Transistors, the building blocks of modern electronics, are typically made of silicon. Because it’s a semiconductor, this material can control the flow of electricity in a circuit. But silicon has fundamental physical limits that restrict how compact and energy-efficient a transistor can be.

MIT researchers have now replaced silicon with a magnetic semiconductor, creating a magnetic transistor that could enable smaller, faster, and more energy-efficient circuits. The material’s magnetism strongly influences its electronic behavior, leading to more efficient control of the flow of electricity. 

The team used a novel magnetic material and an optimization process that reduces the material’s defects, which boosts the transistor’s performance.

The material’s unique magnetic properties also allow for transistors with built-in memory, which would simplify circuit design and unlock new applications for high-performance electronics.

“People have known about magnets for thousands of years, but there are very limited ways to incorporate magnetism into electronics. We have shown a new way to efficiently utilize magnetism that opens up a lot of possibilities for future applications and research,” says Chung-Tao Chou, an MIT graduate student in the departments of Electrical Engineering and Computer Science (EECS) and Physics, and co-lead author of a paper on this advance.

Chou is joined on the paper by co-lead author Eugene Park, a graduate student in the Department of Materials Science and Engineering (DMSE); Julian Klein, a DMSE research scientist; Josep Ingla-Aynes, a postdoc in the MIT Plasma Science and Fusion Center; Jagadeesh S. Moodera, a senior research scientist in the Department of Physics; and senior authors Frances Ross, TDK Professor in DMSE; and Luqiao Liu, an associate professor in EECS, and a member of the Research Laboratory of Electronics; as well as others at the University of Chemistry and Technology in Prague. The paper appears today in Physical Review Letters.

Overcoming the limits

In an electronic device, silicon semiconductor transistors act like tiny light switches that turn a circuit on and off, or amplify weak signals in a communication system. They do this using a small input voltage.

But a fundamental physical limit of silicon semiconductors prevents a transistor from operating below a certain voltage, which hinders its energy efficiency.

To make more efficient electronics, researchers have spent decades working toward magnetic transistors that utilize electron spin to control the flow of electricity. Electron spin is a fundamental property that enables electrons to behave like tiny magnets.

So far, scientists have mostly been limited to using certain magnetic materials. These lack the favorable electronic properties of semiconductors, constraining device performance.

“In this work, we combine magnetism and semiconductor physics to realize useful spintronic devices,” Liu says.

The researchers replace the silicon in the surface layer of a transistor with chromium sulfur bromide, a two-dimensional material that acts as a magnetic semiconductor.

Due to the material’s structure, researchers can switch between two magnetic states very cleanly. This makes it ideal for use in a transistor that smoothly switches between “on” and “off.”

“One of the biggest challenges we faced was finding the right material. We tried many other materials that didn’t work,” Chou says.

They discovered that changing these magnetic states modifies the material’s electronic properties, enabling low-energy operation. And unlike many other 2D materials, chromium sulfur bromide remains stable in air.

To make a transistor, the researchers pattern electrodes onto a silicon substrate, then carefully align and transfer the 2D material on top. They use tape to pick up a tiny piece of material, only a few tens of nanometers thick, and place it onto the substrate.

“A lot of researchers will use solvents or glue to do the transfer, but transistors require a very clean surface. We eliminate all those risks by simplifying this step,” Chou says.

Leveraging magnetism

This lack of contamination enables their device to outperform existing magnetic transistors. Most others can only create a weak magnetic effect, changing the flow of current by a few percent or less. Their new transistor can switch or amplify the electric current by a factor of 10.

They use an external magnetic field to change the magnetic state of the material, switching the transistor using significantly less energy than would usually be required.

The material also allows them to control the magnetic states with electric current. This is important because engineers cannot apply magnetic fields to individual transistors in an electronic device. They need to control each one electrically.

The material’s magnetic properties could also enable transistors with built-in memory, simplifying the design of logic or memory circuits.

A typical memory device has a magnetic cell to store information and a transistor to read it out. Their method can combine both into one magnetic transistor.

“Now, not only are transistors turning on and off, they are also remembering information. And because we can switch the transistor with greater magnitude, the signal is much stronger so we can read out the information faster, and in a much more reliable way,” Liu says.

Building on this demonstration, the researchers plan to further study the use of electrical current to control the device. They are also working to make their method scalable so they can fabricate arrays of transistors.

This research was supported, in part, by the Semiconductor Research Corporation, the U.S. Defense Advanced Research Projects Agency (DARPA), the U.S. National Science Foundation (NSF), the U.S. Department of Energy, the U.S. Army Research Office, and the Czech Ministry of Education, Youth, and Sports. The work was partially carried out at the MIT.nano facilities.

James J. Collins, the Termeer Professor of Medical Engineering and Science at MIT and faculty co-lead of the Abdul Latif Jameel Clinic for Machine Learning in Health, is embarking on a multidisciplinary research project that applies synthetic biology and generative artificial intelligence to the growing global threat of antimicrobial resistance (AMR).

The research project is sponsored by Jameel Research, part of the Abdul Latif Jameel International network. The initial three-year, $3 million research project in MIT’s Department of Biological Engineering and Institute of Medical Engineering and Science focuses on developing and validating programmable antibacterials against key pathogens.

AMR — driven by the overuse and misuse of antibiotics — has accelerated the rise of drug-resistant infections, while the development of new antibacterial tools has slowed. The impact is felt worldwide, especially in low- and middle-income countries, where limited diagnostic infrastructure causes delays or ineffective treatment.

The project centers on developing a new generation of targeted antibacterials using AI to design small proteins to disable specific bacterial functions. These designer molecules would be produced and delivered by engineered microbes, providing a more precise and adaptable approach than traditional antibiotics.

“This project reflects my belief that tackling AMR requires both bold scientific ideas and a pathway to real-world impact,” Collins says. “Jameel Research is keen to address this crisis by supporting innovative, translatable research at MIT.”

Mohammed Abdul Latif Jameel ’78, chair of Abdul Latif Jameel, says, “antimicrobial resistance is one of the most urgent challenges we face today, and addressing it will require ambitious science and sustained collaboration. We are pleased to support this new research, building on our long-standing relationship with MIT and our commitment to advancing research across the world, to strengthen global health and contribute to a more resilient future.”

