There are a lot of good ideas floating around about the conditions necessary for life to develop, said Peter Girguis, professor of organismic and evolutionary biology in the Faculty of Arts and Sciences and co-director of the Harvard Microbial Sciences Initiative. 

What’s not known — and perhaps not provable — is whether life began on Earth, or if it arrived here from somewhere else. It’s an idea known as panspermia. 

It’s a “far out” hypothesis Girguis acknowledged, and one that makes a lot of scientists uncomfortable. It’s theoretically possible that some proto-microbe or bacterial spore hitched a ride to Earth on a meteorite billions of years ago. But science requires reproducibility, and with a data point of one, it’s difficult to prove either way. 

“We’re running into the very edges of our ability to use science to address some of these questions,” he said. 

Without a time machine, it’s unlikely we’ll prove panspermia one way or the other. But the broader question — whether life exists beyond Earth — is perhaps easier to tackle. In September, NASA reported that a Mars rover discovered chemical compounds that could be evidence of microbial life from billions of years before. It’s the closest sign to date of life on other worlds, but it’s far from a smoking gun. 

“I can’t assert that life exists on another planet because I have no data to support that,” Girguis said. “But as a curious and open-minded person, I would strongly argue that I have no scientific data to refute that, either. So if we’re talking about life elsewhere in the solar system or in the universe, I personally lean towards, ‘Yeah, maybe.’” 

Also interested in the question of life’s origin is David Kring, the 2025-2026 Edward, Frances, and Shirley B. Daniels Fellow at the Harvard Radcliffe Institute and a principal scientist at the Lunar and Planetary Institute of the Universities Space Research Association. 

Kring is a principal author of the impact origin of life hypothesis, which suggests that heavy asteroid bombardment about 4 billion years ago created hydrothermal environments rich in the kinds of chemicals that could lead to life. His team found the remnants of a microbial ecosystem in a hydrothermal system beneath the floor of the Chicxulub impact crater in Mexico. 

But, he said, it’s one of several reasonable hypotheses. 

“I don’t champion the idea. That’s a notion in science that I find offensive: It means you no longer have an open mind. I just want to know what the right answer is; whatever nature did is going to be interesting.” 

Between about 4.4 and 3.8 billion years ago, Earth and other inner solar system planets experienced a period of bombardment heavy enough, in some cases, to vaporize entire oceans. The impacts would have churned the Earth’s crust and created subsurface hydrothermal systems and chemical environments conducive to early life.

Unfortunately, Kring explained, “That period in Earth history is largely erased from the geologic record on our planet.” But there may be another way to get some answers: The same asteroids that bombarded Earth also did a number on the moon. 

“By collecting samples of impact craters or impact basins on the moon, we can determine their age and the types of asteroids or comets hitting the Earth-moon system,” Kring said. “Because those impacts jettisoned pieces of early Earth toward the moon, we also have an opportunity, if our minds are open to it, to find bits and pieces of early Earth history there.”  Kring has his eye on NASA’s 2028 Artemis IV mission, which is expected to send a crew to the surface of the moon for the first time in more than 50 years. 

The incidence of colorectal cancer among people under 50 has been rising by about 2 percent per year since the 1990s, and it is now the leading cause of cancer-related death among that age group. 

Kimmie Ng wants to know why. 

“The risk factors you may have heard of — obesity, sedentary behavior, ultra-processed foods — are associated with a higher risk of developing colorectal cancer under the age of 50. But I’m not sure those factors fully account for the rise,” said Ng, Harvard Medical School professor of medicine and director of the Young-Onset Colorectal Cancer Center at the Dana-Farber Cancer Institute.

“Obesity has been by far the leading hypothesis, but I and most of my colleagues will tell you that most of our patients are not obese,” she went on. “In fact, many are triathletes and marathon runners. They eat organic. There are no lifestyle factors that could explain why they ended up diagnosed with, unfortunately, Stage 4 colorectal cancer.”

Ng suspects the answer may lie in some combination of environmental and other factors that change the microbiome and the immune system, which surveil against cancer, in ways researchers still don’t understand. 

Her team is also studying whether the tumors they see in younger people are different from those in older adults.

“Young people get more treatment and more surgery; they tolerate treatment better and they’re healthier to begin with. But their survival is not necessarily better than that of somebody in their 70s. Why is that? Answering that question could lead to targeted treatment options specifically for younger patients.” 

Finding the answers, she said, will require the collaboration of cancer centers and scientists around the world in a variety of disciplines. “We just don’t have any other choice if we want to accelerate the pace of discovery.”

For Thomas Fel, the field of artificial intelligence is full of unanswered questions: How have we built something we don’t understand? What, or how, does AI think? How do we make sure AI doesn’t destroy us? And can we control something we don’t understand? 

Fel, a former research fellow at the Kempner Institute and current research affiliate in the Department of Psychology, says the field is still articulating the right way to ask the questions, let alone answer them. 

