Bryan Reimer, a research scientist at the MIT Center for Transportation and Logistics (CTL), and the founder and co-leader of the Advanced Vehicle Technology Consortium and the Human Factors Evaluator for Automotive Demand Consortium in the MIT AgeLab, has been appointed to the Task Force on Human Factors in Aviation Safety Aviation Rulemaking Committee (HF Task Force ARC). The HF Task Force ARC will provide recommendations to the U.S. Federal Aviation Administration (FAA) on the most significant human factors and the relative contribution of these factors to aviation safety risk.

Reimer, who has worked at MIT since 2003, joins a committee whose operational or academic expertise includes air carrier operations, air traffic control, pilot experience, aeronautical information, aircraft maintenance and mechanics psychology, human-machine integration, and general aviation operations. Their recommendations to the FAA will help ensure safety for passengers, aircraft crews, and cargo for years to come. His appointment follows a year of serving on the Transforming Transportation Advisory Committee (TTAC) for the U.S. Department of Transportation, where he has taken on the role of vice chair on the Artificial Intelligence subcommittee. The TTAC recently released a report to the Secretary of Transportation in response to its charter.

As a mobility and technology futurist working at the intersection of technology, human behavior, and public policy, Reimer brings his expertise in human-machine integration, transportation safety, and AI to the committee. The committee, chartered by congressional mandate through the bipartisan FAA Reauthorization Act of 2024, specifically calls for a portion of the committee to have expertise on human factors but whose experience and training are not primarily in aviation, which Reimer will provide.

MIT CTL creates supply chain innovation and drives it into practice through the three pillars of research, outreach, and education, working with businesses, government, and nongovernmental organizations. As a longtime advocate of collaboration across public and private sectors to ensure consumers’ safety in transportation, Reimer’s particular expertise will help the FAA more broadly consider the human element of aviation safety. Yossi Sheffi, director of MIT CTL, says, “Aviation plays a critical role in the rapid and reliable transportation of goods across vast distances, making it essential for delivering time-sensitive products globally. We must understand the current human factors involved in this process to help ensure smooth operation of this indispensable service amid potential disruptions.”

Reimer recently discussed his research on an episode of The Ojo-Yoshida Report with Phil Koopman, a professor of electrical and computer engineering.

HF Task Force ARC members will serve a two-year term. The first ARC plenary meeting was held Jan. 15-16 in Washington.

Artificial intelligence has become vital in business and financial dealings, medical care, technology development, research, and much more. Without realizing it, consumers rely on AI when they stream a video, do online banking, or perform an online search. Behind these capabilities are more than 10,000 data centers globally, each one a huge warehouse containing thousands of computer servers and other infrastructure for storing, managing, and processing data. There are now over 5,000 data centers in the United States, and new ones are being built every day — in the U.S. and worldwide. Often dozens are clustered together right near where people live, attracted by policies that provide tax breaks and other incentives, and by what looks like abundant electricity.

And data centers do consume huge amounts of electricity. U.S. data centers consumed more than 4 percent of the country’s total electricity in 2023, and by 2030 that fraction could rise to 9 percent, according to the Electric Power Research Institute. A single large data center can consume as much electricity as 50,000 homes.

The sudden need for so many data centers presents a massive challenge to the technology and energy industries, government policymakers, and everyday consumers. Research scientists and faculty members at the MIT Energy Initiative (MITEI) are exploring multiple facets of this problem — from sourcing power to grid improvement to analytical tools that increase efficiency, and more. Data centers have quickly become the energy issue of our day.

Unexpected demand brings unexpected solutions

Several companies that use data centers to provide cloud computing and data management services are announcing some surprising steps to deliver all that electricity. Proposals include building their own small nuclear plants near their data centers and even restarting one of the undamaged nuclear reactors at Three Mile Island, which has been shuttered since 2019. (A different reactor at that plant partially melted down in 1979, causing the nation’s worst nuclear power accident.) Already the need to power AI is causing delays in the planned shutdown of some coal-fired power plants and raising prices for residential consumers. Meeting the needs of data centers is not only stressing power grids, but also setting back the transition to clean energy needed to stop climate change.

There are many aspects to the data center problem from a power perspective. Here are some that MIT researchers are focusing on, and why they’re important.

An unprecedented surge in the demand for electricity

“In the past, computing was not a significant user of electricity,” says William H. Green, director of MITEI and the Hoyt C. Hottel Professor in the MIT Department of Chemical Engineering. “Electricity was used for running industrial processes and powering household devices such as air conditioners and lights, and more recently for powering heat pumps and charging electric cars. But now all of a sudden, electricity used for computing in general, and by data centers in particular, is becoming a gigantic new demand that no one anticipated.”

Why the lack of foresight? Usually, demand for electric power increases by roughly half-a-percent per year, and utilities bring in new power generators and make other investments as needed to meet the expected new demand. But the data centers now coming online are creating unprecedented leaps in demand that operators didn’t see coming. In addition, the new demand is constant. It’s critical that a data center provides its services all day, every day. There can be no interruptions in processing large datasets, accessing stored data, and running the cooling equipment needed to keep all the packed-together computers churning away without overheating.

Moreover, even if enough electricity is generated, getting it to where it’s needed may be a problem, explains Deepjyoti Deka, a MITEI research scientist. “A grid is a network-wide operation, and the grid operator may have sufficient generation at another location or even elsewhere in the country, but the wires may not have sufficient capacity to carry the electricity to where it’s wanted.” So transmission capacity must be expanded — and, says Deka, that’s a slow process.

Then there’s the “interconnection queue.” Sometimes, adding either a new user (a “load”) or a new generator to an existing grid can cause instabilities or other problems for everyone else already on the grid. In that situation, bringing a new data center online may be delayed. Enough delays can result in new loads or generators having to stand in line and wait for their turn. Right now, much of the interconnection queue is already filled up with new solar and wind projects. The delay is now about five years. Meeting the demand from newly installed data centers while ensuring that the quality of service elsewhere is not hampered is a problem that needs to be addressed.

Finding clean electricity sources

To further complicate the challenge, many companies — including so-called “hyperscalers” such as Google, Microsoft, and Amazon — have made public commitments to having net-zero carbon emissions within the next 10 years. Many have been making strides toward achieving their clean-energy goals by buying “power purchase agreements.” They sign a contract to buy electricity from, say, a solar or wind facility, sometimes providing funding for the facility to be built. But that approach to accessing clean energy has its limits when faced with the extreme electricity demand of a data center.

Meanwhile, soaring power consumption is delaying coal plant closures in many states. There are simply not enough sources of renewable energy to serve both the hyperscalers and the existing users, including individual consumers. As a result, conventional plants fired by fossil fuels such as coal are needed more than ever.

As the hyperscalers look for sources of clean energy for their data centers, one option could be to build their own wind and solar installations. But such facilities would generate electricity only intermittently. Given the need for uninterrupted power, the data center would have to maintain energy storage units, which are expensive. They could instead rely on natural gas or diesel generators for backup power — but those devices would need to be coupled with equipment to capture the carbon emissions, plus a nearby site for permanently disposing of the captured carbon.

Because of such complications, several of the hyperscalers are turning to nuclear power. As Green notes, “Nuclear energy is well matched to the demand of data centers, because nuclear plants can generate lots of power reliably, without interruption.”

In a much-publicized move in September, Microsoft signed a deal to buy power for 20 years after Constellation Energy reopens one of the undamaged reactors at its now-shuttered nuclear plant at Three Mile Island, the site of the much-publicized nuclear accident in 1979. If approved by regulators, Constellation will bring that reactor online by 2028, with Microsoft buying all of the power it produces. Amazon also reached a deal to purchase power produced by another nuclear plant threatened with closure due to financial troubles. And in early December, Meta released a request for proposals to identify nuclear energy developers to help the company meet their AI needs and their sustainability goals.

Other nuclear news focuses on small modular nuclear reactors (SMRs), factory-built, modular power plants that could be installed near data centers, potentially without the cost overruns and delays often experienced in building large plants. Google recently ordered a fleet of SMRs to generate the power needed by its data centers. The first one will be completed by 2030 and the remainder by 2035.

Some hyperscalers are betting on new technologies. For example, Google is pursuing next-generation geothermal projects, and Microsoft has signed a contract to purchase electricity from a startup’s fusion power plant beginning in 2028 — even though the fusion technology hasn’t yet been demonstrated.

Reducing electricity demand

Other approaches to providing sufficient clean electricity focus on making the data center and the operations it houses more energy efficient so as to perform the same computing tasks using less power. Using faster computer chips and optimizing algorithms that use less energy are already helping to reduce the load, and also the heat generated.

Another idea being tried involves shifting computing tasks to times and places where carbon-free energy is available on the grid. Deka explains: “If a task doesn’t have to be completed immediately, but rather by a certain deadline, can it be delayed or moved to a data center elsewhere in the U.S. or overseas where electricity is more abundant, cheaper, and/or cleaner? This approach is known as ‘carbon-aware computing.’” We’re not yet sure whether every task can be moved or delayed easily, says Deka. “If you think of a generative AI-based task, can it easily be separated into small tasks that can be taken to different parts of the country, solved using clean energy, and then be brought back together? What is the cost of doing this kind of division of tasks?”

That approach is, of course, limited by the problem of the interconnection queue. It’s difficult to access clean energy in another region or state. But efforts are under way to ease the regulatory framework to make sure that critical interconnections can be developed more quickly and easily.

What about the neighbors?

A major concern running through all the options for powering data centers is the impact on residential energy consumers. When a data center comes into a neighborhood, there are not only aesthetic concerns but also more practical worries. Will the local electricity service become less reliable? Where will the new transmission lines be located? And who will pay for the new generators, upgrades to existing equipment, and so on? When new manufacturing facilities or industrial plants go into a neighborhood, the downsides are generally offset by the availability of new jobs. Not so with a data center, which may require just a couple dozen employees.

There are standard rules about how maintenance and upgrade costs are shared and allocated. But the situation is totally changed by the presence of a new data center. As a result, utilities now need to rethink their traditional rate structures so as not to place an undue burden on residents to pay for the infrastructure changes needed to host data centers.

MIT’s contributions

At MIT, researchers are thinking about and exploring a range of options for tackling the problem of providing clean power to data centers. For example, they are investigating architectural designs that will use natural ventilation to facilitate cooling, equipment layouts that will permit better airflow and power distribution, and highly energy-efficient air conditioning systems based on novel materials. They are creating new analytical tools for evaluating the impact of data center deployments on the U.S. power system and for finding the most efficient ways to provide the facilities with clean energy. Other work looks at how to match the output of small nuclear reactors to the needs of a data center, and how to speed up the construction of such reactors.

MIT teams also focus on determining the best sources of backup power and long-duration storage, and on developing decision support systems for locating proposed new data centers, taking into account the availability of electric power and water and also regulatory considerations, and even the potential for using what can be significant waste heat, for example, for heating nearby buildings. Technology development projects include designing faster, more efficient computer chips and more energy-efficient computing algorithms.

In addition to providing leadership and funding for many research projects, MITEI is acting as a convenor, bringing together companies and stakeholders to address this issue. At MITEI’s 2024 Annual Research Conference, a panel of representatives from two hyperscalers and two companies that design and construct data centers together discussed their challenges, possible solutions, and where MIT research could be most beneficial.

As data centers continue to be built, and computing continues to create an unprecedented increase in demand for electricity, Green says, scientists and engineers are in a race to provide the ideas, innovations, and technologies that can meet this need, and at the same time continue to advance the transition to a decarbonized energy system.

Ammonia is the most widely produced chemical in the world today, used primarily as a source for nitrogen fertilizer. Its production is also a major source of greenhouse gas emissions — the highest in the whole chemical industry.

Now, a team of researchers at MIT has developed an innovative way of making ammonia without the usual fossil-fuel-powered chemical plants that require high heat and pressure. Instead, they have found a way to use the Earth itself as a geochemical reactor, producing ammonia underground. The processes uses Earth’s naturally occurring heat and pressure, provided free of charge and free of emissions, as well as the reactivity of minerals already present in the ground.

The trick the team devised is to inject water underground, into an area of iron-rich subsurface rock. The water carries with it a source of nitrogen and particles of a metal catalyst, allowing the water to react with the iron to generate clean hydrogen, which in turn reacts with the nitrogen to make ammonia. A second well is then used to pump that ammonia up to the surface.