The signals that drive many of the brain and body’s most essential functions — consciousness, sleep, breathing, heart rate, and motion — course through bundles of “white matter” fibers in the brainstem, but imaging systems so far have been unable to finely resolve these crucial neural cables. That has left researchers and doctors with little capability to assess how they are affected by trauma or neurodegeneration. 

In a new study, a team of MIT, Harvard University, and Massachusetts General Hospital researchers unveil AI-powered software capable of automatically segmenting eight distinct bundles in any diffusion MRI sequence.

In the open-access study, published Feb. 6 in the Proceedings of the National Academy Sciences, the research team led by MIT graduate student Mark Olchanyi reports that their BrainStem Bundle Tool (BSBT), which they’ve made publicly available, revealed distinct patterns of structural changes in patients with Parkinson’s disease, multiple sclerosis, and traumatic brain injury, and shed light on Alzheimer’s disease as well. Moreover, the study shows, BSBT retrospectively enabled tracking of bundle healing in a coma patient that reflected the patient’s seven-month road to recovery.

“The brainstem is a region of the brain that is essentially not explored because it is tough to image,” says Olchanyi, a doctoral candidate in MIT’s Medical Engineering and Medical Physics Program. “People don't really understand its makeup from an imaging perspective. We need to understand what the organization of the white matter is in humans and how this organization breaks down in certain disorders.”

Adds Professor Emery N. Brown, Olchanyi’s thesis supervisor and co-senior author of the study, “the brainstem is one of the body’s most important control centers. Mark’s algorithms are a significant contribution to imaging research and to our ability to the understand regulation of fundamental physiology. By enhancing our capacity to image the brainstem, he offers us new access to vital physiological functions such as control of the respiratory and cardiovascular systems, temperature regulation, how we stay awake during the day and how sleep at night.”

Brown is the Edward Hood Taplin Professor of Computational Neuroscience and Medical Engineering in The Picower Institute for Learning and Memory, the Institute for Medical Engineering and Science, and the Department of Brain and Cognitive Sciences at MIT. He is also an anesthesiologist at MGH and a professor at Harvard Medical School.

Building the algorithm

Diffusion MRI helps trace the long branches, or “axons,” that neurons extend to communicate with each other. Axons are typically clad in a sheath of fat called myelin, and water diffuses along the axons within the myelin, which is also called the brain’s “white matter.” Diffusion MRI can highlight this very directed displacement of water. But segmenting the distinct bundles of axons in the brainstem has proved challenging, because they are small and masked by flows of brain fluids and the motions produced by breathing and heart beats.

As part of his thesis work to better understand the neural mechanisms that underpin consciousness, Olchanyi wanted to develop an AI algorithm to overcome these obstacles. BSBT works by tracing fiber bundles that plunge into the brainstem from neighboring areas higher in the brain, such as the thalamus and the cerebellum, to produce a “probabilistic fiber map.” An artificial intelligence module called a “convolutional neural network” then combines the map with several channels of imaging information from within the brainstem to distinguish eight individual bundles.

To train the neural network to segment the bundles, Olchanyi “showed” it 30 live diffusion MRI scans from volunteers in the Human Connectome Project (HCP). The scans were manually annotated to teach the neural network how to identify the bundles. Then he validated BSBT by testing its output against “ground truth” dissections of post-mortem human brains where the bundles were well delineated via microscopic inspection or very slow but ultra-high-resolution imaging. After training, BSBT became proficient in automatically identifying the eight distinct fiber bundles in new scans.

In an experiment to test its consistency and reliability, Olchanyi tasked BSBT with finding the bundles in 40 volunteers who underwent separate scans two months apart. In each case, the tool was able to find the same bundles in the same patients in each of their two scans. Olchanyi also tested BSBT with multiple datasets (not just the HCP), and even inspected how each component of the neural network contributed to BSBT’s analysis by hobbling them one by one.

“We put the neural network through the wringer,” Olchanyi says. “We wanted to make sure that it’s actually doing these plausible segmentations and it is leveraging each of its individual components in a way that improves the accuracy.”

Potential novel biomarkers

Once the algorithm was properly trained and validated, the research team moved on to testing whether the ability to segment distinct fiber bundles in diffusion MRI scans could enable tracking of how each bundle’s volume and structure varied with disease or injury, creating a novel kind of biomarker. Although the brainstem has been difficult to examine in detail, many studies show that neurodegenerative diseases affect the brainstem, often early on in their progression.

Olchanyi, Brown and their co-authors applied BSBT to scores of datasets of diffusion MRI scans from patients with Alzheimer’s, Parkinson’s, MS, and traumatic brain injury (TBI). Patients were compared to controls and sometimes to themselves over time. In the scans, the tool measured bundle volume and “fractional anisotropy,” (FA) which tracks how much water is flowing along the myelinated axons versus how much is diffusing in other directions, a proxy for white matter structural integrity.

In each condition, the tool found consistent patterns of changes in the bundles. While only one bundle showed significant decline in Alzheimer’s, in Parkinson’s the tool revealed a reduction in FA in three of the eight bundles. It also revealed volume loss in another bundle in patients between a baseline scan and a two-year follow-up. Patients with MS showed their greatest FA reductions in four bundles and volume loss in three. Meanwhile, TBI patients didn’t show significant volume loss in any bundles, but FA reductions were apparent in the majority of bundles.

Testing in the study showed that BSBT proved more accurate than other classifier methods in discriminating between patients with health conditions versus controls.

BSBT, therefore, can be “a key adjunct that aids current diagnostic imaging methods by providing a fine-grained assessment of brainstem white matter structure and, in some cases, longitudinal information,” the authors wrote.

Finally, in the case of a 29-year-old man who suffered a severe TBI, Olchanyi applied BSBT to a scans taken during the man’s seven-month coma. The tool showed that the man’s brainstem bundles had been displaced, but not cut, and showed that over his coma, the lesions on the nerve bundles decreased by a factor of three in volume. As they healed, the bundles moved back into place as well.