“This is not like understanding a clock or any other system that was built by humans,” Fel said. “This is kind of like reverse-engineering a brain that we accidentally built.” 

As AI bots get better at inferring context, building websites, and even chatting among themselves on their own social networks, it gets harder to think of them only as token-predicting machines. However, Fel said, referencing the infamous Chinese Room Argument, “Syntactic manipulation doesn’t entail semantic understanding.” Or, in other words, bots could act conscious, seem conscious, and tell us they’re conscious, all without actually being conscious. 

“Right now, we don’t have a theory of consciousness that can definitely say what is and isn’t consciousness,” Fel said. “So the crazy thing is, we might be building conscious machines and not even know it.” 

Of course, Fel added, AI doesn’t need to be conscious to be dangerous. 

“If it really is, as we think, a really smart system able to distribute itself efficiently, then we can easily imagine having some kind of a bad, bad experience. And the worst thing is, it wouldn’t even be intentional.”

In the AI industry, the subfield of AI focused on making sure the technology acts in humanity’s best interest is known as alignment. To some, the question of AI consciousness is separate from the question of alignment. But Fel doesn’t think that’s right.  “Meaningful alignment,” he said, “is fundamentally precarious without a comprehensive understanding of the agency we are attempting to direct.”

Melanie Wood thinks of prime numbers as the fundamental building blocks of integers: They’re to math what elements are to chemists.

“Water is two hydrogen atoms and an oxygen; to me, that’s just like 12 is two times two times three,” she said. 

But where chemists have more or less filled out the periodic table of elements (though whether or not heavier elements can be synthesized is another unanswered question), the list of prime numbers is infinite. It’s also random: The distribution of prime numbers among all natural numbers doesn’t follow any regular pattern. 

“We can tell you how many primes there are in the first 100 numbers, and then the first million numbers, and up and up, until it’s too big to compute on a computer,” said Wood, the William Caspar Graustein Professor of Mathematics at FAS. But identifying a rule that’s always true remains one of the great unsolved problems in math — in fact, this problem, dubbed the Riemann hypothesis, is one of the six as-yet-unsolved Millennium Prize Problems, with $1 million on the line for the mathematician who cracks it. 

“We have all the evidence in the world for the Riemann hypothesis, but we don’t know how to prove it,” Wood said. “We just don’t have any good idea how to proceed.”

Menopause — the end of women’s menstrual cycle and fertility — comes on like a domino effect. Spurred by the gradual loss of function of the ovarian follicles, the menstrual cycle becomes irregular, and the level of estrogen, which is made in the ovaries, fluctuates and progressively declines. Body composition changes, accompanied by changes in cholesterol and blood sugar levels, brain fog, and memory lapses. 

For decades, researchers have been exploring whether raising the level of estrogen and related hormones with hormone therapy can slow those physiological changes — and the chronic conditions that can follow. 

“Two-thirds of Alzheimer’s disease patients are women, and heart disease is the leading killer of women, especially after menopause” said JoAnn Manson, chief of preventive medicine and the Harvard Medical School Michael and Lee Bell Professor of Women’s Health at Brigham and Women’s Hospital. “So it’s a very natural question to ask: Can estrogen be used to delay the biological aging that occurs? Can it, essentially, preserve health and delay chronic disease?” 

Since 1993, Manson has been one of the leading researchers with the Women’s Health Initiative, a landmark study that, among other things, investigated the benefits and risks of hormone therapy in the prevention of chronic disease. But it studied the most common formulation of hormone therapy at the time, oral conjugated estrogens with and without a progestin, not the contemporary “bioidentical” formulation delivered transdermally.

Hormone therapy has been used since the 1940s to help manage menopause symptoms, Manson explained. However, a medical trend started in the 1980s to prescribe hormone treatment to prevent heart disease, stroke, and cognitive decline — all without clinical trials to show whether it was a good idea. 

“We really needed the Women’s Health Initiative randomized trials, and they did provide important answers,” Manson said.

The WHI found that younger women closer to the onset of menopause tended to benefit more from hormone therapy than women who were distant from menopause, and that the therapy might be protective against heart disease in younger women. But surprisingly, the WHI also found that there could be serious consequences from the oral hormone therapies available at the time, especially when started later in menopause. 

The oral hormone therapy formulations used in the WHI are infrequently prescribed now.

“Contemporary formulations of hormone therapy have never been tested in large-scale randomized trials to understand their effects on clinical outcomes,” Manson said. “That means we need the next-generation large trial to answer the big question — what is the role of hormone therapy in healthy aging? — but this time with the formulations matched more closely to women’s natural hormones and started earlier. This will provide key information relevant to women now and for many generations to come.” 

Manson and her colleagues are planning a new large-scale clinical trial to help find the answer. 

