This summer, a pill intended to treat a chronic, incurable lung disease entered mid-phase human trials. Previous studies have demonstrated that the drug is safe to swallow, although whether it will improve symptoms of the painful fibrosis that it targets remains unknown; this is what the current trial will determine, perhaps by next year. Such a tentative advance would hardly be newsworthy, except for a wrinkle in the medicine’s genesis: It is likely the first drug fully designed by artificial intelligence to come this far in the development pipeline.

The pill’s maker, the biotech company Insilico Medicine, used hundreds of AI models to discover both a new target in the body that could treat the fibrosis and which molecules might be synthesized for the drug itself. Those programs allowed Insilico to go from scratch to putting this drug through the first phase of human trials in two and a half years, rather than the typical five or so. Even if the pill proves useless, a real possibility, plenty of other drugs designed with the help of AI are in the wings. Scientists and companies alike hope that these will reach pharmacies far faster than traditionally designed medicine—bringing a drug to market typically takes well over a decade, and failure rates are high.

Medicine is just one aspect of a broader transformation in science. In only the past few months, AI has appeared to predict tropical storms with similar accuracy and much more speed than conventional models; Meta has released a model that can analyze brain scans to reproduce what a person is looking at; Google recently used AI to propose millions of new materials that could enhance supercomputers, electric vehicles, and more. Just as the technology has blurred the line between human-created and computer-generated text and images—upending how people work, learn, and socialize—AI tools are accelerating and refashioning some of the basic elements of science. “We can really make discoveries that would not be possible without the use of AI,” Marinka Zitnik, a biomedical and AI researcher at Harvard, told me.

Science has never been faster than it is today. But the introduction of AI is also, in some ways, making science less human. For centuries, knowledge of the world has been rooted in observing and explaining it. Many of today’s AI models twist this endeavor, providing answers without justifications and leading scientists to study their own algorithms as much as they study nature. In doing so, AI may be challenging the very nature of discovery.

AI exists to derive impossibly intricate patterns from data sets that are too large for any person to fathom, a mystifying phenomenon that has grown more familiar since ChatGPT was released last year. The chatbot—a tool suddenly at everyone’s fingertips that appears to synthesize the entire internet—changed how we can access and apply knowledge, but it simultaneously tainted much of our thinking with doubt. We do not understand exactly how generative-AI chatbots determine their responses, only that they sound remarkably human, making it hard to parse what is real, logical, or trustworthy, and whether writing, even our own, is fully human or bears a silicon touch. When a response does make sense, it can seem to offer a shortcut rather than any true understanding of how or why the answer came to be.

AI may be doing something similar in a broad range of scientific disciplines. Among the most notable scientific advances achieved via AI may be those in molecular biology from DeepMind, a leading AI research lab now based at Google. After DeepMind’s programs conquered the game of Go in 2016—a game so much more complex than chess that many thought that computers could never master it—Demis Hassabis, DeepMind’s CEO, told me that he began considering how to build an AI program for the decades-old challenge of protein folding. All sorts of biological processes depend on proteins, and every protein is made of a sequence of amino acids. How those molecules fold into a three-dimensional shape determines a protein’s function, and mapping those structures could help scientists develop new vaccines, kill antibiotics-resistant bacteria, and explore new cancer treatments. Without a protein’s 3-D shape, scientists have little more than a bunch of Lego bricks without instructions for putting them together.

Figuring out a single protein structure from a sequence of amino acids used to take years. But in 2022, DeepMind’s flagship scientific model, AlphaFold, found the most likely structure of almost every protein known to science—some 200 million of them. Much like the company’s chess- and Go-playing programs, which search for the best possible move, AlphaFold searches the number of possible structures for an amino-acid sequence to find the most probable one. The program compresses what could have been an entire Ph.D.’s worth of work into seconds, and it has been widely lauded for its “revolutionary impact” on basic biology and the development of novel treatments alike. Still, independent researchers have noted that despite inhuman speed, the model does not fully explain why a specific structure is likely. As a result, scientists are trying to demystify AlphaFold’s predictions, and Hassabis noted that those efforts are making good progress.

[Read: Welcome to the big blur]

AI allows researchers to study complex systems in “the world of bits” at a much faster pace than in the “world of atoms,” Hassabis said, and then physically test their hypotheses as a final step. The technology is pushing forward advances in numerous other disciplines—not just improving speed and scale, but changing what kind of research is thought possible. Neuroscientists at Meta and elsewhere, for instance, are turning artificial neural networks trained to “see” photographs or “read” text into hypotheses for how the brain processes both images and language. Biologists are using AI trained on genetic data to study rare diseases, improve immunotherapies, and better understand SARS-CoV-2 variants of concern. “Now we have viable hypotheses, where before we had mysteries,” Jim DiCarlo, a cognitive scientist at MIT who has pioneered the use of AI to study vision in the brain, told me.