The authors wrote that BSBT “has substantial prognostic potential by identifying preserved brainstem bundles that can facilitate coma recovery.”

The study’s other senior authors are Juan Eugenio Iglesias and Brian Edlow. Other co-authors are David Schreier, Jian Li, Chiara Maffei, Annabel Sorby-Adams, Hannah Kinney, Brian Healy, Holly Freeman, Jared Shless, Christophe Destrieux, and Hendry Tregidgo.

Funding for the study came from the National Institutes of Health, U.S. Department of Defense, James S. McDonnell Foundation, Rappaport Foundation, American SidS Institute, American Brain Foundation, American Academy of Neurology, Center for Integration of Medicine and Innovative Technology, Blueprint for Neuroscience Research, and Massachusetts Life Sciences Center.

3D bioprinting, in which living tissues are printed with cells mixed into soft hydrogels, or “bio-inks,” is widely used in the field of bioengineering for modeling or replacing the tissues in our bodies. The print quality and reproducibility of tissues, however, can face challenges. One of the most significant challenges is created simply by gravity — cells naturally sink to the bottom of the bioink-extruding printer syringe because the cells are heavier than the hydrogel around them.

“This cell settling, which becomes worse during the long print sessions required to print large tissues, leads to clogged nozzles, uneven cell distribution, and inconsistencies between printed tissues,” explains Ritu Raman, the Eugene Bell Career Development Professor of Tissue Engineering and assistant professor of mechanical engineering at MIT. “Existing solutions, such as manually stirring bioinks before loading them into the printer, or using passive mixers, cannot maintain uniformity once printing begins.”

In a study published Feb. 2 in the journal Device, Raman’s team introduces a new approach that aims to solve this core limitation by actively preventing cell sedimentation within bioinks during printing, allowing for more reliable and biologically consistent 3D printed tissues.

“Precise control over the bioink’s physical and biological properties is essential for recreating the structure and function of native tissues,” says Ferdows Afghah, a postdoc in mechanical engineering at MIT and lead author of the study.

“If we can print tissues that more closely mimic those in our bodies, we can use them as models to understand more about human diseases, or to test the safety and efficacy of new therapeutic drugs,” adds Raman. Such models could help researchers move away from techniques like animal testing, which supports recent interest from the U.S. Food and Drug Administration in developing faster, less expensive, and more informative new approaches to establish the safety and efficacy of new treatment paths.

“Eventually, we are working towards regenerative medicine applications such as replacing diseased or injured tissues in our bodies with 3D printed tissues that can help restore healthy function,” says Raman.

MagMix, a magnetically actuated mixer, is composed of two parts: a small magnetic propeller that fits inside the syringes used by bioprinters to deposit bioinks, layer by layer, into 3D tissues, and a permanent magnet attached to a motor that moves up and down near the syringe, controlling the movement of the propeller inside. Together, this compact system can be mounted onto any standard 3D bioprinter, keeping bioinks uniformly mixed during printing without changing the bioink formulation or interfering with the printer’s normal operation. To test the approach, the team used computer simulations to design the optimal mixing propeller geometry and speed and then validated its performance experimentally.

“Across multiple bioink types, MagMix prevented cell settling for more than 45 minutes of continuous printing, reducing clogging and preserving high cell viability,” says Raman. “Importantly, we showed that mixing speeds could be adjusted to balance effective homogenization for different bioinks while inducing minimal stress on the cells. As a proof-of-concept, we demonstrated that MagMix could be used to 3D print cells that could mature into muscle tissues over the course of several days.”

By maintaining uniform cell distribution throughout long or complex print jobs, MagMix enables the fabrication of high-quality tissues with more consistent biological function. Because the device is compact, low-cost, customizable, and easily integrated into existing 3D printers, it offers a broadly accessible solution for laboratories and industries working toward reproducible engineered tissues for applications in human health including disease modeling, drug screening, and regenerative medicine.

This work was supported, in part, by the Safety, Health, and Environmental Discovery Lab (SHED) at MIT, which provides infrastructure and interdisciplinary expertise to help translate biofabrication innovations from lab-scale demonstrations to scalable, reproducible applications.

“At the SHED, we focus on accelerating the translation of innovative methods into practical tools that researchers can reliably adopt,” says Tolga Durak, the SHED’s founding director. “MagMix is a strong example of how the right combination of technical infrastructure and interdisciplinary support can move biofabrication technologies toward scalable, real-world impact.”

The SHED’s involvement reflects a broader vision of strengthening technology pathways that enhance reproducibility and accessibility across engineering and the life sciences by providing equitable access to advanced equipment and fostering cross-disciplinary collaboration.

“As the field advances toward larger-scale and more standardized systems, integrated labs like SHED are essential for building sustainable capacity,” Durak adds. “Our goal is not only to enable discovery, but to ensure that new technologies can be reliably adopted and sustained over time.”

The team is also interested in non-medical applications of engineered tissues, such as using printed muscles to power safer and more efficient “biohybrid” robots.

The researchers believe this work can improve the reliability and scalability of 3D bioprinting, making the potential impacts on the field of 3D bioprinting and on human health significant. Their paper, “Advancing Bioink Homogeneity in Extrusion 3D Bioprinting with Active In Situ Magnetic Mixing,” is available now from the journal Device. 

Olympic figure skating looks effortless. Athletes sail across the ice, then soar into the air, spinning like a top, before landing on a single blade just 4-5 millimeters wide. To help figure skaters land quadruple axels, Salchows, Lutzes, and maybe even the elusive quintuple without looking the least bit stressed, Jerry Lu MFin ’24 developed an optical tracking system called OOFSkate that uses artificial intelligence to analyze video of a figure skater’s jump and make recommendations on how to improve. Lu, a former researcher at the MIT Sports Lab, has been aiding elite skaters on Team USA with their technical performance and will be working with NBC Sports during the 2026 Winter Olympics to help commentators and TV viewers make better sense of the complex scoring system in figure skating, snowboarding, and skiing. He’ll be applying AI technologies to explain nuanced judging decisions and demonstrate just how technically challenging these sports can be.