The process, which has been demonstrated in the lab but not yet in a natural setting, is described today in the journal Joule. The paper’s co-authors are MIT professors of materials science and engineering Iwnetim Abate and Ju Li, graduate student Yifan Gao, and five others at MIT.

“When I first produced ammonia from rock in the lab, I was so excited,” Gao recalls. “I realized this represented an entirely new and never-reported approach to ammonia synthesis.’”

The standard method for making ammonia is called the Haber-Bosch process, which was developed in Germany in the early 20th century to replace natural sources of nitrogen fertilizer such as mined deposits of bat guano, which were becoming depleted. But the Haber-Bosch process is very energy intensive: It requires temperatures of 400 degrees Celsius and pressures of 200 atmospheres, and this means it needs huge installations in order to be efficient. Some areas of the world, such as sub-Saharan Africa and Southeast Asia, have few or no such plants in operation.  As a result, the shortage or extremely high cost of fertilizer in these regions has limited their agricultural production.

The Haber-Bosch process “is good. It works,” Abate says. “Without it, we wouldn’t have been able to feed 2 out of the total 8 billion people in the world right now, he says, referring to the portion of the world’s population whose food is grown with ammonia-based fertilizers. But because of the emissions and energy demands, a better process is needed, he says.

Burning fuel to generate heat is responsible for about 20 percent of the greenhouse gases emitted from plants using the Haber-Bosch process. Making hydrogen accounts for the remaining 80 percent.  But ammonia, the molecule NH3, is made up only of nitrogen and hydrogen. There’s no carbon in the formula, so where do the carbon emissions come from? The standard way of producing the needed hydrogen is by processing methane gas with steam, breaking down the gas into pure hydrogen, which gets used, and carbon dioxide gas that gets released into the air.

Other processes exist for making low- or no-emissions hydrogen, such as by using solar or wind-generated electricity to split water into oxygen and hydrogen, but that process can be expensive. That’s why Abate and his team worked on developing a system to produce what they call geological hydrogen. Some places in the world, including some in Africa, have been found to naturally generate hydrogen underground through chemical reactions between water and iron-rich rocks. These pockets of naturally occurring hydrogen can be mined, just like natural methane reservoirs, but the extent and locations of such deposits are still relatively unexplored.

Abate realized this process could be created or enhanced by pumping water, laced with copper and nickel catalyst particles to speed up the process, into the ground in places where such iron-rich rocks were already present. “We can use the Earth as a factory to produce clean flows of hydrogen,” he says.

He recalls thinking about the problem of the emissions from hydrogen production for ammonia: “The ‘aha!’ moment for me was thinking, how about we link this process of geological hydrogen production with the process of making Haber-Bosch ammonia?”

That would solve the biggest problem of the underground hydrogen production process, which is how to capture and store the gas once it’s produced. Hydrogen is a very tiny molecule — the smallest of them all — and hard to contain. But by implementing the entire Haber-Bosch process underground, the only material that would need to be sent to the surface would be the ammonia itself, which is easy to capture, store, and transport.

The only extra ingredient needed to complete the process was the addition of a source of nitrogen, such as nitrate or nitrogen gas, into the water-catalyst mixture being injected into the ground. Then, as the hydrogen gets released from water molecules after interacting with the iron-rich rocks, it can immediately bond with the nitrogen atoms also carried in the water, with the deep underground environment providing the high temperatures and pressures required by the Haber-Bosch process. A second well near the injection well then pumps the ammonia out and into tanks on the surface.

“We call this geological ammonia,” Abate says, “because we are using subsurface temperature, pressure, chemistry, and geologically existing rocks to produce ammonia directly.”

Whereas transporting hydrogen requires expensive equipment to cool and liquefy it, and virtually no pipelines exist for its transport (except near oil refinery sites), transporting ammonia is easier and cheaper. It’s about one-sixth the cost of transporting hydrogen, and there are already more than 5,000 miles of ammonia pipelines and 10,000 terminals in place in the U.S. alone. What’s more, Abate explains, ammonia, unlike hydrogen, already has a substantial commercial market in place, with production volume projected to grow by two to three times by 2050, as it is used not only for fertilizer but also as feedstock for a wide variety of chemical processes.

For example, ammonia can be burned directly in gas turbines, engines, and industrial furnaces, providing a carbon-free alternative to fossil fuels. It is being explored for maritime shipping and aviation as an alternative fuel, and as a possible space propellant.

Another upside to geological ammonia is that untreated wastewater, including agricultural runoff, which tends to be rich in nitrogen already, could serve as the water source and be treated in the process. “We can tackle the problem of treating wastewater, while also making something of value out of this waste,” Abate says.

Gao adds that this process “involves no direct carbon emissions, presenting a potential pathway to reduce global CO2 emissions by up to 1 percent.” To arrive at this point, he says, the team “overcame numerous challenges and learned from many failed attempts. For example, we tested a wide range of conditions and catalysts before identifying the most effective one.”

The project was seed-funded under a flagship project of MIT’s Climate Grand Challenges program, the Center for the Electrification and Decarbonization of Industry. Professor Yet-Ming Chiang, co-director of the center, says “I don’t think there’s been any previous example of deliberately using the Earth as a chemical reactor. That’s one of the key novel points of this approach.”  Chiang emphasizes that even though it is a geological process, it happens very fast, not on geological timescales. “The reaction is fundamentally over in a matter of hours,” he says. “The reaction is so fast that this answers one of the key questions: Do you have to wait for geological times? And the answer is absolutely no.”

Professor Elsa Olivetti, a mission director of the newly established Climate Project at MIT, says, “The creative thinking by this team is invaluable to MIT’s ability to have impact at scale. Coupling these exciting results with, for example, advanced understanding of the geology surrounding hydrogen accumulations represent the whole-of-Institute efforts the Climate Project aims to support.”

“This is a significant breakthrough for the future of sustainable development,” says Geoffrey Ellis, a geologist at the U.S. Geological Survey, who was not associated with this work. He adds, “While there is clearly more work that needs to be done to validate this at the pilot stage and to get this to the commercial scale, the concept that has been demonstrated is truly transformative.  The approach of engineering a system to optimize the natural process of nitrate reduction by Fe2+ is ingenious and will likely lead to further innovations along these lines.”

The initial work on the process has been done in the laboratory, so the next step will be to prove the process using a real underground site. “We think that kind of experiment can be done within the next one to two years,” Abate says. This could open doors to using a similar approach for other chemical production processes, he adds.

The team has applied for a patent and aims to work towards bringing the process to market.

“Moving forward,” Gao says, “our focus will be on optimizing the process conditions and scaling up tests, with the goal of enabling practical applications for geological ammonia in the near future.”

The research team also included Ming Lei, Bachu Sravan Kumar, Hugh Smith, Seok Hee Han, and Lokesh Sangabattula, all at MIT. Additional funding was provided by the National Science Foundation and was carried out, in part, through the use of MIT.nano facilities.

The roundworm C. elegans is a simple animal whose nervous system has exactly 302 neurons. Each of the connections between those neurons has been comprehensively mapped, allowing researchers to study how they work together to generate the animal’s different behaviors.

Steven Flavell, an MIT associate professor of brain and cognitive sciences and investigator with The Picower Institute for Learning and Memory at MIT and the Howard Hughes Medical Institute, uses the worm as a model to study motivated behaviors such as feeding and navigation, in hopes of shedding light on the fundamental mechanisms that may also determine how similar behaviors are controlled in other animals.

In recent studies, Flavell’s lab has uncovered neural mechanisms underlying adaptive changes in the worms’ feeding behavior, and his lab has also mapped how the activity of each neuron in the animal’s nervous system affects the worms’ different behaviors.

Such studies could help researchers gain insight into how brain activity generates behavior in humans. “It is our aim to identify molecular and neural circuit mechanisms that may generalize across organisms,” he says, noting that many fundamental biological discoveries, including those related to programmed cell death, microRNA, and RNA interference, were first made in C. elegans.

“Our lab has mostly studied motivated state-dependent behaviors, like feeding and navigation. The machinery that’s being used to control these states in C. elegans — for example, neuromodulators — are actually the same as in humans. These pathways are evolutionarily ancient,” he says.

Drawn to the lab

Born in London to an English father and a Dutch mother, Flavell came to the United States in 1982 at the age of 2, when his father became chief scientific officer at Biogen. The family lived in Sudbury, Massachusetts, and his mother worked as a computer programmer and math teacher. His father later became a professor of immunology at Yale University.

Though Flavell grew up in a science family, he thought about majoring in English when he arrived at Oberlin College. A musician as well, Flavell took jazz guitar classes at Oberlin’s conservatory, and he also plays the piano and the saxophone. However, taking classes in psychology and physiology led him to discover that the field that most captivated him was neuroscience.

“I was immediately sold on neuroscience. It combined the rigor of the biological sciences with deep questions from psychology,” he says.

While in college, Flavell worked on a summer research project related to Alzheimer’s disease, in a lab at Case Western Reserve University. He then continued the project, which involved analyzing post-mortem Alzheimer’s tissue, during his senior year at Oberlin.

“My earliest research revolved around mechanisms of disease. While my research interests have evolved since then, my earliest research experiences were the ones that really got me hooked on working at the bench: running experiments, looking at brand new results, and trying to understand what they mean,” he says.

By the end of college, Flavell was a self-described lab rat: “I just love being in the lab.” He applied to graduate school and ended up going to Harvard Medical School for a PhD in neuroscience. Working with Michael Greenberg, Flavell studied how sensory experience and resulting neural activity shapes brain development. In particular, he focused on a family of gene regulators called MEF2, which play important roles in neuronal development and synaptic plasticity.

All of that work was done using mouse models, but Flavell transitioned to studying C. elegans during a postdoctoral fellowship working with Cori Bargmann at Rockefeller University. He was interested in studying how neural circuits control behavior, which seemed to be more feasible in simpler animal models.

“Studying how neurons across the brain govern behavior felt like it would be nearly intractable in a large brain — to understand all the nuts and bolts of how neurons interact with each other and ultimately generate behavior seemed daunting,” he says. “But I quickly became excited about studying this in C. elegans because at the time it was still the only animal with a full blueprint of its brain: a map of every brain cell and how they are all wired up together.”

That wiring diagram includes about 7,000 synapses in the entire nervous system. By comparison, a single human neuron may form more than 10,000 synapses. “Relative to those larger systems, the C. elegans nervous system is mind-bogglingly simple,” Flavell says.

Despite their much simpler organization, roundworms can execute complex behaviors such as feeding, locomotion, and egg-laying. They even sleep, form memories, and find suitable mating partners. The neuromodulators and cellular machinery that give rise to those behaviors are similar to those found in humans and other mammals.

“C. elegans has a fairly well-defined, smallish set of behaviors, which makes it attractive for research. You can really measure almost everything that the animal is doing and study it,” Flavell says.

How behavior arises

Early in his career, Flavell’s work on C. elegans revealed the neural mechanisms that underlie the animal’s stable behavioral states. When worms are foraging for food, they alternate between stably exploring the environment and pausing to feed. “The transition rates between those states really depend on all these cues in the environment. How good is the food environment? How hungry are they? Are there smells indicating a better nearby food source? The animal integrates all of those things and then adjusts their foraging strategy,” Flavell says.

These stable behavioral states are controlled by neuromodulators like serotonin. By studying serotonergic regulation of the worm’s behavioral states, Flavell’s lab has been able to uncover how this important system is organized. In a recent study, Flavell and his colleagues published an “atlas” of the C. elegans serotonin system. They identified every neuron that produces serotonin, every neuron that has serotonin receptors, and how brain activity and behavior change across the animal as serotonin is released.

“Our studies of how the serotonin system works to control behavior have already revealed basic aspects of serotonin signaling that we think ought to generalize all the way up to mammals,” Flavell says. “By studying the way that the brain implements these long-lasting states, we can tap into these basic features of neuronal function. With the resolution that you can get studying specific C. elegans neurons and the way that they implement behavior, we can uncover fundamental features of the way that neurons act.”

In parallel, Flavell’s lab has also been mapping out how neurons across the C. elegans brain control different aspects of behavior. In a 2023 study, Flavell’s lab mapped how changes in brain-wide activity relate to behavior. His lab uses special microscopes that can move along with the worms as they explore, allowing them to simultaneously track every behavior and measure the activity of every neuron in the brain. Using these data, the researchers created computational models that can accurately capture the relationship between brain activity and behavior.

This type of research requires expertise in many areas, Flavell says. When looking for faculty jobs, he hoped to find a place where he could collaborate with researchers working in different fields of neuroscience, as well as scientists and engineers from other departments.