Astronomers and physicists are using machine learning to process data sets from the universe that were too immense to touch before, Brice Ménard, an astrophysicist at Johns Hopkins, told me. Some experiments, such as the CERN particle collider, produce too much information to physically store. Researchers rely on AI to throw out familiar observations while keeping unknowns for analysis. “We don’t know what the needle looks like, because these are undetected physics events, but we know what the hay looks like,” Alexander Szalay, the director of the Institute for Data Intensive Science at Johns Hopkins, told me. “So computers are trained to recognize the hay and basically throw it away.”

[Read: Computers are learning to smell]

The long-term vision could even involve combining AI models and physical experiments in a sort of “self-driving lab,” Zitnik said, wherein computer programs and robots generate hypotheses, plan experiments to test them, and analyze the results. Such labs are a ways off, although prototypes do exist, such as the Scientific Autonomous Reasoning Agent, a robotic system that has already discovered new materials for renewable energy. SARA uses a laser to analyze and alter materials iteratively, with each loop lasting a few seconds, Carla Gomes, a computer scientist at Cornell, told me—reducing days of research to hours. This future, if it comes to pass, will elevate software and robots from tools to collaborators, even co-creators of knowledge.

Quantum observations too numerous for humans to store, experiments too rapid for humans to run, neuroscientific hypotheses too complex for humans to derive—even as AI enables scientific work never before thought possible, those same tools pose an epistemic dilemma. They will produce groundbreaking knowledge while breaking apart what it means to know in the first place.

“The holy grail of science is understanding,” Zitnik said. “To be able to understand a phenomenon, whether that’s the behavior of a cell or a planetary system, requires being able to identify causes and effects.” But AI models are famously opaque. They detect patterns based on gargantuan data sets via software architectures whose inner workings baffle human intuition and reasoning. Experts have taken to calling them “black boxes.”

This presents obvious problems for the scientific method. “We have to understand what is going on inside this black box so we can see where this discovery is coming from,” Szalay told me. To predict events without understanding why those predictions are accurate might gesture toward a different type of science, in which knowledge and the resulting actions are not always accompanied by an explanation. An AI model might predict a thunderstorm’s arrival but struggle to explain the underlying physics and atmospheric changes that triggered it, analyze an X-ray without showing how it arrived at its diagnosis, or propose abstract mathematical conjectures without proving them. Such shifts from observations and grounded reasoning to mathematical probability have happened in the sciences before: The equations of quantum mechanics, which emerged in the 20th century, accurately predict subatomic phenomena that physicists still don’t fully understand—leading Albert Einstein himself to doubt quantum theory.

[From the January 1951 issue: Faith in science]

Science itself may offer a solution to this conundrum. Physical experiments have uncovered a great deal in the past century about the quantum world, and similarly, AI tools might appear inscrutable partly because researchers haven’t spent enough time probing them. “You have to build the artifact first before you can pull it apart and scientifically analyze it,” Hassabis told me, and scientists have only recently begun to build AI models worthy of study. Even older numerical simulations, although far less complex than today’s AI models, are hard to interpret in an intuitive way, but they have nevertheless informed new discoveries for decades.

If researchers understand how artificial neurons respond to an image, they might be able to translate those predictions to biological neurons; if researchers understand which parts of an AI model link a mutation to a disease, scientists could gain new insights into the human genome. Such models are “fully observable systems. You can measure all the parts,” DiCarlo said. Whereas he cannot measure every neuron and synapse in a monkey brain during surgery, he can do that for an AI model. With the right access, AI programs might present scientists not with black boxes so much as with a new type of object requiring a new sort of inquiry—not “models” of the natural world so much as addendums to it. Some scientists even hope to build “digital twins” to simulate cells, organs, and the planet.

AI is not a silver bullet, though. AlphaFold may be revolutionary, and perhaps Insilico will indeed drastically reduce the time it takes to develop new medicine. But the technology has significant limitations. For instance, AI models need to train on large amounts of relevant data. AlphaFold is a “spectacular success,” Jennifer Listgarten, a computational biologist and computer scientist at UC Berkeley, told me—but it also “relied on very expensive, highly curated data that was generated over decades in the laboratory on a very crisply defined problem that could be evaluated extremely cleanly.” The lack of high-quality data in other disciplines can prevent or limit the use of AI.

