Imagine a billionaire with an abiding interest in science, but also in having sex with young girls. He’s famous, our billionaire, and he associates with what used to be called boldface names, some of whom know—they’d have to, right?—about his habits. But they let it go. And eventually the billionaire’s name gets associated with all sorts of do-gooderish public endeavors, before the truth gets told—mostly after the billionaire’s death under salacious and suspicious circumstances.

To be clear—not that billionaire. All the things I just said were true about Jeffrey Epstein, the financier, convicted sex trafficker, and accused child rapist whose decades of personal and financial connections to entire TED conferences’ worth of marquee scientists and intellectuals have at last begun to have consequences for his enablers (but not Epstein, who died in jail in early August). I was actually talking about Howard Hughes—aviator, film producer, mogul, creep … and vitally important philanthropist. The Howard Hughes Medical Institute has $20 billion in assets today and claims more than 2,000 employees; in 2018 alone it awarded $562 million in competitive, highly sought-after grants to fund biomedical research. In the middle of the 20th century, Hughes was a predator; today the institute with his name on the door is one of the most prestigious and robust supporters of lifesaving scientific innovation on earth. No scientist would dream of saying no to a grant from HHMI.

Association with Jeffrey Epstein, on the other hand, now turns research institutions into reputational superfund sites. Money corrupts—which, duh—but the Epstein episode tells an even bigger story. The entire system for metabolizing philanthropic gifts, particularly private ones, into academic research is a poorly illuminated pile of broken guardrails. Even if most institutions and foundations are cautious internally, even if the unfolding scandal with Epstein and the Massachusetts Institute of Technology is an outlier, the system is essentially a pool of dark money. Its sources and goals are often unclear, or occult. But it’s money that research institutions need—or, at least, want. Stipulated, most donors want to help the world. Some also want to build a legacy. Most institutions want the same. But those desires are threaded through an ethical minefield.

In mid-August, the director of MIT's Media Lab (and former WIRED contributor) Joichi Ito apologized for taking donations from Epstein to support both the Lab and his personal investments. Two researchers affiliated with the Lab quit, and last week an article in The New Yorker revealed that Ito’s financial connections ran deeper than he had acknowledged—more money had changed hands, Ito had spearheaded attempts to keep Epstein’s involvement secret, and Lab researchers feared the headquarters of the $75 million-a-year think tank dedicated to making people’s lives “safer, cleaner, healthier, fairer, and more productive” had hosted meetings to which Epstein had brought women against their will. Ito resigned the next day.

A memo this week from MIT president L. Rafael Reif previewed some of the findings of an investigation into all that. They’re damning, and they include a letter acknowledging a gift from Epstein to MIT physicist Seth Lloyd four years after Epstein’s 2008 conviction. “I apparently signed this letter on August 16, 2012, about six weeks into my presidency. Although I do not recall it, it does bear my signature,” Reif wrote. Not only did senior administrators know about multiple donations from Epstein, but “information shared with us last night also indicates that Epstein gifts were discussed at at least one of MIT’s regular senior team meetings, and I was present.”

Late Thursday, while a Democratic presidential debate was absorbing most of the available attentional oxygen, Harvard president Lawrence Bacow released a letter acknowledging the $6.5 million Epstein gave to Harvard to start the Program for Evolutionary Dynamics, as well as a fellowship Epstein had at Harvard and $2.4 million in “other gifts” given before Epstein’s conviction. The university will donate the $186,000 remaining of those funds to groups that help human trafficking and sexual assault victims. The rest of the money, Bacow wrote, has been spent. “Epstein’s behavior, not just at Harvard, but elsewhere, raises significant questions about how institutions like ours review and vet potential donors,” Bacow wrote. As apologies go, it was a little mea culpa, but also a little sua culpa.

Depending on how you feel about billionaire plutocrats, their money is always tainted. The Media Lab’s building is just a few minutes’ walk from the Koch Institute for Integrative Cancer Research and the Koch Biology Building, named—as MIT science writing professor Seth Mnookin pointed out in Stat earlier this week—for the recently deceased David Koch. With his brother Christopher, Koch funded the political denial of climate change. Immoral? Sure. Like Thanos-scale immoral. But it’s not illegal. Koch was an MIT alumnus and, by some accounts, a smart and decent guy, if you can put aside the planet destruction and the LGBTQ rights obstruction. Which, don’t.