“Being at MIT has allowed my lab to be much more multidisciplinary than it could have been elsewhere,” he says. “My lab members have had undergrad degrees in physics, math, computer science, biology, neuroscience, and we use tools from all of those disciplines. We engineer microscopes, we build computational models, we come up with molecular tricks to perturb neurons in the C. elegans nervous system. And I think being able to deploy all those kinds of tools leads to exciting research outcomes.”

In a two-part series, MIT News explores the environmental implications of generative AI. In this article, we look at why this technology is so resource-intensive. A second piece will investigate what experts are doing to reduce genAI’s carbon footprint and other impacts.

The excitement surrounding potential benefits of generative AI, from improving worker productivity to advancing scientific research, is hard to ignore. While the explosive growth of this new technology has enabled rapid deployment of powerful models in many industries, the environmental consequences of this generative AI “gold rush” remain difficult to pin down, let alone mitigate.

The computational power required to train generative AI models that often have billions of parameters, such as OpenAI’s GPT-4, can demand a staggering amount of electricity, which leads to increased carbon dioxide emissions and pressures on the electric grid.

Furthermore, deploying these models in real-world applications, enabling millions to use generative AI in their daily lives, and then fine-tuning the models to improve their performance draws large amounts of energy long after a model has been developed.

Beyond electricity demands, a great deal of water is needed to cool the hardware used for training, deploying, and fine-tuning generative AI models, which can strain municipal water supplies and disrupt local ecosystems. The increasing number of generative AI applications has also spurred demand for high-performance computing hardware, adding indirect environmental impacts from its manufacture and transport.

“When we think about the environmental impact of generative AI, it is not just the electricity you consume when you plug the computer in. There are much broader consequences that go out to a system level and persist based on actions that we take,” says Elsa A. Olivetti, professor in the Department of Materials Science and Engineering and the lead of the Decarbonization Mission of MIT’s new Climate Project.

Olivetti is senior author of a 2024 paper, “The Climate and Sustainability Implications of Generative AI,” co-authored by MIT colleagues in response to an Institute-wide call for papers that explore the transformative potential of generative AI, in both positive and negative directions for society.

Demanding data centers

The electricity demands of data centers are one major factor contributing to the environmental impacts of generative AI, since data centers are used to train and run the deep learning models behind popular tools like ChatGPT and DALL-E.

A data center is a temperature-controlled building that houses computing infrastructure, such as servers, data storage drives, and network equipment. For instance, Amazon has more than 100 data centers worldwide, each of which has about 50,000 servers that the company uses to support cloud computing services.

While data centers have been around since the 1940s (the first was built at the University of Pennsylvania in 1945 to support the first general-purpose digital computer, the ENIAC), the rise of generative AI has dramatically increased the pace of data center construction.

“What is different about generative AI is the power density it requires. Fundamentally, it is just computing, but a generative AI training cluster might consume seven or eight times more energy than a typical computing workload,” says Noman Bashir, lead author of the impact paper, who is a Computing and Climate Impact Fellow at MIT Climate and Sustainability Consortium (MCSC) and a postdoc in the Computer Science and Artificial Intelligence Laboratory (CSAIL).

Scientists have estimated that the power requirements of data centers in North America increased from 2,688 megawatts at the end of 2022 to 5,341 megawatts at the end of 2023, partly driven by the demands of generative AI. Globally, the electricity consumption of data centers rose to 460 terawatts in 2022. This would have made data centers the 11th largest electricity consumer in the world, between the nations of Saudi Arabia (371 terawatts) and France (463 terawatts), according to the Organization for Economic Co-operation and Development.

By 2026, the electricity consumption of data centers is expected to approach 1,050 terawatts (which would bump data centers up to fifth place on the global list, between Japan and Russia).

While not all data center computation involves generative AI, the technology has been a major driver of increasing energy demands.

“The demand for new data centers cannot be met in a sustainable way. The pace at which companies are building new data centers means the bulk of the electricity to power them must come from fossil fuel-based power plants,” says Bashir.

The power needed to train and deploy a model like OpenAI’s GPT-3 is difficult to ascertain. In a 2021 research paper, scientists from Google and the University of California at Berkeley estimated the training process alone consumed 1,287 megawatt hours of electricity (enough to power about 120 average U.S. homes for a year), generating about 552 tons of carbon dioxide.

While all machine-learning models must be trained, one issue unique to generative AI is the rapid fluctuations in energy use that occur over different phases of the training process, Bashir explains.

Power grid operators must have a way to absorb those fluctuations to protect the grid, and they usually employ diesel-based generators for that task.

Increasing impacts from inference

Once a generative AI model is trained, the energy demands don’t disappear.

Each time a model is used, perhaps by an individual asking ChatGPT to summarize an email, the computing hardware that performs those operations consumes energy. Researchers have estimated that a ChatGPT query consumes about five times more electricity than a simple web search.

“But an everyday user doesn’t think too much about that,” says Bashir. “The ease-of-use of generative AI interfaces and the lack of information about the environmental impacts of my actions means that, as a user, I don’t have much incentive to cut back on my use of generative AI.”

With traditional AI, the energy usage is split fairly evenly between data processing, model training, and inference, which is the process of using a trained model to make predictions on new data. However, Bashir expects the electricity demands of generative AI inference to eventually dominate since these models are becoming ubiquitous in so many applications, and the electricity needed for inference will increase as future versions of the models become larger and more complex.

Plus, generative AI models have an especially short shelf-life, driven by rising demand for new AI applications. Companies release new models every few weeks, so the energy used to train prior versions goes to waste, Bashir adds. New models often consume more energy for training, since they usually have more parameters than their predecessors.

While electricity demands of data centers may be getting the most attention in research literature, the amount of water consumed by these facilities has environmental impacts, as well.

Chilled water is used to cool a data center by absorbing heat from computing equipment. It has been estimated that, for each kilowatt hour of energy a data center consumes, it would need two liters of water for cooling, says Bashir.

“Just because this is called ‘cloud computing’ doesn’t mean the hardware lives in the cloud. Data centers are present in our physical world, and because of their water usage they have direct and indirect implications for biodiversity,” he says.

The computing hardware inside data centers brings its own, less direct environmental impacts.

While it is difficult to estimate how much power is needed to manufacture a GPU, a type of powerful processor that can handle intensive generative AI workloads, it would be more than what is needed to produce a simpler CPU because the fabrication process is more complex. A GPU’s carbon footprint is compounded by the emissions related to material and product transport.

There are also environmental implications of obtaining the raw materials used to fabricate GPUs, which can involve dirty mining procedures and the use of toxic chemicals for processing.

Market research firm TechInsights estimates that the three major producers (NVIDIA, AMD, and Intel) shipped 3.85 million GPUs to data centers in 2023, up from about 2.67 million in 2022. That number is expected to have increased by an even greater percentage in 2024.

The industry is on an unsustainable path, but there are ways to encourage responsible development of generative AI that supports environmental objectives, Bashir says.

He, Olivetti, and their MIT colleagues argue that this will require a comprehensive consideration of all the environmental and societal costs of generative AI, as well as a detailed assessment of the value in its perceived benefits.

“We need a more contextual way of systematically and comprehensively understanding the implications of new developments in this space. Due to the speed at which there have been improvements, we haven’t had a chance to catch up with our abilities to measure and understand the tradeoffs,” Olivetti says.

The MIT Global Supply Chain and Logistics Excellence (SCALE) Network has once again been ranked as the world’s top master’s program for supply chain and logistics management by Eduniversal’s 2024/2025 Best Masters Rankings. This recognition marks the eighth consecutive No. 1 ranking since 2016, reaffirming MIT’s unparalleled leadership in supply chain education, research, and practice.

Eduniversal evaluates more than 20,000 postgraduate programs globally each year, considering academic reputation, graduate employability, and student satisfaction.

The MIT SCALE Network’s sustained top ranking reflects its commitment to fostering international diversity; delivering hands-on, project-based learning; and success in developing a generation of supply chain leaders ready to tackle global supply chain challenges.

A growing global network with local impact

This year’s ranking coincides with the MIT SCALE Network’s expansion of its global footprint, highlighted by the recent announcement of the UK SCALE Center at Loughborough University. The center, which will welcome its inaugural cohort in fall 2025, underscores MIT’s commitment to advancing supply chain innovation and creating transformative opportunities for students and researchers.

The UK SCALE Center joins the network’s global community of centers in the United States, China, Spain, Colombia, and Luxembourg. Together, these centers deliver world-class education and practical solutions that address critical supply chain challenges across industries, empowering a global alumni base of more than 1,900 leaders representing over 50 different countries.

"The launch of the UK SCALE Center represents a fantastic opportunity for Loughborough University to showcase our cutting-edge research and data-driven, forward-thinking approach to supporting the U.K. supply chain industry,” says Jan Godsell, dean of Loughborough Business School. “Through projects like the InterAct Network and our implementation of the Made Smarter Innovation 'Leading Digital Transformation' program, we’re offering businesses and industry professionals the essential training and leading insights into the future of the supply chain ecosystem, which I’m excited to build on with the creation of this new MSc in supply chain management."

Other MIT SCALE centers also emphasized the network’s transformative impact:

“The MIT SCALE Network provides NISCI students with the tools, expertise, and global connections to lead in today’s rapidly evolving supply chain environment,” says Jay Guo, director of the Ningbo China Institute for Supply Chain Innovation.

Susana Val, director of Zaragoza Logistics Center (ZLC), highlights the program’s reach and influence: “For the last 21 years, ZLC has educated over 5,000 logistics professionals from more than 90 nationalities. We are proud of this recognition and look forward to continuing our alliance with the MIT SCALE Network, upholding the rigor and quality that define our teaching.”

From Luxembourg, Benny Mantin, director of the Luxembourg Center for Logistics and Supply Chain Management (LCL), adds, “Our students greatly appreciate the LCL’s SCALE Network membership as it provides them with superb experience and ample opportunities to network and expand their scope.”

The global presence and collaborative approach of the MIT SCALE Network help define its mission: to deliver education and research that drive transformative impact in every corner of the world.

A legacy of leadership

This latest recognition from Eduniversal underscores the MIT SCALE Network’s leadership in supply chain education. For over two decades, its master’s programs have shaped graduates who tackle pressing challenges across industries and geographies.

"This recognition reflects the dedication of our faculty, researchers, and global partners to delivering excellence in supply chain education," says Yossi Sheffi, director of the MIT Center for Transportation and Logistics. “The MIT SCALE Network’s alumni are proof of the program’s impact, applying their skills to tackle challenges across every industry and continent.”

Maria Jesus Saenz, executive director of the MIT SCM Master’s Program, emphasizes the strength of the global alumni network: “The MIT SCALE Network doesn’t just prepare graduates — it connects them to a global community of supply chain leaders. This powerful ecosystem fosters collaboration and innovation that transcends borders, enabling our graduates to tackle the world’s most pressing supply chain challenges.”

Founded in 2003, the MIT SCALE Network connects world-class research centers across multiple continents, offering top-ranked master’s and executive education programs that combine academic rigor with real-world application. Graduates are among the most sought-after professionals in the global supply chain field.

In the world of high-priced art, galleries usually act as gatekeepers. Their selective curation process is a key reason galleries in major cities often feature work from the same batch of artists. The system limits opportunities for emerging artists and leaves great art undiscovered.

NALA was founded by Benjamin Gulak ’22 to disrupt the gallery model. The company’s digital platform, which was started as part of an MIT class project, allows artists to list their art and uses machine learning and data science to offer personalized recommendations to art lovers.

By providing a much larger pool of artwork to buyers, the company is dismantling the exclusive barriers put up by traditional galleries and efficiently connecting creators with collectors.

“There’s so much talent out there that has never had the opportunity to be seen outside of the artists’ local market,” Gulak says. “We’re opening the art world to all artists, creating a true meritocracy.”

NALA takes no commission from artists, instead charging buyers an 11.5 percent commission on top of the artist’s listed price. Today more than 20,000 art lovers are using NALA's platform, and the company has registered more than 8,500 artists.

“My goal is for NALA to become the dominant place where art is discovered, bought, and sold online,” Gulak says. “The gallery model has existed for such a long period of time that they are the tastemakers in the art world. However, most buyers never realize how restrictive the industry has been.”

From founder to student to founder again

Growing up in Canada, Gulak worked hard to get into MIT, participating in science fairs and robotic competitions throughout high school. When he was 16, he created an electric, one-wheeled motorcycle that got him on the popular television show “Shark Tank” and was later named one of the top inventions of the year by Popular Science.

