of your work on society,” Rotblat said. “Precepts such as ‘science is neutral’ or ‘science has nothing to do with politics’ still prevail. They are remnants of the ivory-tower mentality.”

Amid the growing presence of research funded by private enterprise, it is time for the ivory tower to shed those remnants. Many researchers think they are working in the best interests of society, but without full disclosure, it is impossible to know. In their desire to snag funding, what compromises are universities making?

* This article originally referred to the AAUP as the American Association of University Professionals. We regret the error.

The McGraw Center for Business Journalism at the CUNY Graduate School of Journalism provided partial funding for this investigation.