Meanwhile, Professor Anette “Peko” Hosoi, co-founder and faculty director of the MIT Sports Lab, is embarking on new research aimed at understanding how AI systems evaluate aesthetic performance in figure skating. Hosoi and Lu recently chatted with MIT News about applying AI to sports, whether AI systems could ever be used to judge Olympic figure skating, and when we might see a skater land a quint.

Q: Why apply AI to figure skating?

Lu: Skaters can always keep pushing, higher, faster, stronger. OOFSkate is all about helping skaters figure out a way to rotate a little bit faster in their jumps or jump a little bit higher. The system helps skaters catch things that perhaps could pass an eye test, but that might allow them to target some high-value areas of opportunity. The artistic side of skating is much harder to evaluate than the technical elements because it’s subjective.

To use mobile training app, you just need to take a video of an athlete’s jump, and it will spit out the physical metrics that drive how many rotations you can do. It tracks those metrics and builds in all of the other current elite and former elite athletes. You can see your data and then see, “This is how an Olympic champion did this element, perhaps I should try that.” You get the comparison and the automated classifier, which shows you if you did this trick at World Championships and it were judged by an international panel, this is approximately the grade of execution score they would give you.

Hosoi: There are a lot of AI tools that are coming online, especially things like pose estimators, where you can approximate skeletal configurations from video. The challenge with these pose estimators is that if you only have one camera angle, they do very well in the plane of the camera, but they do very poorly with depth. For example, if you’re trying to critique somebody’s form in fencing, and they’re moving toward the camera, you get very bad data. But with figure skating, Jerry has found one of the few areas where depth challenges don’t really matter. In figure skating, you need to understand: How high did this person jump, how many times did they go around, and how well did they land? None of those rely on depth. He’s found an application that pose estimators do really well, and that doesn’t pay a penalty for the things they do badly.

Q: Could you ever see a world in which AI is used to evaluate the artistic side of figure skating?

Hosoi: When it comes to AI and aesthetic evaluation, we have new work underway thanks to a MIT Human Insight Collaborative (MITHIC) grant. This work is in collaboration with Professor Arthur Bahr and IDSS graduate student Eric Liu. When you ask an AI platform for an aesthetic evaluation such as “What do you think of this painting?” it will respond with something that sounds like it came from a human. What we want to understand is, to get to that assessment, are the AIs going through the same sort of reasoning pathways or using the same intuitive concepts that humans go through to arrive at, “I like that painting,” or “I don’t like that painting”? Or are they just parrots? Are they just mimicking what they heard a person say? Or is there some concept map of aesthetic appeal? Figure skating is a perfect place to look for this map because skating is aesthetically judged. And there are numbers. You can’t go around a museum and find scores, “This painting is a 35.” But in skating, you’ve got the data.

That brings up another even more interesting question, which is the difference between novices and experts. It’s known that expert humans and novice humans will react differently to seeing the same thing. Somebody who is an expert judge may have a different opinion of a skating performance than a member of the general population. We’re trying to understand differences between reactions from experts, novices, and AI. Do these reactions have some common ground in where they are coming from, or is the AI coming from a different place than both the expert and the novice?

Lu: Figure skating is interesting because everybody working in the field of AI is trying to figure out AGI or artificial general intelligence and trying to build this extremely sound AI that replicates human beings. Working on applying AI to sports like figure skating helps us understand how humans think and approach judging. This has down-the-line impacts for AI research and companies that are developing AI models. By gaining a deeper understanding of how current state-of-the-art AI models work with these sports, and how you need to do training and fine-tuning of these models to make them work for specific sports, it helps you understand how AI needs to advance.

Q: What will you be watching for in the Milan Cortina Olympics figure skating competitions, now that you’ve been studying and working in this area? Do you think someone will land a quint?

Lu: For the winter games, I am working with NBC for the figure skating, ski, and snowboarding competitions to help them tell a data-driven story for the American people. The goal is to make these sports more relatable. Skating looks slow on television, but it’s not. Everything is supposed to look effortless. If it looks hard, you are probably going to get penalized. Skaters need to learn how to spin very fast, jump extremely high, float in the air, and land beautifully on one foot. The data we are gathering can help showcase how hard skating actually is, even though it is supposed to look easy.

I’m glad we are working in the Olympics sports realm because the world watches once every four years, and it is traditionally coaching-intensive and talent-driven sports, unlike a sport like baseball, where if you don’t have an elite-level optical tracking system you are not maximizing the value that you currently have. I’m glad we get to work with these Olympic sports and athletes and make an impact here.

Hosoi: I have always watched Olympic figure skating competitions, ever since I could turn on the TV. They’re always incredible. One of the things that I’m going to be practicing is identifying the jumps, which is very hard to do if you’re an amateur “judge.”

I have also done some back-of-the-envelope calculations to see if a quint is possible. I am now totally convinced it’s possible. We will see one in our lifetime, if not relatively soon. Not in this Olympics, but soon. When I saw we were so close on the quint, I thought, what about six? Can we do six rotations? Probably not. That’s where we start to come up against the limits of human physical capability. But five, I think, is in reach.

The 2026 Times Higher Education World University Ranking has ranked MIT first in three subject categories: Arts and Humanities, Business and Economics, and Social Sciences, repeating the Institute’s top spot in the same subjects in 2025.

The Times Higher Education World University Ranking is an annual publication of university rankings by Times Higher Education, a leading British education magazine. The subject rankings are based on 18 rigorous performance indicators categorized under five core pillars: teaching, research environment, research quality, industry, and international outlook.

Disciplines included in MIT’s top-ranked subjects are housed in the School of Humanities, Arts, and Social Sciences (SHASS), the School of Architecture and Planning (SA+P), and the MIT Sloan School of Management.

“SHASS is a vibrant crossroads of ideas, bringing together extraordinary people,” says Agustín Rayo, the Kenan Sahin Dean of SHASS. “These rankings reflect the strength of this remarkable community and MIT’s ongoing commitment to the humanities, arts, and social sciences.” 

“The human dimension is capital to our school's mission and programs, be they architecture, planning, media arts and sciences, or the arts, and whether at the scale of individuals, communities, or societies,” says Hashim Sarkis, dean of SA+P. “The acknowledgment and celebration of their centrality by the Times Higher Education only renews our deep commitment to human values.”