Gulak was accepted into MIT in 2009 but withdrew from his undergrad program shortly after entering to launch a business around the media exposure and capital from “Shark Tank.” Following a whirlwind decade in which he raised more than $12 million and sold thousands of units globally, Gulak decided to return to MIT to complete his degree, switching his major from mechanical engineering to one combining computer science, economics, and data science.

“I spent 10 years of my life building my business, and realized to get the company where I wanted it to be, it would take another decade, and that wasn’t what I wanted to be doing,” Gulak says. “I missed learning, and I missed the academic side of my life. I basically begged MIT to take me back, and it was the best decision I ever made.”

During the ups and downs of running his company, Gulak took up painting to de-stress. Art had always been a part of Gulak’s life, and he had even done a fine arts study abroad program in Italy during high school. Determined to try selling his art, he collaborated with some prominent art galleries in London, Miami, and St. Moritz. Eventually he began connecting artists he’d met on travels from emerging markets like Cuba, Egypt, and Brazil to the gallery owners he knew.

“The results were incredible because these artists were used to selling their work to tourists for $50, and suddenly they’re hanging work in a fancy gallery in London and getting 5,000 pounds,” Gulak says. “It was the same artist, same talent, but different buyers.”

At the time, Gulak was in his third year at MIT and wondering what he’d do after graduation. He thought he wanted to start a new business, but every industry he looked at was dominated by tech giants. Every industry, that is, except the art world.

“The art industry is archaic,” Gulak says. “Galleries have monopolies over small groups of artists, and they have absolute control over the prices. The buyers are told what the value is, and almost everywhere you look in the industry, there’s inefficiencies.”

At MIT, Gulak was studying the recommender engines that are used to populate social media feeds and personalize show and music suggestions, and he envisioned something similar for the visual arts.

“I thought, why, when I go on the big art platforms, do I see horrible combinations of artwork even though I’ve had accounts on these platforms for years?” Gulak says. “I’d get new emails every week titled ‘New art for your collection,’ and the platform had no idea about my taste or budget.”

For a class project at MIT, Gulak built a system that tried to predict the types of art that would do well in a gallery. By his final year at MIT, he had realized that working directly with artists would be a more promising approach.

“Online platforms typically take a 30 percent fee, and galleries can take an additional 50 percent fee, so the artist ends up with a small percentage of each online sale, but the buyer also has to pay a luxury import duty on the full price,” Gulak explains. “That means there’s a massive amount of fat in the middle, and that’s where our direct-to-artist business model comes in.”

Today NALA, which stands for Networked Artistic Learning Algorithm, onboards artists by having them upload artwork and fill out a questionnaire about their style. They can begin uploading work immediately and choose their listing price.

The company began by using AI to match art with its most likely buyer. Gulak notes that not all art will sell — “if you’re making rock paintings there may not be a big market” — and artists may price their work higher than buyers are willing to pay, but the algorithm works to put art in front of the most likely buyer based on style preferences and budget. NALA also handles sales and shipments, providing artists with 100 percent of their list price from every sale.

“By not taking commissions, we’re very pro artists,” Gulak says. “We also allow all artists to participate, which is unique in this space. NALA is built by artists for artists.”

Last year, NALA also started allowing buyers to take a photo of something they like and see similar artwork from its database.

“In museums, people will take a photo of masterpieces they’ll never be able to afford, and now they can find living artists producing the same style that they could actually put in their home,” Gulak says. “It makes art more accessible.”

Championing artists

Ten years ago, Ben Gulak was visiting Egypt when he discovered an impressive mural on the street. Gulak found the local artist, Ahmed Nofal, on Instagram and bought some work. Later, he brought Nofal to Dubai to participate in World Art Dubai. The artist’s work was so well-received he ended up creating murals for the Royal British Museum in London and Red Bull. Most recently, Nofal and Gulak collaborated together during Art Basel 2024 doing a mural at the Museum of Graffiti in Miami.

Gulak has worked personally with many of the artists on his platform. For more than a decade he’s travelled to Cuba buying art and delivering art supplies to friends. He’s also worked with artists as they work to secure immigration visas.

“Many people claim they want to help the art world, but in reality, they often fall back on the same outdated business models,” says Gulak. “Art isn’t just my passion — it’s a way of life for me. I’ve been on every side of the art world: as a painter selling my work through galleries, as a collector with my office brimming with art, and as a collaborator working alongside incredible talents like Raheem Saladeen Johnson. When artists visit, we create together, sharing ideas and brainstorming. These experiences, combined with my background as both an artist and a computer scientist, give me a unique perspective. I’m trying to use technology to provide artists with unparalleled access to the global market and shake things up.”

Karl K. Berggren, the Joseph F. and Nancy P. Keithley Professor of Electrical Engineering at MIT, has been named the new faculty head of electrical engineering in the Department of Electrical Engineering and Computer Science (EECS), effective Jan. 15.

“Karl’s exceptional interdisciplinary research combining electrical engineering, physics, and materials science, coupled with his experience working with industry and government organizations, makes him an ideal fit to head electrical engineering. I’m confident electrical engineering will continue to grow under his leadership,” says Anantha Chandrakasan, chief innovation and strategy officer, dean of engineering, and Vannevar Bush Professor of Electrical Engineering and Computer Science.

“Karl has made an incredible impact as a researcher and educator over his two decades in EECS. Students and faculty colleagues praise his thoughtful approach to teaching, and the care with which he oversaw the teaching labs in his prior role as undergraduate lab officer for the department. He will undoubtedly be an excellent leader, bringing his passion for education and collaborative spirit to this new role,” adds Daniel Huttenlocher, dean of the MIT Schwarzman College of Computing and the Henry Ellis Warren Professor of Electrical Engineering and Computer Science.

Berggren joins the leadership of EECS, which jointly reports to the MIT Schwarzman College of Computing and the School of Engineering. The largest academic department at MIT, EECS was reorganized in 2019 as part of the formation of the college into three overlapping sub-units in electrical engineering, computer science, and artificial intelligence and decision-making. The restructuring has enabled each of the three sub-units to concentrate on faculty recruitment, mentoring, promotion, academic programs, and community building in coordination with the others.

A member of the EECS faculty since 2003, Berggren has taught a range of subjects in the department, including Digital Communications, Circuits and Electronics, Fundamentals of Programming, Applied Quantum and Statistical Physics, Introduction to EECS via Interconnected Embedded Systems, Introduction to Quantum Systems Engineering, and Introduction to Nanofabrication. Before joining EECS, Berggren worked as a staff member at MIT Lincoln Laboratory for seven years. Berggren also maintains an active consulting practice and has experience working with industrial and government organizations.

Berggren’s current research focuses on superconductive circuits, electronic devices, single-photon detectors for quantum applications, and electron-optical systems. He heads the Quantum Nanostructures and Nanofabrication Group, which develops nanofabrication technology at the few-nanometer length scale. The group uses these technologies to push the envelope of what is possible with photonic and electrical devices, focusing on superconductive and free-electron devices.

Berggren has received numerous prestigious awards and honors throughout his career. Most recently, he was named an MIT MacVicar Fellow in 2024. Berggren is also a fellow of the AAAS, IEEE, and the Kavli Foundation, and a recipient of the 2015 Paul T. Forman Team Engineering Award from the Optical Society of America (now Optica). In 2016, he received a Bose Fellowship and was also a recipient of the EECS department’s Frank Quick Innovation Fellowship and the Burgess (’52) & Elizabeth Jamieson Award for Excellence in Teaching.

Berggren succeeds Joel Voldman, who has served as the inaugural electrical engineering faculty head since January 2020.

“Joel has been in leadership roles since 2018, when he was named associate department head of EECS. I am deeply grateful to him for his invaluable contributions to EECS since that time,” says Asu Ozdaglar, MathWorks Professor and head of EECS, who also serves as the deputy dean of the MIT Schwarzman College of Computing. “I look forward to working with Karl now and continuing along the amazing path we embarked on in 2019.”

Shreya Mogulothu is naturally curious. As a high school student in New Jersey, she was interested in mathematics and theoretical computer science (TCS). So, when her curiosity compelled her to learn more, she turned to MIT Open Learning’s online resources and completed the Paradox and Infinity course on MITx Online. 

“Coming from a math and TCS background, the idea of pushing against the limits of assumptions was really interesting,” says Mogulothu, now a junior at MIT. “I mean, who wouldn’t want to learn more about infinity?”

The class, taught by Agustín Rayo, professor of philosophy and the current dean of the School of Humanities, Arts, and Social Sciences, and David Balcarras, a former instructor in philosophy and fellow in the Digital Learning Lab at Open Learning, explores the intersection of math and philosophy and guides learners through thinking about paradoxes and open-ended problems, as well as the boundaries of theorizing and the limits of standard mathematical tools.

“We talked about taking regular assumptions about numbers and objects and pushing them to extremes,” Mogulothu says. “For example, what contradictions arise when you talk about an infinite set of things, like the infinite hats paradox?” 

The infinite hats paradox, also known as Bacon’s Puzzle, involves an infinite line of people, each wearing one of two colors of hats. The puzzle posits that each individual can see only the hat of the person in front of them and must guess the color of their own hat. The puzzle challenges students to identify if there is a strategy that can ensure the least number of incorrect answers and to consider how strategy may change if there is a finite number of people. Mogulothu was thrilled that a class like this was available to her even though she wasn’t yet affiliated with MIT. 

“My MITx experience was one of the reasons I came to MIT,” she says. “I really liked the course, and I was happy it was shared with people like me, who didn’t even go to the school. I thought that a place that encouraged even people outside of campus to learn like that would be a pretty good place to study.” 

Looking back at the course, Balcarras says, “Shreya may have been the most impressive student in our online community of approximately 3,900 learners and 100 verified learners. I cannot single out another student whose performance rivaled hers.”

Because of her excellent performance, Mogulothu was invited to submit her work to the 2021 MITx Philosophy Awards. She won. In fact, Balcarras remembers, both papers she wrote for the course would have won. They demonstrated, he says, “an unusually high degree of precision, formal acumen, and philosophical subtlety for a high school student.”

Completing the course and winning the award was rewarding, Mogulothu says. It motivated her to keep exploring new things as a high school student, and then as a new student enrolled at MIT.

She came to college thinking she would declare a major in math or computer science. But when she looked at the courses she was most interested in, she realized she should pursue a physics major. 

She has enjoyed the courses in her major, especially class STS.042J/8.225J (Einstein, Oppenheimer, Feynman: Physics in the 20th Century), taught by David Kaiser, the Germeshausen Professor of the History of Science and professor of physics. She took the course on campus, but it is also available on Open Learning’s MIT OpenCourseWare. As a student, she continues to use MIT Open Learning resources to check out courses and review syllabi as she plans her coursework. 

In summer 2024, Mogulothu did research on gravitational wave detection at PIER, the partnership between research center DESY and the University of Hamburg, in Hamburg, Germany. She wants to pursue a PhD in physics to keep researching, expanding her mind, and indulging the curiosity that led her to MITx in the first place. She encourages all learners to feel comfortable and confident trying something entirely new. 

“I went into the Paradox and Infinity course thinking, ‘yeah, math is cool, computer science is cool,’” she says. “But, actually taking the course and learning about things you don’t even expect to exist is really powerful. It increases your curiosity and is super rewarding to stick with something and realize how much you can learn and grow.”  

Advisors are meant to guide students academically, supporting their research and career objectives. For MIT graduate students, the Committed to Caring program recognizes those who go above and beyond.

Professors Iain Stewart and Roberto Fernandez are two of the 2023-25 Committed to Caring cohort, supporting their students through self-doubt, developing a welcoming environment, and serving as a friend.

Iain Stewart: Supportive, equitable, and inclusive

Iain Stewart is the Otto and Jane Morningstar Professor of Science and former director of the Center for Theoretical Physics (CTP). His research interests center around nuclear and particle physics, where he develops and applies effective field theories to understand interactions between elementary particles and particularly strong interactions described by quantum chromodynamics.

Stewart shows faith in his students’ abilities even when they doubt themselves. According to his nominators, the field of physics, like many areas of intellectual pursuit, can attract a wide range of personalities, including those who are highly confident as well as those who may grapple with self-doubt. He explains concepts in a down-to-earth manner and does not make his students feel less than they are.

For his students, Stewart’s research group comes as a refreshing change. Stewart emphasizes that graduate school is for learning, and that one is not expected to know everything from the onset.