Sticking to sticking it to MIT, what about the $350 million donation that put Stephen Schwarzman’s name on a college of computing there? Schwarzman, head of a private equity firm called the Blackstone Group, gave $150 million to Yale in 2015, but faced opposition at MIT because of his association with President Trump and Blackstone’s ties to Saudi royal family, especially after they were involved with the murder of the journalist Jamal Khashoggi. Nothing illegal about any of that, right?

The minefields don’t end at the banks of the Charles River, either. If you’re a museum-goer, you’ve probably walked through a Sackler wing or two; this year the Sackler family came under (deserved) fire for their role in the epidemic of opioid addiction, fostered in part by their company, Purdue Pharma. London’s Tate museum and National Portrait Gallery both said they’d stop taking Sackler money, and then, under pressure, so did the Guggenheim in New York City. Nice, except the Guggenheim family made its money with environmentally vile lead and copper mines in the early 20th century, and then shifted to Chilean nitrate mines for fertilizer and explosives. Enjoy the museum!

Or take the Michael Milken Foundation. Milken helped invent junk bonds and leveraged buyouts in the 1980s, went to prison for almost two years for securities and tax fraud, and is now mostly known for charitable giving to help cure cancer. “It’s almost a point of pride today to have a grant from the Milken Foundation,” says Gene Tempel, founding dean emeritus of the Lilly Family School of Philanthropy at IUPUI. “And yeah, he was convicted. Very clear.” (Er … that school would be funded by money the Lilly family made through the Eli Lilly pharmaceutical firm, inventor of the antidepressant drug Prozac.)

American history is shot through with these kind of stories. This country wouldn’t have Carnegie libraries, the Howard Hughes Medical Institute, the Ford Foundation, or the Rockefeller Foundation if very rich men didn’t want to burnish their legacies, which is what people used to call reputation laundering. At least one of those guys was a vicious anti-Semite and another one was a sex predator. Yet today getting a Ford Foundation or HHMI grant is a sign that your work is a high-value contribution to American arts and letters.

Donations that come from sources with potential legal or ethical issues are called, in the language of this world, “tainted.” Different places have different rules for handling such funds, both among and within institutions. Turning down money from someone with legal issues, that’s easy—or at least it’s supposed to be, Jeffrey Epstein notwithstanding. And gifts with explicit quos attached to the quid? Those are easy to say no to, too. “Most institutions will have policies to make certain that research is not tainted by outside influence. There’s a pretty high standard in academic research so that findings are trustworthy,” Tempel says. But that’s not even the biggest problem: “If you’re taking money and associate the institution with someone whose value system might not be in alignment with the values of the institution, then its an assault on institutional integrity.”

The countervailing pressure against all that philosophical hand-wringing is, like, money. The idea that government should pay for scientific research isn’t even a century old in the US—it’s a post–World War II innovation. The 2018 federal research and development budget was $176.8 billion, which is even more than Jeff Bezos is worth, if you can believe it. But even that amount doesn’t keep up with the number of new scientists entering the field, looking for funding. The pressures of “soft money,” of jobs funded entirely by grants, are significant—a person funded that way loses their job if they don’t do fund-raising. “The rubber hits the road with these parts of universities trying to stay afloat with whatever funding they can get, and they start sacrificing their moral druthers,” says Aaron Horvath, a PhD candidate in sociology and a fellow at Stanford’s Center on Philanthropy and Civil Society (which has, it may give you some giggles to learn, three Marc and Laura Andreessen Faculty Codirectors—jobs endowed by the partner at famed Silicon Valley venture capital firm Andreessen Horowitz and his wife, who has her own family billions and is herself the founder and chair of the center).

Bad money might not necessarily lead to bad science. A smart donor-recipient team can ensure that the donor doesn’t have any say in methodology. Lots of researchers might never even have contact with the person who contributes a name to the building they work in. But for better or worse, big-ticket donors redirect research emphasis. They bend the arc of the innovation universe. The really massive influxes of money from superdreadnought givers like the Bill and Melinda Gates Foundation or the Chan Zuckerberg Initiative unbalanced entire fields of research (emerging disease in the former case and education in the latter).