For the last 780,000 years, the Earth’s magnetic north pole has hovered somewhere around the Arctic Ocean. But episodically, due in part to the chaotic movements of molten metals in the Earth’s liquid outer core, the magnetic polarity reverses. 

Paleomagnetic records show that the Earth’s polarity has reversed hundreds of times in the planet’s history. Sometimes, reversals last tens of millions of years; sometimes, only tens of thousands. Planetary scientists don’t have a great handle on why the timelines vary so greatly. 

“You can construct computer simulations of the Earth’s core that produce outwardly similar behavior,” explained Roger Fu, professor of Earth and planetary sciences in the Faculty of Arts and Sciences. “But the best supercomputers would still take many decades to really get close to a truly realistic answer.

“We don’t know exactly what would happen if the field reversed, because this hasn’t happened since we’ve been tracking it,” he continued. But models suggest power lines might burn out, satellites would go down, and animals that rely on Earth’s magnetic field for navigation might become confused. 

Luckily, reversals take tens of thousands of years, so we should have plenty of time to adjust. “This is mainly a curiosity-driven question,” Fu said. “But people go back and forth on whether the magnetic field is or is not important for life, and this question gets at that.” 

To predict how technological advances will impact the labor market, Lawrence Katz has always seen economic theory as a better guide than the fantastic or dystopian visions of science fiction. But when it comes to artificial intelligence, he’s not so sure. 

“One view is that economic theory and our historical experience is the way to understand AI. But maybe science fiction writers — and philosophers — who have thought a lot about what it means to be human may be very insightful going forward.” 

Katz, the Elisabeth Allison Professor of Economics in the Faculty of Arts and Sciences, is closely watching the impact of AI on jobs, which as of now, he says, is unknowable. 

“We’re still in the early stages of seeing what artificial intelligence can do,” he said, “although it’s moving very rapidly.” 

Some employers, including Amazon, Meta, and the payments firm Block, have blamed recent or planned layoffs on productivity gains attributed to AI. One Goldman Sachs analysis found AI could automate tasks that make up 25 percent of all work hours. 

But the link between the automation of certain tasks and the layoff of a given employee is far from clear, Katz said. 

“Most jobs are not one task. So a lot of the open questions are: How does a job get rebundled or recharacterized when some of it gets automated?”

Katz is most worried about entry-level jobs, where young workers got paid to do “grunt work” while learning the tacit knowledge necessary to advance in their fields. Indeed, labor market conditions have worsened for recent college grads, with unemployment ticking up from 5.3 percent in the third quarter of 2025 to 5.7 percent in the fourth quarter. 

“New pathways to work, possibly with internships or cooperative education more integrated with traditional college courses, may be necessary to help get workers started on more fruitful career pathways,” he said.

One of the longstanding questions in quantum physics is what’s called the measurement problem: why different rules of physics apply depending on the presence or the absence of an observer. The classic thought experiment is “Schrödinger’s cat,” a creature somehow both alive and dead until somebody opens the box to check. 

“Do observers have some kind of magical power to change the rule, so the cat snaps into being either alive or dead but not some blend of alive and dead?” asked Jacob Barandes, a senior preceptor in physics and associated faculty in philosophy at FAS. 

Experiments have shown that the act of observation can change the thing being observed; even the order in which one observes certain properties can change the object’s other properties. Other experiments have shown that in some cases, the behavior of two particles can exhibit strange statistical correlations despite the particles being very far apart. 

It’s a frustrating conundrum for those of us who live in the world of Newtonian physics, where the rules of gravity and mass are doggedly consistent. 

The first proposed solution was the Copenhagen interpretation, Barandes said. That theory proposed a cutoff: Very small things made out of just a few atoms or subatomic particles might behave according to quantum rules, while anything larger behaved according to the rules of classical physics. 

But, he added, “Most physicists don’t think there is a line; it would lead to all sorts of questions, like, what if it’s one atom too few? It does seem a little arbitrary.” 

A later theory, decoherence, gets a lot closer, Barandes said. Decoherence argues that quantum systems lose their quantum behaviors due to interactions with other molecules and systems. Some in the field consider the measurement problem solved. Barandes doesn’t buy it: “Decoherence doesn’t explain how we get one outcome over the others,” he said. 

“I don’t know if this is a problem that’s solvable; I don’t know if it can be solved by an experiment. I don’t know whether solving it would require changing how we think about quantum theory in some more profound way.” 

But, he added, the solution may just as well be sociological as physical. 

“A variety of questions at the foundation of quantum physics were branded as unscientific, uninteresting, and unworthy of attention by many people in the scientific community, to the extent that working on them in a serious way was actively harmful to a person’s career,” Barandes said. “Much of the work on these problems came from people working on them on the side or in secret, with basically no money.” 

That work has already led to major advancements in the field, and to practical applications like quantum cryptography and computing. Just imagine, Barandes said, what would be possible with more investment in the field.