Stewart shows a great level of empathy and emotional support for his students. For example, one of the nominators recounted a story about preparing for their oral qualification exam. The student had temporarily suspended research, and another faculty member made a disparaging comment about the student’s grasp of their research. The student approached Stewart in distress.

"As your advisor,” Stewart reassured them, “I can tell you confidently that you know your research and you are doing well, and it’s totally OK to put it off for a while to prepare for the qual."

Stewart’s words gave the student a sense of relief and validation, reminding them that progress is a journey, not a race, and that taking time to prepare thoughtfully was both wise and necessary.

Always emphasizing positivity in his feedback, Stewart reminds advisees of their achievements and progress, helping them develop a more optimistic mindset. Stewart’s mentorship style recognizes individual student needs, a trait that his students find uncommon. His research group flourishes due to this approach, and a large number of his graduate and postdoc students have achieved great success.

During his six years as director, Stewart has made significant contributions to the CTP. He has improved the culture and demographics due to strong and inclusive leadership. In particular, a noteworthy number of women have joined the CTP.

In his own research group, a large number of international and female students have found a place, which is uncommon for groups in theoretical physics. Currently, three out of seven group members are female in a field where fewer than 10 percent are women.

Stewart’s nominators believe that given the number of women he has mentored in his career, he is single-handedly contributing to improving the diversity in his field. His nominators say he supports diverse backgrounds, and financially supports and encourages participation for marginalized groups.

Roberto Fernandez: Professor and friend

Roberto Fernandez is the William F. Pounds Professor of Organization Studies at the MIT Sloan School of Management as well as the co-director of the Economic Sociology PhD Program. His research focuses on organizations, social networks, and race and gender stratification. He has extensive experience doing field research in organizations, and he currently focuses on the organizational processes surrounding the hiring of new talent.

Fernandez describes himself as a “full-service professor.” He tries to attend to differing needs and circumstances of students and the situations they find themselves in, offering advice and consolation.

Fernandez is very understanding of his students, and is happy to speak to them about academic and personal problems alike. He acknowledges that each student comes from a different background with individual experience, and Fernandez attempts to accommodate each one in an ideal manner.

He advises in a way that respects a student’s personal life, but still expects a reasonable amount of produced work that motivates the student, allows for them to excel, and keeps them to a high standard.

Fernandez says, “It is just my sense of duty to pay forward how my mentors treated me. I feel like I would dishonor their work if I were not to pass it on.”

A nominator shared that Fernandez serves as both a professor and a friend. He has gone out of his way to check in and chat with them. They said that Fernandez is the only professor who has taken the time to truly get to know their story, and Fernandez speaks to students like an equal.

The nominator noted that many people at MIT enjoy a high level of privilege. Despite the differences in their circumstances, however, the nominator feels comfortable talking to Fernandez.

Happily, the professor continued to touch base with the nominator long after their class had finished, and he is the one person who really made them feel like MIT was their home. This experience stood out as unique for the nominator, and played a large role in their experience.

In addition to providing genuine connections, Fernandez advises incoming graduate students about the need for a mindset shift. Graduate school is not like undergrad. Being an excellent student is necessary, but it is not sufficient to succeed in a PhD program. Excellent undergraduate students are consumers of knowledge; on the other hand, excellent graduate students are producers of knowledge.

The nominator enthused, “[Fernandez] really went above and beyond, and this means a lot.”

MIT philosopher Sally Haslanger has been named the 2024 recipient of the prestigious Philip L. Quinn Prize from the American Philosophical Association (APA).

The award recognizes Haslanger’s lifelong contributions to philosophy and philosophers. Haslanger, the Ford Professor of Philosophy and Women’s and Gender Studies, says she is deeply honored by the recognition.

“So many philosophers I deeply respect have come before me as awardees, including Judith Jarvis Thomson, my former colleague and lifelong inspiration,” Haslanger says. “Judy and I both were deeply engaged in doing metaphysics with an eye toward the moral/political domain. Both of us were committed feminists in a time when it was not professionally easy. Both of us believed in the power of institutions, such as the APA and the American Association of University Professors (AAUP), to sustain a flourishing intellectual community. Both of us have demanded that institutions we are part of abide by their values.”

Haslanger joined the MIT faculty in 1998.

Her research features explorations of the social construction of categories like gender, race, and the family; social explanation and social structure; and topics in feminist epistemology. She has also published in metaphysics and critical race theory. Broadly speaking, her work links issues of social justice with contemporary work in epistemology, metaphysics, philosophy of language, and philosophy of mind.

Her book, “Resisting Reality: Social Construction and Social Critique” (Oxford University Press, 2012), was awarded the Joseph B. Gittler prize for outstanding work in the philosophy of social science. She also co-authored “What is Race: Four Philosophical Views” (Oxford University Press, 2019). Her current book, “Doing Justice to the Social” (under contract with Oxford University Press), develops an account of social practices and structures, emphasizing their materiality, the role of ideology, and potential grounds for critique. She continues to document and ameliorate the underrepresentation of women and other minorities in philosophy.

Haslanger, a former president of the Eastern Division of the APA, singles out the collaborative nature of the field while also celebrating her peers’ recognition, noting her work is “inspired, nourished, and scaffolded by others.”

“Judy was a notable inspiration (and a clear example of how hard such work can be), but there are so many others who have been on this journey with me and kept me going, including feminist colleagues across the country and abroad, graduate students, staff members, and allies from many different disciplines and professions,” Haslanger says.

Awarded annually since 2007, the Quinn Prize honors the memory of Philip L. Quinn, a noted philosopher from the University of Notre Dame who served as president of the APA Central Division for many years. The prize carries a $2,500 award and an engraved plaque.

Kieran Setiya, the Peter de Florez Professor of Philosophy and head of the Department of Linguistics and Philosophy, says Haslanger has played a “transformative role in philosophy.”

“Sally’s influence on the field has been vast. Bridging a deep divide, she has brought critical social theory into conversation with analytic philosophy, arguing for an account of social structures and practices that does justice to their materiality,” Setiya says. “This work earned her a Guggenheim Fellowship as well as membership in the American Academy of Arts and Sciences, along with invitations to give lectures named after canonical philosophers past and present: Wittgenstein, Benjamin, Hempel, Kant, Spinoza, and others.”

Setiya noted Haslanger’s substantial contributions to the field, including her role in founding the Philosophy in an Inclusive Key Summer Institute (PIKSI) in Boston, which for 10 years has brought diverse undergraduates to MIT to show them that graduate study in philosophy is a meaningful option for them and to mentor them as they apply to graduate school.

“As Sally’s colleague, I am in awe not just of her extraordinary philosophical and professional achievements, but of her integrity and the seemingly limitless energy she invests in her students, in the Philosophy Section, in MIT, in the profession, and in fighting for social justice in the world from which academia is inextricable,” Setiya adds.

Three MIT students — Yutao Gong, Brandon Man, and Andrii Zahorodnii — have been awarded 2025 Schwarzman Scholarships and will join the program’s 10th cohort to pursue a master’s degree in global affairs at Tsinghua University in Beijing, China.

The MIT students were selected from a pool of over 5,000 applicants. This year’s class of 150 scholars represents 38 countries and 105 universities from around the world.

The Schwarzman Scholars program aims to develop leadership skills and deepen understanding of China’s changing role in the world. The fully funded one-year master’s program at Tsinghua University emphasizes leadership, global affairs, and China. Scholars also gain exposure to China through mentoring, internships, and experiential learning.

MIT’s Schwarzman Scholar applicants receive guidance and mentorship from the distinguished fellowships team in Career Advising and Professional Development and the Presidential Committee on Distinguished Fellowships.

Yutao Gong will graduate this spring from the Leaders for Global Operations program at the MIT Sloan School of Management, earning a dual MBA and a MS degree in civil and environmental engineering with a focus on manufacturing and operations. Gong, who hails from Shanghai, China, has academic, work, and social engagement experiences in China, the United States, Jordan, and Denmark. She was previously a consultant at Boston Consulting Group working on manufacturing, agriculture, sustainability, and renewable energy-related projects, and spent two years in Chicago and one year in Greater China as a global ambassador. Gong graduated magna cum laude from Duke University with double majors in environmental science and statistics, where she organized the Duke China-U.S. Summit.

Brandon Man, from Canada and Hong Kong, is a master’s student in the Department of Mechanical Engineering at MIT, where he studies generative artificial intelligence (genAI) for engineering design. Previously, he graduated from Cornell University magna cum laude with honors in computer science. With a wealth of experience in robotics — from assistive robots to next-generation spacesuits for NASA to Tencent’s robot dog, Max — he is now a co-founder of Sequestor, a genAI-powered data aggregation platform that enables carbon credit investors to perform faster due diligence. His goal is to bridge the best practices of the Eastern and Western tech worlds.

Andrii Zahorodnii, from Ukraine, will graduate this spring with a bachelor of science and a master of engineering degree in computer science and cognitive sciences. An engineer as well as a neuroscientist, he has conducted research at MIT with Professor Guangyu Robert Yang’s MetaConscious Group and the Fiete Lab. Zahorodnii is passionate about using AI to uncover insights into human cognition, leading to more-informed, empathetic, and effective global decision-making and policy. Besides driving the exchange of ideas as a TEDxMIT organizer, he strives to empower and inspire future leaders internationally and in Ukraine through the Ukraine Leadership and Technology Academy he founded.

With a more efficient method for artificial pollination, farmers in the future could grow fruits and vegetables inside multilevel warehouses, boosting yields while mitigating some of agriculture’s harmful impacts on the environment.

To help make this idea a reality, MIT researchers are developing robotic insects that could someday swarm out of mechanical hives to rapidly perform precise pollination. However, even the best bug-sized robots are no match for natural pollinators like bees when it comes to endurance, speed, and maneuverability.

Now, inspired by the anatomy of these natural pollinators, the researchers have overhauled their design to produce tiny, aerial robots that are far more agile and durable than prior versions.

The new bots can hover for about 1,000 seconds, which is more than 100 times longer than previously demonstrated. The robotic insect, which weighs less than a paperclip, can fly significantly faster than similar bots while completing acrobatic maneuvers like double aerial flips.

The revamped robot is designed to boost flight precision and agility while minimizing the mechanical stress on its artificial wing flexures, which enables faster maneuvers, increased endurance, and a longer lifespan.

The new design also has enough free space that the robot could carry tiny batteries or sensors, which could enable it to fly on its own outside the lab.

“The amount of flight we demonstrated in this paper is probably longer than the entire amount of flight our field has been able to accumulate with these robotic insects. With the improved lifespan and precision of this robot, we are getting closer to some very exciting applications, like assisted pollination,” says Kevin Chen, an associate professor in the Department of Electrical Engineering and Computer Science (EECS), head of the Soft and Micro Robotics Laboratory within the Research Laboratory of Electronics (RLE), and the senior author of an open-access paper on the new design.

Chen is joined on the paper by co-lead authors Suhan Kim and Yi-Hsuan Hsiao, who are EECS graduate students; as well as EECS graduate student Zhijian Ren and summer visiting student Jiashu Huang. The research appears today in Science Robotics.

Boosting performance

Prior versions of the robotic insect were composed of four identical units, each with two wings, combined into a rectangular device about the size of a microcassette.

“But there is no insect that has eight wings. In our old design, the performance of each individual unit was always better than the assembled robot,” Chen says.

This performance drop was partly caused by the arrangement of the wings, which would blow air into each other when flapping, reducing the lift forces they could generate.

The new design chops the robot in half. Each of the four identical units now has one flapping wing pointing away from the robot’s center, stabilizing the wings and boosting their lift forces. With half as many wings, this design also frees up space so the robot could carry electronics.

In addition, the researchers created more complex transmissions that connect the wings to the actuators, or artificial muscles, that flap them. These durable transmissions, which required the design of longer wing hinges, reduce the mechanical strain that limited the endurance of past versions.

“Compared to the old robot, we can now generate control torque three times larger than before, which is why we can do very sophisticated and very accurate path-finding flights,” Chen says.

Yet even with these design innovations, there is still a gap between the best robotic insects and the real thing. For instance, a bee has only two wings, yet it can perform rapid and highly controlled motions.

“The wings of bees are finely controlled by a very sophisticated set of muscles. That level of fine-tuning is something that truly intrigues us, but we have not yet been able to replicate,” he says.