It’s true, too, that one strain of the philosophy of science says that, by definition, an unethical research environment can never produce good results. If you don’t get informed consent from human subjects, your data is garbage. If you don’t take good care of your research animals, your data is garbage. If you harass and abuse your students, your data is … well, no one’s exactly sure what to do with science published by monstrous men. What to do about science produced with monstrous men’s money is a harder question. People working in those Koch buildings at MIT are going to do great work. Maybe taking the money was the salient unethical act—but the university did that, not the researchers.

In fact, one argument goes that you’re making the world worse by not taking the money, wherever it comes from. “What if you can make some lifesaving discovery, or provide scholarships? It might be helping the institution with faculty members for a new program,” Tempel says. “Whatever it is, that’s not going to happen if that money is turned down.” In this purely utilitarian view, even tainted money can maximize overall good. Take Koch money and use it to fight climate change!

Now, that argument runs aground against the Media Lab’s almost unsullied reputation for never actually making anything. It certainly falls apart as soon as someone tries to apply a metric. What counts as good, and how does it spread? Maybe I’ll leave that, too, for philosophers. For institutions that pay taxes and have to answer questions from reporters, rule utilitarianism provides an easier way out. Or deontology—just do the right thing, for heaven’s sake. MIT has acknowledged violating its own policies, to the extent anyone understands what it means for a potential donor to be on a “disqualified” list.

Even if an institution has clear rules and a team of people who understand its values and can apply them to would-be donors (or can afford to hire a high-priced consultant who specializes in such matters), that doesn’t stop individual labs or researchers from freelancing. Rich patients at elite hospitals sometimes give their doctors big checks directly, for example. That’s nice for a physician-researcher’s career, but it can be a real headache for their institution. Those individual donations rarely come with stipulations for indirect costs like facilities or support staff, which means the researcher has money for a lab but the institution has to pay for all the things that make a lab function.

You can understand why the donors do it, of course. Yes, sure, they want to advance some field of science that they care about. But they also want to control tomorrow, to chisel the story of their lives into plaques and cornerstones before some historian can overwrite them with an uglier truth. And it works. How do I know? Before I started working on this story, I’d not only forgotten all that nasty stuff about Howard Hughes, I’d all but forgotten that his name was what the “HH” in HHMI stood for. Hughes didn’t buy societal amnesia; he bought prosopagnosia. We didn’t forget his name, but we forgot his true face.

None of those challenges excuse taking money from a child rapist and covering it up. Nobody who let that happen should still have a job tomorrow. But the system that lets this kind of thing happen needs fixing, too. Clearly some equation must link the number of zeros on a check to the number of years it takes to untaint the dollars. But unless the requirements for disclosure and data collection change, no one will ever be able to work out the constants in that equation. “As far as aggregate data goes, it’s kind of shit. It really is pretty bad. There isn’t a good system for me to go find all cases where person A gives to organization B,” Horvath says.

More transparency on who’s donating what would help, but all the data in the world doesn’t matter if no one’s looking for violations and enforcing the rules. “Any project I do, I have to run through an institutional review board for research ethics. But my funding doesn’t,” Horvath says. “It’s interesting to wonder what that would look like.” But then, what donor would ever want to fund such a program?

Have a tip about science philanthropy? Contact the author at Adam_rogers@wired.com or via Signal at 415-244-4557.

Updated 9-16-19, 5:30 pm EDT: This story has been updated to correct the spelling of Jamal Khashoggi’s name and to correct descriptions of Stephen Schwarzman’s donations to Yale and MIT.

• Concerning consent—and canceling cancel culture
• Watch how to solve a Rubik's cube, step by step
• Why the Porsche Taycan's two-speed gearbox is such a big deal
• Elon Musk: Humanity is a kind of “biological boot loader” for AI
• The ethics of hiding your data from the machines
• 👁 How do machines learn? Plus, read the latest news on artificial intelligence
• 📱 Torn between the latest phones? Never fear—check out our iPhone buying guide and favorite Android phones