“MIT and MIT Sloan are providing students with an education that ensures they have the skills, experience, and problem-solving abilities they need in order to succeed in our world today,” says Richard M. Locke, the John C Head III Dean at the MIT Sloan School of Management. “It’s not just what we teach them, but how we teach them. The interdisciplinary nature of a school like MIT combines analytical reasoning skills, deep functional knowledge, and, at MIT Sloan, a hands-on management education that teaches students how to collaborate, lead teams, and navigate challenges, now and in the future."

The Arts and Humanities ranking evaluated 817 universities from 74 countries in the disciplines of languages; literature and linguistics; history; philosophy; theology; architecture; archaeology; and art, performing arts, and design. This is the second consecutive year MIT has earned the top spot in this subject.

The ranking for Business and Economics evaluated 1,067 institutions from 91 countries and territories across three core disciplines: business and management; accounting and finance; and economics and econometrics. This is the fifth consecutive year MIT has been ranked first in this subject.

The Social Sciences ranking evaluated 1,202 institutions from 104 countries and territories in the disciplines of political science and international studies, sociology, geography, communication and media studies, and anthropology. MIT claimed the top spot in this subject for the second consecutive year.

In other subjects, MIT was also named among the top universities, ranking third in Engineering and Life Sciences, and fourth in Computer Science and Physical Sciences. Overall, MIT ranked second in the Times Higher Education 2026 World University Ranking. 

Most materials have an inherent capacity to handle heat. Plastic, for instance, is typically a poor thermal conductor, whereas materials like marble move heat more efficiently. If you were to place one hand on a marble countertop and the other on a plastic cutting board, the marble would conduct more heat away from your hand, creating a colder sensation compared to the plastic.

Typically, a material’s thermal conductivity cannot be changed without re-manufacturing it. But MIT engineers have now found that a relatively common material can switch its thermal conductivity. Simply stretching the material quickly dials up its heat conductance, from a baseline similar to that of plastic to a higher capacity closer to that of marble. When the material springs back to its unstretched form, it returns to its plastic-like properties.

The thermally reversible material is an olefin block copolymer — a soft and flexible polymer that is used in a wide range of commercial products. The team found that when the material is quickly stretched, its ability to conduct heat more than doubles. This transition occurs within just 0.22 seconds, which is the fastest thermal switching that has been observed in any material.

This material could be used to engineer systems that adapt to changing temperatures in real time. For instance, switchable fibers could be woven into apparel that normally retains heat. When stretched, the fabric would instantly conduct heat away from a person’s body to cool them down. Similar fibers can be built into laptops and infrastructure to keep devices and buildings from overheating. The researchers are working on further optimizing the polymer and on engineering new materials with similar properties.

“We need cheap and abundant materials that can quickly adapt to environmental temperature changes,” says Svetlana Boriskina, principal research scientist in MIT’s Department of Mechanical Engineering. “Now that we’ve seen this thermal switching, this changes the direction where we can look for and build new adaptive materials.”

Boriskina and her colleagues have published their results in a study appearing today in the journal Advanced Materials. The study’s co-authors include Duo Xu, Buxuan Li, You Lyu, and Vivian Santamaria-Garcia of MIT, and Yuan Zhu of Southern University of Science and Technology in Shenzhen, China.

Elastic chains

The key to the new phenomenon is that when the material is stretched, its microscopic structures align in ways that suddenly allow heat to travel through easily, increasing the material’s thermal conductivity. In its unstretched state, the same microstructures are tangled and bunched, effectively blocking heat’s path.

As it happens, Boriskina and her colleagues didn’t set out to find a heat-switching material. They were initially looking for more sustainable alternatives to spandex, which is a synthetic fabric made from petroleum-based plastics that is traditionally difficult to recycle. As a potential replacement, the team was investigating fibers made from a different polymer known as polyethylene.

“Once we started working with the material, we realized it had other properties that were more interesting than the fact that it was elastic,” Boriskina says. “What makes polyethylene unique is it has this backbone of carbon atoms arranged along a simple chain. And carbon is a very good conductor of heat.”

The microstructure of most polymer materials, including polyethylene, contains many carbon chains. However, these chains exist in a messy, spaghetti-like tangle known as an amorphous phase. Despite the fact that carbon is a good heat conductor, the disordered arrangement of chains typically impedes heat flow. Polyethylene and most other polymers, therefore, generally have low thermal conductivity.

In previous work, MIT Professor Gang Chen and his collaborators found ways to untangle the mess of carbon chains and push polyethylene to shift from a disordered amorphous state to a more aligned, crystalline phase. This transition effectively straightened the carbon chains, providing clear highways for heat to flow through and increasing the material’s thermal conductivity. In those experiments however, the switch was permanent; once the material’s phase changed, it could not be reversed.

As Boriskina’s team explored polyethylene, they also considered other closely related materials, including olefin block copolymer (OBC). OBC is predominantly an amorphous material, made from highly tangled chains of carbon and hydrogen atoms. Scientists had therefore assumed that OBC would exhibit low thermal conductivity. If its conductance could be increased, it would likely be permanent, similar to polyethylene.

But when the team carried out experiments to test the elasticity of OBC, they found something quite different.

“As we stretched and released the material, we realized that its thermal conductivity was really high when it was stretched and lower when it was relaxed, over thousands of cycles,” says study co-author and MIT graduate student Duo Xu. “This switch was reversible, while the material stayed mostly amorphous. That was unexpected.”

A stretchy mess

The team then took a closer look at OBC, and how it might be changing as it was stretched. The researchers used a combination of X-ray and Raman spectroscopy to observe the material’s microscopic structure as they stretched and relaxed it repeatedly. They observed that, in its unstretched state, the material consists mainly of amorphous tangles of carbon chains, with just a few islands of ordered, crystalline domains scattered here and there. When stretched, the crystalline domains seemed to align and the amorphous tangles straightened out, similar to what Gang Chen observed in polyethylene.