Even with those data, the real world can be more complex and dynamic than a silicon simulation. Translating the static structure of a molecule into its interactions with various systems in the body, for instance, is a problem that researchers are still working on, Andreas Bender, who studies molecular informatics at the University of Cambridge, told me. AI can propose new medicines quickly, but “you still need to run the drug-discovery process, which is, of course, quite long,” John Jumper, a researcher at DeepMind who led the development of AlphaFold, told me.

Clinical trials take years and many are unsuccessful; plenty of AI drug start-ups and initiatives have scaled back. Those failures are, in some sense, evidence of science working. Experimental results, along with known physical laws, allow scientists to prevent their models from hallucinating, Anima Anandkumar, a computer scientist at Caltech, told me. No analogous laws of linguistic accuracy exist for chatbots—consumers have to trust Big Tech.

In a lab, novel predictions can be physically and safely tested in an isolated setting. But when developing drugs or treating patients, the stakes are much higher. Existing maps of the human genome, for instance, skew toward white Europeans, but the expression of many conditions, such as diabetes, varies significantly by race and ethnicity. Just as biased data sets produce racist chatbots, skewed biological data might mean that “models are not applicable to people of non-caucasian origin,” Bender told me, or those of different ages, or with existing diseases and on co-medications. A cancer-diagnosis program or treatment designed by AI might be especially effective only on a small slice of the population.

AI models might transform not just how we understand the world but how we understand understanding itself. If so, we must build out new models of knowledge as well—what we can trust, why, and when. Otherwise, our reliance on a chatbot, drug-discovery tool, or AI hurricane forecast might depart from the realm of science. It might be more akin to faith.

Maresuke Nogi was always his own toughest critic. Emperor Meiji trusted him and appointed him to high military posts in Japan: general in the imperial army, governor-general of Taiwan. But we all make mistakes, and Nogi’s lapses gnawed at him. Twice he requested the emperor’s leave to commit ritual suicide. Each time, the emperor refused. In Nogi’s home, now a quiet shrine in a Tokyo meadow, you can see pictures of Nogi reading the newspaper on September 13, 1912, the morning of his boss’s funeral. No one was left to stop him. Near the photo you can see the sword he used later that day to disembowel himself.

I raise the example of General Nogi to encourage present-day leaders (military, political, educational) to take a much more modest step. They should offer to resign—often, and both in times of trouble and in times of calm. This weekend, the president of the University of Pennsylvania, Liz Magill, did the honorable thing, and the chair of Penn’s board, Scott Bok, followed his kōhai’s example shortly after. Magill resigned because she, along with Harvard President Claudine Gay and MIT President Sally Kornbluth, performed abysmally under questioning in Congress. Their inquisitor, upstate New York’s Elise Stefanik, a Republican, asked them whether chanting genocidal slogans violated their universities’ policies. It depends on the context, they all said, on the advice of counsel and the worst PR teams money can buy. Within days, Magill and Gay conceded that their answers had not been ideal. Gay is facing calls for her resignation, too.

Resign. Resign. Everyone: resign. Resignation has come to mean failure, something one does when cornered, caught dead to rights, incapable of continuing for even another day. It should be an act of honor—a high point in a career of service. It isn’t shameful. It is noble. It is the first and sometimes only step in the expiation of shame, and (ironically) the ultimate sign of one’s fitness for office.

No one demonstrates the value of these traits better than those who lack them entirely. I thought of Nogi’s katana, flashing from its scabbard, last week when the House voted to expel George Santos, Stefanik’s colleague in New York’s Republican delegation. The House almost never kicks anyone out, mainly because those facing expulsion have in the past tended to resign rather than weather the indignity of an expulsion vote. Santos is taking his ouster well and posting prolifically on TikTok. A psychologically normal person would have resigned the instant his tower of lies showed signs of wobbling. To let it crash down, then dance around the rubble of that tower until the orderlies arrive and pull you away, is truly mad behavior, and a demonstration of unfitness for the job, or indeed any job other than TikTok star.

[David Frum: There is no right to bully and harass]

I cannot prove it, but I believe the tendency to stick it out rather than resign started roughly when Representative Anthony Weiner (New York again, this time a Democrat) called a press conference to discuss whether he had, in fact, tweeted a picture of his penis, tumescent in his underwear. He could have just quit, and eventually he did (but lived to humiliate himself another day). But that pause to hold a press conference broke the seal on something dangerous, the idea that one can talk one’s way through a mortification like this. To take the podium and subject oneself to hostile questioning under those circumstances bespoke a delusionary chutzpah.