Less strain, more force

The motion of the robot’s wings is driven by artificial muscles. These tiny, soft actuators are made from layers of elastomer sandwiched between two very thin carbon nanotube electrodes and then rolled into a squishy cylinder. The actuators rapidly compress and elongate, generating mechanical force that flaps the wings.

In previous designs, when the actuator’s movements reach the extremely high frequencies needed for flight, the devices often start buckling. That reduces the power and efficiency of the robot. The new transmissions inhibit this bending-buckling motion, which reduces the strain on the artificial muscles and enables them to apply more force to flap the wings.

Another new design involves a long wing hinge that reduces torsional stress experienced during the flapping-wing motion. Fabricating the hinge, which is about 2 centimeters long but just 200 microns in diameter, was among their greatest challenges.

“If you have even a tiny alignment issue during the fabrication process, the wing hinge will be slanted instead of rectangular, which affects the wing kinematics,” Chen says.

After many attempts, the researchers perfected a multistep laser-cutting process that enabled them to precisely fabricate each wing hinge.

With all four units in place, the new robotic insect can hover for more than 1,000 seconds, which equates to almost 17 minutes, without showing any degradation of flight precision.

“When my student Nemo was performing that flight, he said it was the slowest 1,000 seconds he had spent in his entire life. The experiment was extremely nerve-racking,” Chen says.

The new robot also reached an average speed of 35 centimeters per second, the fastest flight researchers have reported, while performing body rolls and double flips. It can even precisely track a trajectory that spells M-I-T.

“At the end of the day, we’ve shown flight that is 100 times longer than anyone else in the field has been able to do, so this is an extremely exciting result,” he says.

From here, Chen and his students want to see how far they can push this new design, with the goal of achieving flight for longer than 10,000 seconds.

They also want to improve the precision of the robots so they could land and take off from the center of a flower. In the long run, the researchers hope to install tiny batteries and sensors onto the aerial robots so they could fly and navigate outside the lab.

“This new robot platform is a major result from our group and leads to many exciting directions. For example, incorporating sensors, batteries, and computing capabilities on this robot will be a central focus in the next three to five years,” Chen says.

This research is funded, in part, by the U.S. National Science Foundation and a Mathworks Fellowship.

Nearly 50 years ago, neuroscientists discovered cells within the brain’s hippocampus that store memories of specific locations. These cells also play an important role in storing memories of events, known as episodic memories. While the mechanism of how place cells encode spatial memory has been well-characterized, it has remained a puzzle how they encode episodic memories.

A new model developed by MIT researchers explains how those place cells can be recruited to form episodic memories, even when there’s no spatial component. According to this model, place cells, along with grid cells found in the entorhinal cortex, act as a scaffold that can be used to anchor memories as a linked series.

“This model is a first-draft model of the entorhinal-hippocampal episodic memory circuit. It’s a foundation to build on to understand the nature of episodic memory. That’s the thing I’m really excited about,” says Ila Fiete, a professor of brain and cognitive sciences at MIT, a member of MIT’s McGovern Institute for Brain Research, and the senior author of the new study.

The model accurately replicates several features of biological memory systems, including the large storage capacity, gradual degradation of older memories, and the ability of people who compete in memory competitions to store enormous amounts of information in “memory palaces.”

MIT Research Scientist Sarthak Chandra and Sugandha Sharma PhD ’24 are the lead authors of the study, which appears today in Nature. Rishidev Chaudhuri, an assistant professor at the University of California at Davis, is also an author of the paper.

An index of memories

To encode spatial memory, place cells in the hippocampus work closely with grid cells — a special type of neuron that fires at many different locations, arranged geometrically in a regular pattern of repeating triangles. Together, a population of grid cells forms a lattice of triangles representing a physical space.

In addition to helping us recall places where we’ve been, these hippocampal-entorhinal circuits also help us navigate new locations. From human patients, it’s known that these circuits are also critical for forming episodic memories, which might have a spatial component but mainly consist of events, such as how you celebrated your last birthday or what you had for lunch yesterday.

“The same hippocampal and entorhinal circuits are used not just for spatial memory, but also for general episodic memory,” Fiete says. “The question you can ask is what is the connection between spatial and episodic memory that makes them live in the same circuit?”

Two hypotheses have been proposed to account for this overlap in function. One is that the circuit is specialized to store spatial memories because those types of memories — remembering where food was located or where predators were seen — are important to survival. Under this hypothesis, this circuit encodes episodic memories as a byproduct of spatial memory.

An alternative hypothesis suggests that the circuit is specialized to store episodic memories, but also encodes spatial memory because location is one aspect of many episodic memories.

In this work, Fiete and her colleagues proposed a third option: that the peculiar tiling structure of grid cells and their interactions with hippocampus are equally important for both types of memory — episodic and spatial. To develop their new model, they built on computational models that her lab has been developing over the past decade, which mimic how grid cells encode spatial information.

“We reached the point where I felt like we understood on some level the mechanisms of the grid cell circuit, so it felt like the time to try to understand the interactions between the grid cells and the larger circuit that includes the hippocampus,” Fiete says.

In the new model, the researchers hypothesized that grid cells interacting with hippocampal cells can act as a scaffold for storing either spatial or episodic memory. Each activation pattern within the grid defines a “well,” and these wells are spaced out at regular intervals. The wells don’t store the content of a specific memory, but each one acts as a pointer to a specific memory, which is stored in the synapses between the hippocampus and the sensory cortex.

When the memory is triggered later from fragmentary pieces, grid and hippocampal cell interactions drive the circuit state into the nearest well, and the state at the bottom of the well connects to the appropriate part of the sensory cortex to fill in the details of the memory. The sensory cortex is much larger than the hippocampus and can store vast amounts of memory.

“Conceptually, we can think about the hippocampus as a pointer network. It’s like an index that can be pattern-completed from a partial input, and that index then points toward sensory cortex, where those inputs were experienced in the first place,” Fiete says. “The scaffold doesn’t contain the content, it only contains this index of abstract scaffold states.”

Furthermore, events that occur in sequence can be linked together: Each well in the grid cell-hippocampal network efficiently stores the information that is needed to activate the next well, allowing memories to be recalled in the right order.

Modeling memory cliffs and palaces

The researchers’ new model replicates several memory-related phenomena much more accurately than existing models that are based on Hopfield networks — a type of neural network that can store and recall patterns.

While Hopfield networks offer insight into how memories can be formed by strengthening connections between neurons, they don’t perfectly model how biological memory works. In Hopfield models, every memory is recalled in perfect detail until capacity is reached. At that point, no new memories can form, and worse, attempting to add more memories erases all prior ones. This “memory cliff” doesn’t accurately mimic what happens in the biological brain, which tends to gradually forget the details of older memories while new ones are continually added.

The new MIT model captures findings from decades of recordings of grid and hippocampal cells in rodents made as the animals explore and forage in various environments. It also helps to explain the underlying mechanisms for a memorization strategy known as a memory palace. One of the tasks in memory competitions is to memorize the shuffled sequence of cards in one or several card decks. They usually do this by assigning each card to a particular spot in a memory palace — a memory of a childhood home or other environment they know well. When they need to recall the cards, they mentally stroll through the house, visualizing each card in its spot as they go along. Counterintuitively, adding the memory burden of associating cards with locations makes recall stronger and more reliable.

The MIT team’s computational model was able to perform such tasks very well, suggesting that memory palaces take advantage of the memory circuit’s own strategy of associating inputs with a scaffold in the hippocampus, but one level down: Long-acquired memories reconstructed in the larger sensory cortex can now be pressed into service as a scaffold for new memories. This allows for the storage and recall of many more items in a sequence than would otherwise be possible.

The researchers now plan to build on their model to explore how episodic memories could become converted to cortical “semantic” memory, or the memory of facts dissociated from the specific context in which they were acquired (for example, Paris is the capital of France), how episodes are defined, and how brain-like memory models could be integrated into modern machine learning.

The research was funded by the U.S. Office of Naval Research, the National Science Foundation under the Robust Intelligence program, the ARO-MURI award, the Simons Foundation, and the K. Lisa Yang ICoN Center.

Quantum computing promises to solve complex problems exponentially faster than a classical computer, by using the principles of quantum mechanics to encode and manipulate information in quantum bits (qubits).

Qubits are the building blocks of a quantum computer. One challenge to scaling, however, is that qubits are highly sensitive to background noise and control imperfections, which introduce errors into the quantum operations and ultimately limit the complexity and duration of a quantum algorithm. To improve the situation, MIT researchers and researchers worldwide have continually focused on improving qubit performance. 

In new work, using a superconducting qubit called fluxonium, MIT researchers in the Department of Physics, the Research Laboratory of Electronics (RLE), and the Department of Electrical Engineering and Computer Science (EECS) developed two new control techniques to achieve a world-record single-qubit fidelity of 99.998 percent. This result complements then-MIT researcher Leon Ding’s demonstration last year of a 99.92 percent two-qubit gate fidelity. 

The paper’s senior authors are David Rower PhD ’24, a recent physics postdoc in MIT’s Engineering Quantum Systems (EQuS) group and now a research scientist at the Google Quantum AI laboratory; Leon Ding PhD ’23 from EQuS, now leading the Calibration team at Atlantic Quantum; and William D. Oliver, the Henry Ellis Warren Professor of EECS and professor of physics, leader of EQuS, director of the Center for Quantum Engineering, and RLE associate director. The paper recently appeared in the journal PRX Quantum.

Decoherence and counter-rotating errors

A major challenge with quantum computation is decoherence, a process by which qubits lose their quantum information. For platforms such as superconducting qubits, decoherence stands in the way of realizing higher-fidelity quantum gates.

Quantum computers need to achieve high gate fidelities in order to implement sustained computation through protocols like quantum error correction. The higher the gate fidelity, the easier it is to realize practical quantum computing.

MIT researchers are developing techniques to make quantum gates, the basic operations of a quantum computer, as fast as possible in order to reduce the impact of decoherence. However, as gates get faster, another type of error, arising from counter-rotating dynamics, can be introduced because of the way qubits are controlled using electromagnetic waves. 

Single-qubit gates are usually implemented with a resonant pulse, which induces Rabi oscillations between the qubit states. When the pulses are too fast, however, “Rabi gates” are not so consistent, due to unwanted errors from counter-rotating effects. The faster the gate, the more the counter-rotating error is manifest. For low-frequency qubits such as fluxonium, counter-rotating errors limit the fidelity of fast gates.

“Getting rid of these errors was a fun challenge for us,” says Rower. “Initially, Leon had the idea to utilize circularly polarized microwave drives, analogous to circularly polarized light, but realized by controlling the relative phase of charge and flux drives of a superconducting qubit. Such a circularly polarized drive would ideally be immune to counter-rotating errors.”

While Ding’s idea worked immediately, the fidelities achieved with circularly polarized drives were not as high as expected from coherence measurements.

“Eventually, we stumbled on a beautifully simple idea,” says Rower. “If we applied pulses at exactly the right times, we should be able to make counter-rotating errors consistent from pulse-to-pulse. This would make the counter-rotating errors correctable. Even better, they would be automatically accounted for with our usual Rabi gate calibrations!”

They called this idea “commensurate pulses,” since the pulses needed to be applied at times commensurate with intervals determined by the qubit frequency through its inverse, the time period. Commensurate pulses are defined simply by timing constraints and can be applied to a single linear qubit drive. In contrast, circularly polarized microwaves require two drives and some extra calibration.

“I had much fun developing the commensurate technique,” says Rower. “It was simple, we understood why it worked so well, and it should be portable to any qubit suffering from counter-rotating errors!”

“This project makes it clear that counter-rotating errors can be dealt with easily. This is a wonderful thing for low-frequency qubits such as fluxonium, which are looking more and more promising for quantum computing.”

Fluxonium’s promise

Fluxonium is a type of superconducting qubit made up of a capacitor and Josephson junction; unlike transmon qubits, however, fluxonium also includes a large “superinductor,” which by design helps protect the qubit from environmental noise. This results in performing logical operations, or gates, with greater accuracy.

Despite having higher coherence, however, fluxonium has a lower qubit frequency that is generally associated with proportionally longer gates.

“Here, we’ve demonstrated a gate that is among the fastest and highest-fidelity across all superconducting qubits,” says Ding. “Our experiments really show that fluxonium is a qubit that supports both interesting physical explorations and also absolutely delivers in terms of engineering performance.”

With further research, they hope to reveal new limitations and yield even faster and higher-fidelity gates.