However, rather than transitioning entirely into a crystalline phase, the straightened tangles stayed in their amorphous state. In this way, the team found that the tangles were able to switch back and forth, from straightened to bunched and back again, as the material was stretched and relaxed repeatedly.

“Our material is always in a mostly amorphous state; it never crystallizes under strain,” Xu notes. “So it leaves you this opportunity to go back and forth in thermal conductivity a thousand times. It’s very reversible.”

The team also found that this thermal switching happens extremely fast: The material’s thermal conductivity more than doubled within just 0.22 seconds of being stretched.

“The resulting difference in heat dissipation through this material is comparable to a tactile difference between touching a plastic cutting board versus a marble countertop,” Boriskina says.

She and her colleagues are now taking the results of their experiments and working them into models to see how they can tweak a material’s amorphous structure, to trigger an even bigger change when stretched.

“Our fibers can quickly react to dissipate heat, for electronics, fabrics, and building infrastructure.” Boriskina says. “If we could make further improvements to switch their thermal conductivity from that of plastic to that closer to diamond, it would have a huge industrial and societal impact.”

This research was supported, in part, by the U.S. Department of Energy, the Office of Naval Research Global via Tec de Monterrey, MIT Evergreen Graduate Innovation Fellowship, MathWorks MechE Graduate Fellowship, and the MIT-SUSTech Centers for Mechanical Engineering Research and Education, and carried out, in part, with the use of MIT.nano and ISN facilities.

A firm that wants to use a large language model (LLM) to summarize sales reports or triage customer inquiries can choose between hundreds of unique LLMs with dozens of model variations, each with slightly different performance.

To narrow down the choice, companies often rely on LLM ranking platforms, which gather user feedback on model interactions to rank the latest LLMs based on how they perform on certain tasks.

But MIT researchers found that a handful of user interactions can skew the results, leading someone to mistakenly believe one LLM is the ideal choice for a particular use case. Their study reveals that removing a tiny fraction of crowdsourced data can change which models are top-ranked.

They developed a fast method to test ranking platforms and determine whether they are susceptible to this problem. The evaluation technique identifies the individual votes most responsible for skewing the results so users can inspect these influential votes.

The researchers say this work underscores the need for more rigorous strategies to evaluate model rankings. While they didn’t focus on mitigation in this study, they provide suggestions that may improve the robustness of these platforms, such as gathering more detailed feedback to create the rankings.

The study also offers a word of warning to users who may rely on rankings when making decisions about LLMs that could have far-reaching and costly impacts on a business or organization.

“We were surprised that these ranking platforms were so sensitive to this problem. If it turns out the top-ranked LLM depends on only two or three pieces of user feedback out of tens of thousands, then one can’t assume the top-ranked LLM is going to be consistently outperforming all the other LLMs when it is deployed,” says Tamara Broderick, an associate professor in MIT’s Department of Electrical Engineering and Computer Science (EECS); a member of the Laboratory for Information and Decision Systems (LIDS) and the Institute for Data, Systems, and Society; an affiliate of the Computer Science and Artificial Intelligence Laboratory (CSAIL); and senior author of this study.

She is joined on the paper by lead authors and EECS graduate students Jenny Huang and Yunyi Shen as well as Dennis Wei, a senior research scientist at IBM Research. The study will be presented at the International Conference on Learning Representations.

Dropping data

While there are many types of LLM ranking platforms, the most popular variations ask users to submit a query to two models and pick which LLM provides the better response.

The platforms aggregate the results of these matchups to produce rankings that show which LLM performed best on certain tasks, such as coding or visual understanding.

By choosing a top-performing LLM, a user likely expects that model’s top ranking to generalize, meaning it should outperform other models on their similar, but not identical, application with a set of new data.

The MIT researchers previously studied generalization in areas like statistics and economics. That work revealed certain cases where dropping a small percentage of data can change a model’s results, indicating that those studies’ conclusions might not hold beyond their narrow setting.

The researchers wanted to see if the same analysis could be applied to LLM ranking platforms.

“At the end of the day, a user wants to know whether they are choosing the best LLM. If only a few prompts are driving this ranking, that suggests the ranking might not be the end-all-be-all,” Broderick says.

But it would be impossible to test the data-dropping phenomenon manually. For instance, one ranking they evaluated had more than 57,000 votes. Testing a data drop of 0.1 percent means removing each subset of 57 votes out of the 57,000, (there are more than 10194 subsets), and then recalculating the ranking.

Instead, the researchers developed an efficient approximation method, based on their prior work, and adapted it to fit LLM ranking systems.

“While we have theory to prove the approximation works under certain assumptions, the user doesn’t need to trust that. Our method tells the user the problematic data points at the end, so they can just drop those data points, re-run the analysis, and check to see if they get a change in the rankings,” she says.

Surprisingly sensitive

When the researchers applied their technique to popular ranking platforms, they were surprised to see how few data points they needed to drop to cause significant changes in the top LLMs. In one instance, removing just two votes out of more than 57,000, which is 0.0035 percent, changed which model is top-ranked.

A different ranking platform, which uses expert annotators and higher quality prompts, was more robust. Here, removing 83 out of 2,575 evaluations (about 3 percent) flipped the top models.

Their examination revealed that many influential votes may have been a result of user error. In some cases, it appeared there was a clear answer as to which LLM performed better, but the user chose the other model instead, Broderick says.

“We can never know what was in the user’s mind at that time, but maybe they mis-clicked or weren’t paying attention, or they honestly didn’t know which one was better. The big takeaway here is that you don’t want noise, user error, or some outlier determining which is the top-ranked LLM,” she adds.

The researchers suggest that gathering additional feedback from users, such as confidence levels in each vote, would provide richer information that could help mitigate this problem. Ranking platforms could also use human mediators to assess crowdsourced responses.

For the researchers’ part, they want to continue exploring generalization in other contexts while also developing better approximation methods that can capture more examples of non-robustness.

“Broderick and her students’ work shows how you can get valid estimates of the influence of specific data on downstream processes, despite the intractability of exhaustive calculations given the size of modern machine-learning models and datasets,” says Jessica Hullman, the Ginni Rometty Professor of Computer Science at Northwestern University, who was not involved with this work.  “The recent work provides a glimpse into the strong data dependencies in routinely applied — but also very fragile — methods for aggregating human preferences and using them to update a model. Seeing how few preferences could really change the behavior of a fine-tuned model could inspire more thoughtful methods for collecting these data.”