It soon became clear that anyone socially defective enough to persist through a scandal has a good chance of surviving it. By the time then-candidate Donald Trump (Republican, guess where) appeared on the Access Hollywood tape, describing his hobby of sexually assaulting women, it ceased to be obvious that at some point one should tap out and go home. If you have no shame, and you refuse to go, there might not be anyone out there who can make you. Mechanisms exist, as the Santos case shows. But the mechanisms were devised to govern people from another time, sensitive to ridicule and guffaws.

One should be ready for criticism, both earned and unearned. But resignation—more precisely, the offer of resignation—is an expression of confidence, both in oneself and in one’s employers or constituents. A board can reject a resignation. Voters can turn out in the streets to beg you to reconsider, or can turn out at the ballot to vote you back in. In fact, the more defensible one’s position, the greater esteem we should show for the one who offers to leave it. Call this the Nogi rule.

Harvard’s Claudine Gay evidently believed that she’d erred, because she reverted immediately to damage-control mode after leaving Washington. The next day, she told the Crimson that her testimony did not represent “my truth”—that is, that she disapproves of genocidal anti-Semitism. (This an extreme example of the political axiom “if you’re explaining, you’re losing.”) Her original answer before Congress lacked any visceral disapproval of anti-Semitism, certainly none to match Harvard’s recent record of condemning speech deemed offensive to historically disadvantaged groups. Her affect was robotic, neutral. She showed no signs of concern at all.

But her neutrality was born of an honorable principle, well worth defending. It reflected the values of free expression in a modern interpretation of the First Amendment, under which anyone can say just about any foolish thing, as long as saying it isn’t about to cause someone else to break the law. If the “context” of a genocidal chant is a nonviolent rally, the university shouldn’t stop anyone from chanting. (It should examine its soul. But that is another matter.) If the context is a crowd of protesters with bricks in hand, running at a group of Jews, the university should expel or fire every demonstrator there, whether armed with a brick or a bullhorn. All three presidents should have said this, then added a note of contrition over their universities’ failure to uphold these principles of free expression in the past.

But I’ll say it again: Gay should resign. To offer her neck to Harvard’s Board of Overseers would show her confidence that its members, like Emperor Meiji, would see past her error and ask her to endure in her position. It would also demonstrate her willingness to own that error, to acknowledge it publicly and unselfishly. Maybe the board would accept her resignation, and maybe it would not. Either of these fates is better than the one she is courting. At the moment she is trying to wriggle out of her error, and clinging to her job as if her dignity depended on keeping it. Better to teach by example that the reverse is often true, that dignity depends on leaving a job—and that staying suggests that one has nothing else, once it is gone.

[Greg Lukianoff: The latest victims of the free speech crisis]

The greatest legacy a resignation leaves is the creation of a culture of resignation. One institution that, up until now, has had such a culture is the Israeli defense establishment. A few weeks ago, I spoke with a former Mossad official who assured me that the entire leadership of the Mossad and the Israel Defense Forces would, as soon as the Gaza war reached a satisfactory pause, resign from their positions. They would do so, he said, because resignation was the only honorable response to their failure to foresee and prevent Hamas’s attack on October 7. Their predecessors did so in 2006, after the very messy war with Hezbollah in southern Lebanon, and after several other episodes of modest failure in Israeli history. That they might stick around, slinking back to their offices as if hoping everyone forgot about their mistakes, would be inconceivable. In this context, one understands better the popular rage against Prime Minister Benjamin Netanyahu, in whom the spirit of General Nogi is extinct: To this day, he is making the case to the Israeli right for his remaining their leader indefinitely.

One can’t get far in politics without a dogged willingness to destroy one’s critics and step on their corpses to reach the next height. But this is a minimal qualification for success, and everyone who attains high office, having climbed up from decades in the Senate or in departmental meetings, has it to an unusual degree. To persist is just to do what comes naturally for these people. To give up at the right moment—that is a quality against type, and a virtue possessed by the greatest of leaders. It is nevertheless available even to those who have hitherto shown no signs of greatness at all. Let it be said of them what Macbeth’s Duncan said of the Thane of Cawdor: that nothing became them in public service like the leaving it.

More than 500 faculty members have urged Harvard to 'resist political pressures' to force Dr Gay out.

Petition warns that bowing to pressure to fire Gay would be 'at odds with Harvard's commitment to academic freedom'.