“Counter-rotating dynamics have been understudied in the context of superconducting quantum computing because of how well the rotating-wave approximation holds in common scenarios,” says Ding. “Our paper shows how to precisely calibrate fast, low-frequency gates where the rotating-wave approximation does not hold.”

Physics and engineering team up

“This is a wonderful example of the type of work we like to do in EQuS, because it leverages fundamental concepts in both physics and electrical engineering to achieve a better outcome,” says Oliver. “It builds on our earlier work with non-adiabatic qubit control, applies it to a new qubit — fluxonium — and makes a beautiful connection with counter-rotating dynamics.”

The science and engineering teams enabled the high fidelity in two ways. First, the team demonstrated “commensurate” (synchronous) non-adiabatic control, which goes beyond the standard “rotating wave approximation” of standard Rabi approaches. This leverages ideas that won the 2023 Nobel Prize in Physics for ultrafast “attosecond” pulses of light.

Secondly, they demonstrated it using an analog to circularly polarized light. Rather than a physical electromagnetic field with a rotating polarization vector in real x-y space, they realized a synthetic version of circularly polarized light using the qubit’s x-y space, which in this case corresponds to its magnetic flux and electric charge.

The combination of a new take on an existing qubit design (fluxonium) and the application of advanced control methods applied to an understanding of the underlying physics enabled this result.

Platform-independent and requiring no additional calibration overhead, this work establishes straightforward strategies for mitigating counter-rotating effects from strong drives in circuit quantum electrodynamics and other platforms, which the researchers expect to be helpful in the effort to realize high-fidelity control for fault-tolerant quantum computing.

Adds Oliver, “With the recent announcement of Google’s Willow quantum chip that demonstrated quantum error correction beyond threshold for the first time, this is a timely result, as we have pushed performance even higher. Higher-performant qubits will lead to lower overhead requirements for implementing error correction.”  

Other researchers on the paper are RLE’s Helin Zhang, Max Hays, Patrick M. Harrington, Ilan T. Rosen, Simon Gustavsson, Kyle Serniak, Jeffrey A. Grover, and Junyoung An, who is also with EECS; and MIT Lincoln Laboratory’s Jeffrey M. Gertler, Thomas M. Hazard, Bethany M. Niedzielski, and Mollie E. Schwartz.

This research was funded, in part, by the U.S. Army Research Office, the U.S. Department of Energy Office of Science, National Quantum Information Science Research Centers, Co-design Center for Quantum Advantage, U.S. Air Force, the U.S. Office of the Director of National Intelligence, and the U.S. National Science Foundation.  

Angelina Wu has been taking Japanese classes at MIT since arriving as a first-year student.

“I have had such a wonderful experience learning the language, getting to know my classmates, and interacting with the Japanese community at MIT,” says Wu, now a senior majoring in computer science and engineering.

“It’s been an integral part of my MIT experience, supplementing my other technical skills and also giving me opportunities to meet many people outside my major that I likely wouldn’t have had otherwise. As a result, I feel like I get to understand a much broader, more complete version of MIT.”

Now, Wu is sharing her experience and giving back as a Global Languages Student Ambassador. At a recent Global Languages preregistration fair, Wu spoke with other students interested in pursuing Japanese studies.

“I could not be happier to help promote such an experience to curious students and the greater MIT community,” Wu says.

Global Language Student Ambassadors is a group of students who lead outreach efforts to help increase visibility for the program.

In addition to disseminating information and promotional materials to the MIT undergraduate community, student ambassadors are asked to organize and host informal gatherings for Global Languages students around themes related to language and cultural exploration to build community and provide opportunities for learning and fun outside of the classroom.

Global Languages director Per Urlaub isn’t surprised that the Student Ambassadors program is popular with both students and the MIT community.

“The Global Languages program brings people together,” he says. “Providing a caring learning environment and creating a sense of belonging are central to our mission.”

What’s also central to the Global Languages’ mission is centering students’ work and creating spaces in which language learning can help create connections across academic areas. Students who study languages may improve their understanding of the cultural facets that underlie communication across cultures and open new worlds.

“An engaging community that fosters a deep sense of belonging doesn’t just happen automatically,” Urlaub notes. “A stronger community elevates our students’ proficiency gains, and also makes language learning more meaningful and fun.”

Each student ambassador serves for a single academic year in their area of language focus. They work closely with MIT’s academic administrators to plan, communicate, and stage events.

“I love exploring the richness of the Arabic language, especially how it connects to my culture and heritage,” says Heba Hussein, a student ambassador studying Arabic and majoring in electrical science and engineering. “I believe that having a strong grasp of languages and cultural awareness will help me work effectively in diverse teams.”

Student ambassadors, alongside other language learners, discover how other languages, cultures, and countries can guide their communications with others while shaping how they understand the world.

“My Spanish courses at MIT have been a highlight of my college experience thus far — the opportunity to connect on a deeper level with other cultures and force myself out of my comfort zone in conversations is important to me,” says Katie Kempff, another student ambassador who is majoring in climate system science and engineering and Spanish.

“As a heritage speaker, learning Chinese has been a way for me to connect with my culture and my roots,” adds Zixuan Liu, a double major in biological engineering and biology, and a Chinese student ambassador, who says that as a heritage speaker, learning Chinese has been a way for her to connect with her culture and her roots.

“I would highly recommend diving into languages and culture at MIT, where the support and the community really enhances the experience,” Liu says.

Back in the old days — the really old days — the task of designing materials was laborious. Investigators, over the course of 1,000-plus years, tried to make gold by combining things like lead, mercury, and sulfur, mixed in what they hoped would be just the right proportions. Even famous scientists like Tycho Brahe, Robert Boyle, and Isaac Newton tried their hands at the fruitless endeavor we call alchemy.

Materials science has, of course, come a long way. For the past 150 years, researchers have had the benefit of the periodic table of elements to draw upon, which tells them that different elements have different properties, and one can’t magically transform into another. Moreover, in the past decade or so, machine learning tools have considerably boosted our capacity to determine the structure and physical properties of various molecules and substances. New research by a group led by Ju Li — the Tokyo Electric Power Company Professor of Nuclear Engineering at MIT and professor of materials science and engineering — offers the promise of a major leap in capabilities that can facilitate materials design. The results of their investigation are reported in a December 2024 issue of Nature Computational Science.

At present, most of the machine-learning models that are used to characterize molecular systems are based on density functional theory (DFT), which offers a quantum mechanical approach to determining the total energy of a molecule or crystal by looking at the electron density distribution — which is, basically, the average number of electrons located in a unit volume around each given point in space near the molecule. (Walter Kohn, who co-invented this theory 60 years ago, received a Nobel Prize in Chemistry for it in 1998.) While the method has been very successful, it has some drawbacks, according to Li: “First, the accuracy is not uniformly great. And, second, it only tells you one thing: the lowest total energy of the molecular system.”

“Couples therapy” to the rescue

His team is now relying on a different computational chemistry technique, also derived from quantum mechanics, known as coupled-cluster theory, or CCSD(T). “This is the gold standard of quantum chemistry,” Li comments. The results of CCSD(T) calculations are much more accurate than what you get from DFT calculations, and they can be as trustworthy as those currently obtainable from experiments. The problem is that carrying out these calculations on a computer is very slow, he says, “and the scaling is bad: If you double the number of electrons in the system, the computations become 100 times more expensive.” For that reason, CCSD(T) calculations have normally been limited to molecules with a small number of atoms — on the order of about 10. Anything much beyond that would simply take too long.

That’s where machine learning comes in. CCSD(T) calculations are first performed on conventional computers, and the results are then used to train a neural network with a novel architecture specially devised by Li and his colleagues. After training, the neural network can perform these same calculations much faster by taking advantage of approximation techniques. What’s more, their neural network model can extract much more information about a molecule than just its energy. “In previous work, people have used multiple different models to assess different properties,” says Hao Tang, an MIT PhD student in materials science and engineering. “Here we use just one model to evaluate all of these properties, which is why we call it a ‘multi-task’ approach.”

The “Multi-task Electronic Hamiltonian network,” or MEHnet, sheds light on a number of electronic properties, such as the dipole and quadrupole moments, electronic polarizability, and the optical excitation gap — the amount of energy needed to take an electron from the ground state to the lowest excited state. “The excitation gap affects the optical properties of materials,” Tang explains, “because it determines the frequency of light that can be absorbed by a molecule.” Another advantage of their CCSD-trained model is that it can reveal properties of not only ground states, but also excited states. The model can also predict the infrared absorption spectrum of a molecule related to its vibrational properties, where the vibrations of atoms within a molecule are coupled to each other, leading to various collective behaviors.

The strength of their approach owes a lot to the network architecture. Drawing on the work of MIT Assistant Professor Tess Smidt, the team is utilizing a so-called E(3)-equivariant graph neural network, says Tang, “in which the nodes represent atoms and the edges that connect the nodes represent the bonds between atoms. We also use customized algorithms that incorporate physics principles — related to how people calculate molecular properties in quantum mechanics — directly into our model.”

Testing, 1, 2 3

When tested on its analysis of known hydrocarbon molecules, the model of Li et al. outperformed DFT counterparts and closely matched experimental results taken from the published literature.

Qiang Zhu — a materials discovery specialist at the University of North Carolina at Charlotte (who was not part of this study) — is impressed by what’s been accomplished so far. “Their method enables effective training with a small dataset, while achieving superior accuracy and computational efficiency compared to existing models,” he says. “This is exciting work that illustrates the powerful synergy between computational chemistry and deep learning, offering fresh ideas for developing more accurate and scalable electronic structure methods.”

The MIT-based group applied their model first to small, nonmetallic elements — hydrogen, carbon, nitrogen, oxygen, and fluorine, from which organic compounds can be made — and has since moved on to examining heavier elements: silicon, phosphorus, sulfur, chlorine, and even platinum. After being trained on small molecules, the model can be generalized to bigger and bigger molecules. “Previously, most calculations were limited to analyzing hundreds of atoms with DFT and just tens of atoms with CCSD(T) calculations,” Li says. “Now we’re talking about handling thousands of atoms and, eventually, perhaps tens of thousands.”

For now, the researchers are still evaluating known molecules, but the model can be used to characterize molecules that haven’t been seen before, as well as to predict the properties of hypothetical materials that consist of different kinds of molecules. “The idea is to use our theoretical tools to pick out promising candidates, which satisfy a particular set of criteria, before suggesting them to an experimentalist to check out,” Tang says.

It’s all about the apps

Looking ahead, Zhu is optimistic about the possible applications. “This approach holds the potential for high-throughput molecular screening,” he says. “That’s a task where achieving chemical accuracy can be essential for identifying novel molecules and materials with desirable properties.”

Once they demonstrate the ability to analyze large molecules with perhaps tens of thousands of atoms, Li says, “we should be able to invent new polymers or materials” that might be used in drug design or in semiconductor devices. The examination of heavier transition metal elements could lead to the advent of new materials for batteries — presently an area of acute need.

The future, as Li sees it, is wide open. “It’s no longer about just one area,” he says. “Our ambition, ultimately, is to cover the whole periodic table with CCSD(T)-level accuracy, but at lower computational cost than DFT. This should enable us to solve a wide range of problems in chemistry, biology, and materials science. It’s hard to know, at present, just how wide that range might be.”

This work was supported by the Honda Research Institute. Hao Tang acknowledges support from the Mathworks Engineering Fellowship. The calculations in this work were performed, in part, on the Matlantis high-speed universal atomistic simulator, the Texas Advanced Computing Center, the MIT SuperCloud, and the National Energy Research Scientific Computing.

When sound waves reach the inner ear, neurons there pick up the vibrations and alert the brain. Encoded in their signals is a wealth of information that enables us to follow conversations, recognize familiar voices, appreciate music, and quickly locate a ringing phone or crying baby.

Neurons send signals by emitting spikes — brief changes in voltage that propagate along nerve fibers, also known as action potentials. Remarkably, auditory neurons can fire hundreds of spikes per second, and time their spikes with exquisite precision to match the oscillations of incoming sound waves.

With powerful new models of human hearing, scientists at MIT’s McGovern Institute for Brain Research have determined that this precise timing is vital for some of the most important ways we make sense of auditory information, including recognizing voices and localizing sounds.

The open-access findings, reported Dec. 4 in the journal Nature Communications, show how machine learning can help neuroscientists understand how the brain uses auditory information in the real world. MIT professor and McGovern investigator Josh McDermott, who led the research, explains that his team’s models better-equip researchers to study the consequences of different types of hearing impairment and devise more effective interventions.