This research is funded, in part, by the Office of Naval Research, the MIT-IBM Watson AI Lab, the National Science Foundation, Amazon, and a CSAIL seed award.

Today, MIT plays a key role in maintaining U.S. competitiveness, technological leadership, and national defense — and much of the Institute’s work to support the nation’s standing in these areas can be traced back to 1953.

Two months after he took office that year, U.S. President Dwight Eisenhower received a startling report from the military: The USSR had successfully exploded a nuclear bomb nine months sooner than intelligence sources had predicted. The rising Communist power had also detonated a hydrogen bomb using development technology more sophisticated than that of the U.S. And lastly, there was evidence of a new Soviet bomber that rivaled the B-52 in size and range — and the aircraft was of an entirely original design from within the USSR. There was, the report concluded, a significant chance of a surprise nuclear attack on the United States.

Eisenhower’s understanding of national security was vast (he had led the Allies to victory in World War II and served as the first supreme commander of NATO), but the connections he’d made during his two-year stint as president of Columbia University would prove critical to navigating the emerging challenges of the Cold War. He sent his advisors in search of a plan for managing this threat, and he suggested they start with James Killian, then president of MIT.

Killian had an unlikely path to the presidency of MIT. “He was neither a scientist nor an engineer,” says David Mindell, the Dibner Professor of the History of Engineering and Manufacturing and a professor of aeronautics and astronautics at MIT. “But Killian turned out to be a truly gifted administrator.”

While he was serving as editor of MIT Technology Review (where he founded what became the MIT Press), Killian was tapped by then-president Karl Compton to join his staff. As the war effort ramped up on the MIT campus in the 1940s, Compton deputized Killian to lead the RadLab — a 4,000-person effort to develop and deploy the radar systems that proved decisive in the Allied victory.

Killian was named MIT’s 10th president in 1948. In 1951, he launched MIT Lincoln Laboratory, a federally funded research center where MIT and U.S. Air Force scientists and engineers collaborated on new air defense technologies to protect the nation against a nuclear attack.

Two years later, within weeks of Eisenhower’s 1953 request, Killian convened a group of leading scientists at MIT. The group proposed a three-part study: The U.S. needed to reassess its offensive capabilities, its continental defense, and its intelligence operations. Eisenhower agreed.

Killian mobilized 42 engineers and scientists from across the country into three panels matching the committee’s charge. Between September 1954 and February 1955, the panels held 307 meetings with every major defense and intelligence organization in the U.S. government. They had unrestricted access to every project, plan, and program involving national defense. The result, a 190-page report titled “Meeting the Threat of a Surprise Attack,” was delivered to Eisenhower’s desk on Feb. 14, 1955.

The Killian Report, as it came to be known, would go on to play a dramatic role in defining the frontiers of military technology, intelligence gathering, national security policy, and global affairs over the next several decades. Killian’s input would also have dramatic impacts on Eisenhower’s presidency and the relationship between the federal government and higher education.

Foreseeing an evolving competition

The Killian Report opens by anticipating four projected “periods” in the shifting balance of power between the U.S. and the Soviet Union.

In 1955, the U.S. had a decided offensive advantage over the USSR, but it was overly vulnerable to surprise attack. In 1956 and 1957, the U.S. would have an even larger offensive advantage and be only somewhat less vulnerable to surprise. By 1960, the U.S.’ offensive advantage would be narrower, but it would be in a better position to anticipate an attack. Within a decade, the report stated, the two nations would enter “Period IV” — during which “an attack by either side would result in mutual destruction … [a period] so fraught with danger to the U.S. that we should push all promising technological development so that we may stay in Periods II and III as long as possible.”

The report went on to make extensive, detailed recommendations — accelerated development of intercontinental ballistic missiles and high-energy aircraft fuels, expansion and increased ground security for “delivery system” facilities, increased cooperation with Canada and more studies about establishing monitoring stations on polar pack ice, and “studies directed toward better understanding of the radiological hazards that may result from the detonation of large numbers of nuclear weapons,” among others.

“Eisenhower really wanted to draw the perspectives of scientists and engineers into his decision-making,” says Mindell. “Generals and admirals tend to ask for more arms and more boots on the ground. The president didn’t want to be held captive by these views — and Killian’s report really delivered this for him.”

On the day it arrived, President Eisenhower circulated the Killian Report to the head of every department and agency in the federal government and asked them to comment on its recommendations. The Cold War arms race was on — and it would be between scientists and engineers in the United States and those in the Soviet Union.

An odd couple

The Killian Report made many recommendations based on “the correctness of the current national intelligence estimates” — even though “Eisenhower was frustrated with his whole intelligence apparatus,” says Will Hitchcock, the James Madison Professor of History at the University of Virginia and author of “The Age of Eisenhower.” “He felt it was still too much World War II ‘exploding-cigar’ stuff. There wasn’t enough work on advance warning, on seeing what’s over the hill. But that’s what Eisenhower really wanted to know.” The surprise attack on Pearl Harbor still lingered in the minds of many Americans, Hitchcock notes, and “that needed to be avoided.”

Killian needed an aggressive, innovative thinker to assess U.S. intelligence, so he turned to Edwin Land. The cofounder of Polaroid, Land was an astonishingly bold engineer and inventor. He also had military experience, having developed new ordnance targeting systems, aerial photography devices, and other photographic and visual surveillance technologies during World War II. Killian approached Land knowing their methods and work style were quite different. (When the offer to lead the intelligence panel was made, Land was in Hollywood advising filmmakers on the development of 3D movies; Land told Killian he had a personal rule that any committee he served on “must fit into a taxicab.”)

In fall 1954, Land and his five-person panel quickly confirmed Killian and Eisenhower’s suspicions: “We would go in and interview generals and admirals in charge of intelligence and come away worried,” Land reported to Killian later. “We were [young scientists] asking questions — and they couldn’t answer them.” Killian and Land realized this would set their report and its recommendations on a complicated path: While they needed to acknowledge and address the challenges of broadly upgrading intelligence activities, they also needed to make rapid progress on responding to the Soviet threat.