Science of sound

The nervous system’s auditory signals are timed so precisely, researchers have long suspected that timing is important to our perception of sound. Sound waves oscillate at rates that determine their pitch: Low-pitched sounds travel in slow waves, whereas high-pitched sound waves oscillate more frequently. The auditory nerve that relays information from sound-detecting hair cells in the ear to the brain generates electrical spikes that correspond to the frequency of these oscillations. “The action potentials in an auditory nerve get fired at very particular points in time relative to the peaks in the stimulus waveform,” explains McDermott, who is also associate head of the MIT Department of Brain and Cognitive Sciences.

This relationship, known as phase-locking, requires neurons to time their spikes with sub-millisecond precision. But scientists haven’t really known how informative these temporal patterns are to the brain. Beyond being scientifically intriguing, McDermott says, the question has important clinical implications: “If you want to design a prosthesis that provides electrical signals to the brain to reproduce the function of the ear, it’s arguably pretty important to know what kinds of information in the normal ear actually matter,” he says.

This has been difficult to study experimentally; animal models can’t offer much insight into how the human brain extracts structure in language or music, and the auditory nerve is inaccessible for study in humans. So McDermott and graduate student Mark Saddler PhD ’24 turned to artificial neural networks.

Artificial hearing

Neuroscientists have long used computational models to explore how sensory information might be decoded by the brain, but until recent advances in computing power and machine learning methods, these models were limited to simulating simple tasks. “One of the problems with these prior models is that they’re often way too good,” says Saddler, who is now at the Technical University of Denmark. For example, a computational model tasked with identifying the higher pitch in a pair of simple tones is likely to perform better than people who are asked to do the same thing. “This is not the kind of task that we do every day in hearing,” Saddler points out. “The brain is not optimized to solve this very artificial task.” This mismatch limited the insights that could be drawn from this prior generation of models.

To better understand the brain, Saddler and McDermott wanted to challenge a hearing model to do things that people use their hearing for in the real world, like recognizing words and voices. That meant developing an artificial neural network to simulate the parts of the brain that receive input from the ear. The network was given input from some 32,000 simulated sound-detecting sensory neurons and then optimized for various real-world tasks.

The researchers showed that their model replicated human hearing well — better than any previous model of auditory behavior, McDermott says. In one test, the artificial neural network was asked to recognize words and voices within dozens of types of background noise, from the hum of an airplane cabin to enthusiastic applause. Under every condition, the model performed very similarly to humans.

When the team degraded the timing of the spikes in the simulated ear, however, their model could no longer match humans’ ability to recognize voices or identify the locations of sounds. For example, while McDermott’s team had previously shown that people use pitch to help them identify people’s voices, the model revealed that that this ability is lost without precisely timed signals. “You need quite precise spike timing in order to both account for human behavior and to perform well on the task,” Saddler says. That suggests that the brain uses precisely timed auditory signals because they aid these practical aspects of hearing.

The team’s findings demonstrate how artificial neural networks can help neuroscientists understand how the information extracted by the ear influences our perception of the world, both when hearing is intact and when it is impaired. “The ability to link patterns of firing in the auditory nerve with behavior opens a lot of doors,” McDermott says.

“Now that we have these models that link neural responses in the ear to auditory behavior, we can ask, ‘If we simulate different types of hearing loss, what effect is that going to have on our auditory abilities?’” McDermott says. “That will help us better diagnose hearing loss, and we think there are also extensions of that to help us design better hearing aids or cochlear implants.” For example, he says, “The cochlear implant is limited in various ways — it can do some things and not others. What’s the best way to set up that cochlear implant to enable you to mediate behaviors? You can, in principle, use the models to tell you that.”

MIT physicists and colleagues have for the first time measured the geometry, or shape, of electrons in solids at the quantum level. Scientists have long known how to measure the energies and velocities of electrons in crystalline materials, but until now, those systems’ quantum geometry could only be inferred theoretically, or sometimes not at all.

The work, reported in the Nov. 25 issue of Nature Physics, “opens new avenues for understanding and manipulating the quantum properties of materials,” says Riccardo Comin, MIT’s Class of 1947 Career Development Associate Professor of Physics and leader of the work.

“We’ve essentially developed a blueprint for obtaining some completely new information that couldn’t be obtained before,” says Comin, who is also affiliated with MIT’s Materials Research Laboratory and the Research Laboratory of Electronics.

The work could be applied to “any kind of quantum material, not just the one we worked with,” says Mingu Kang PhD ’23, first author of the Nature Physics paper who conducted the work as an MIT graduate student and who is now a Kavli Postdoctoral Fellow at Cornell University’s Laboratory of Atomic and Solid State Physics. 

Kang was also invited to write an accompanying research briefing on the work, including its implications, for the Nov. 25 issue of Nature Physics.

A weird world

In the weird world of quantum physics, an electron can be described as both a point in space and a wave-like shape. At the heart of the current work is a fundamental object known as a wave function that describes the latter. “You can think of it like a surface in a three-dimensional space,” says Comin.

There are different types of wave functions, ranging from the simple to the complex. Think of a ball. That is analogous to a simple, or trivial, wave function. Now picture a Mobius strip, the kind of structure explored by M.C. Escher in his art. That’s analogous to a complex, or nontrivial, wave function. And the quantum world is filled with materials composed of the latter.

But until now, the quantum geometry of wave functions could only be inferred theoretically, or sometimes not at all. And the property is becoming more and more important as physicists find more and more quantum materials with potential applications in everything from quantum computers to advanced electronic and magnetic devices.

The MIT team solved the problem using a technique called angle-resolved photoemission spectroscopy, or ARPES. Comin, Kang, and some of the same colleagues had used the technique in other research. For example, in 2022 they reported discovering the “secret sauce” behind exotic properties of a new quantum material known as a kagome metal. That work, too, appeared in Nature Physics. In the current work, the team adapted ARPES to measure the quantum geometry of a kagome metal.

Close collaborations

Kang stresses that the new ability to measure the quantum geometry of materials “comes from the close cooperation between theorists and experimentalists.”

The Covid-19 pandemic, too, had an impact. Kang, who is from South Korea, was based in that country during the pandemic. “That facilitated a collaboration with theorists in South Korea,” says Kang, an experimentalist.

The pandemic also led to an unusual opportunity for Comin. He traveled to Italy to help run the ARPES experiments at the Italian Light Source Elettra, a national laboratory. The lab was closed during the pandemic, but was starting to reopen when Comin arrived. He found himself alone, however, when Kang tested positive for Covid and couldn’t join him. So he inadvertently ran the experiments himself with the support of local scientists. “As a professor, I lead projects, but students and postdocs actually carry out the work. So this is basically the last study where I actually contributed to the experiments themselves,” he says with a smile.

In addition to Kang and Comin, additional authors of the Nature Physics paper are Sunje Kim of Seoul National University (Kim is a co-first author with Kang); Paul M. Neves, a graduate student in the MIT Department of Physics; Linda Ye of Stanford University; Junseo Jung of Seoul National University; Denny Puntel of the University of Trieste; Federico Mazzola of Consiglio Nazionale delle Ricerche and Ca’ Foscari University of Venice; Shiang Fang of Google DeepMind; Chris Jozwiak, Aaron Bostwick, and Eli Rotenberg of Lawrence Berkeley National Laboratory; Jun Fuji and Ivana Vobornik of Consiglio Nazionale delle Ricerche; Jae-Hoon Park of Max Planck POSTECH/Korea Research Initiative and Pohang University of Science and Technology; Joseph G. Checkelsky, associate professor of physics at MIT; and Bohm-Jung Yang of Seoul National University, who co-led the research project with Comin.

This work was funded by the U.S. Air Force Office of Scientific Research, the U.S. National Science Foundation, the Gordon and Betty Moore Foundation, the National Research Foundation of Korea, the Samsung Science and Technology Foundation, the U.S. Army Research Office, the U.S. Department of Energy Office of Science, the Heising-Simons Physics Research Fellow Program, the Tsinghua Education Foundation, the NFFA-MUR Italy Progetti Internazionali facility, the Samsung Foundation of Culture, and the Kavli Institute at Cornell.

Vijay Gadepally, a senior staff member at MIT Lincoln Laboratory, leads a number of projects at the Lincoln Laboratory Supercomputing Center (LLSC) to make computing platforms, and the artificial intelligence systems that run on them, more efficient. Here, Gadepally discusses the increasing use of generative AI in everyday tools, its hidden environmental impact, and some of the ways that Lincoln Laboratory and the greater AI community can reduce emissions for a greener future.

Q: What trends are you seeing in terms of how generative AI is being used in computing?

A: Generative AI uses machine learning (ML) to create new content, like images and text, based on data that is inputted into the ML system. At the LLSC we design and build some of the largest academic computing platforms in the world, and over the past few years we've seen an explosion in the number of projects that need access to high-performance computing for generative AI. We're also seeing how generative AI is changing all sorts of fields and domains — for example, ChatGPT is already influencing the classroom and the workplace faster than regulations can seem to keep up.

We can imagine all sorts of uses for generative AI within the next decade or so, like powering highly capable virtual assistants, developing new drugs and materials, and even improving our understanding of basic science. We can't predict everything that generative AI will be used for, but I can certainly say that with more and more complex algorithms, their compute, energy, and climate impact will continue to grow very quickly.

Q: What strategies is the LLSC using to mitigate this climate impact?

A: We're always looking for ways to make computing more efficient, as doing so helps our data center make the most of its resources and allows our scientific colleagues to push their fields forward in as efficient a manner as possible.

As one example, we've been reducing the amount of power our hardware consumes by making simple changes, similar to dimming or turning off lights when you leave a room. In one experiment, we reduced the energy consumption of a group of graphics processing units by 20 percent to 30 percent, with minimal impact on their performance, by enforcing a power cap. This technique also lowered the hardware operating temperatures, making the GPUs easier to cool and longer lasting.

Another strategy is changing our behavior to be more climate-aware. At home, some of us might choose to use renewable energy sources or intelligent scheduling. We are using similar techniques at the LLSC — such as training AI models when temperatures are cooler, or when local grid energy demand is low.

We also realized that a lot of the energy spent on computing is often wasted, like how a water leak increases your bill but without any benefits to your home. We developed some new techniques that allow us to monitor computing workloads as they are running and then terminate those that are unlikely to yield good results. Surprisingly, in a number of cases we found that the majority of computations could be terminated early without compromising the end result.

Q: What's an example of a project you've done that reduces the energy output of a generative AI program?

A: We recently built a climate-aware computer vision tool. Computer vision is a domain that's focused on applying AI to images; so, differentiating between cats and dogs in an image, correctly labeling objects within an image, or looking for components of interest within an image.

In our tool, we included real-time carbon telemetry, which produces information about how much carbon is being emitted by our local grid as a model is running. Depending on this information, our system will automatically switch to a more energy-efficient version of the model, which typically has fewer parameters, in times of high carbon intensity, or a much higher-fidelity version of the model in times of low carbon intensity.

By doing this, we saw a nearly 80 percent reduction in carbon emissions over a one- to two-day period. We recently extended this idea to other generative AI tasks such as text summarization and found the same results. Interestingly, the performance sometimes improved after using our technique!

Q: What can we do as consumers of generative AI to help mitigate its climate impact?

A: As consumers, we can ask our AI providers to offer greater transparency. For example, on Google Flights, I can see a variety of options that indicate a specific flight's carbon footprint. We should be getting similar kinds of measurements from generative AI tools so that we can make a conscious decision on which product or platform to use based on our priorities.

We can also make an effort to be more educated on generative AI emissions in general. Many of us are familiar with vehicle emissions, and it can help to talk about generative AI emissions in comparative terms. People may be surprised to know, for example, that one image-generation task is roughly equivalent to driving four miles in a gas car, or that it takes the same amount of energy to charge an electric car as it does to generate about 1,500 text summarizations.

There are many cases where customers would be happy to make a trade-off if they knew the trade-off's impact.

Q: What do you see for the future?

A: Mitigating the climate impact of generative AI is one of those problems that people all over the world are working on, and with a similar goal. We're doing a lot of work here at Lincoln Laboratory, but its only scratching at the surface. In the long term, data centers, AI developers, and energy grids will need to work together to provide "energy audits" to uncover other unique ways that we can improve computing efficiencies. We need more partnerships and more collaboration in order to forge ahead.

If you're interested in learning more, or collaborating with Lincoln Laboratory on these efforts, please contact Vijay Gadepally.