As work on the report progressed, Land and Killian held briefings with Eisenhower. They used these meetings to make two additional proposals — neither of which, President Eisenhower decided, would be spelled out in the final report for security reasons. The first was the development of missile-firing submarines, a long-term prospect that would take a decade to complete. (The technology developed for Polaris-class submarines, Mindell notes, transferred directly to the rockets that powered the Apollo program to the moon.)

The second proposal — to fast-track development of the U-2, a new high-altitude spy plane —could be accomplished within a year, Land told Eisenhower. The president agreed to both ideas, but he put a condition on the U-2 program. As Killian later wrote: “The president asked that it should be handled in an unconventional way so that it would not become entangled in the bureaucracy of the Defense Department or troubled by rivalries among the services.”

Powered by Land’s revolutionary imaging devices, the U-2 would become a critical tool in the U.S.’ ability to assess and understand the Soviet Union’s nuclear capacity. But the spy plane would also go on to have disastrous consequences for the peace process and for Eisenhower.

The aftermath(s)

The Killian Report has a very complex legacy, says Christopher Capozzola, the Elting Morison Professor of History. “There is a series of ironies about the whole undertaking,” he says. “For example, Eisenhower was trying to tamp down interservice rivalries by getting scientists to decide things. But within a couple of years those rivalries have all gotten worse.” Similarly, Capozzola notes, Eisenhower — who famously coined the phrase “military-industrial complex” and warned against it — amplified the militarization of scientific research “more than anyone else.”

Another especially painful irony emerged on May 1, 1960. Two weeks before a meeting between Eisenhower and Khrushchev in Paris to discuss how the U.S. and USSR could ease Cold War tensions and slow the arms race, a U-2 was shot down in Soviet airspace. After a public denial by the U.S. that the aircraft was being used for espionage, the Soviets produced the plane’s wreckage, cameras, and pilot — who admitted he was working for the CIA. The peace process, which had become the centerpiece of Eisenhower’s intended legacy, collapsed.

There were also some brighter outcomes of the Killian Report, Capozzola says. It marked a dramatic reset of the national government’s relationship with academic scientists and engineers — and with MIT specifically. “The report really greased the wheels between MIT scientists and Washington,” he notes. “Perhaps more than the report itself, the deep structures and relationships that Killian set up had implications for MIT and other research universities. They started to orient their missions toward the national interest,” he adds.

The report also cemented Eisenhower’s relationship with Killian. After the launch of Sputnik, which induced a broad public panic in the U.S. about Soviet scientific capabilities, the president called on Killian to guide the national response. Eisenhower later named Killian the first special assistant to the president for science and technology. In the years that followed, Killian would go on to help launch NASA, and MIT engineers would play a critical role in the Apollo mission that landed the first person on the moon. To this day, researchers at MIT and Lincoln Laboratory uphold this legacy of service, advancing knowledge in areas vital to national security, economic competitiveness, and quality of life for all Americans.

As Eisenhower’s special assistant, Killian met with him almost daily and became one of his most trusted advisors. “Killian could talk to the president, and Eisenhower really took his advice,” says Capozzola. “Not very many people can do that. The fact that Killian had that and used it was different.”

A key to their relationship, Capozzola notes, was Killian’s approach to his work. “He exemplified the notion that if you want to get something done, don’t take the credit. At no point did Killian think he was setting science policy. He was advising people on their best options, including decision-makers who would have to make very difficult decisions. That’s it.”

In 1977, after many tours of duty in Washington and his retirement from MIT, Killian summarized his experience working for Eisenhower in his memoir, “Sputnik, Scientists, and Eisenhower.” Killian said of his colleagues: “They were held together in close harmony not only by the challenge of the scientific and technical work they were asked to undertake but by their abiding sense of the opportunity they had to serve a president they admired and the country they loved. They entered the corridors of power in a moment of crisis and served there with a sense of privilege and of admiration for the integrity and high purpose of the White House.”

In a wide-ranging live conversation, MIT President Sally Kornbluth joined Jim Braude and Margery Eagan live in studio for GBH’s Boston Public Radio on Thursday, February 5. They talked about MIT, the pressures facing America’s research enterprise, the importance of science, that Congressional hearing on antisemitism in 2023, and more – including Sally’s experience as a Type 1 diabetic.

Reflecting on how research and innovation in the treatment of diabetes has advanced over decades of work, leading to markedly better patient care, Kornbluth exclaims: “This is science!”

With new financial pressures facing universities, increased competition for talented students and scholars from outside the U.S., as well as unprecedented pressures on university leaders and campuses, co-host Eagan asks Kornbluth what she thinks will happen in years to come.

“For us, one of the hardest things now is the endowment tax,” remarks Kornbluth. “That is $240 million a year. Think about how much science you can get for $240 million a year. Are we managing it? Yes. Are we still forging ahead on all of our exciting initiatives? Yes. But we’ve had to reconfigure things. We’ve had to merge things. And it’s not the way we should be spending our time and money.”   

Watch and listen to the full episode on YouTube. President Kornbluth appears one hour and seven minutes into the broadcast.

Following Kornbluth’s appearance, MIT Assistant Professor John Urschel – also a former offensive lineman for the Baltimore Ravens –   joined Edgar B. Herwick III, host of GBH’s newest show, The Curiosity Desk, to talk about his love of his family, linear algebra, and football.

On how he eventually chose math over football, Urschel quips: “Well, I hate to break it to you, I like math better… let me tell you, when I started my PhD at MIT, I just fell in love with the place. I fell in love with this idea of being in this environment [where] everyone loves math, everyone wants to learn. I was just constantly excited every day showing up.”

Prof. Urschel appears about 2 hours and 40 minutes into the webcast on YouTube.

Coming up on Curiosity Desk later this month…

Airing weekday afternoons from 1-2 p.m., The Curiosity Desk will welcome additional MIT guests in the coming weeks. On Thursday, Feb. 12 Anette “Peko” Hosoi, Pappalardo Professor of Mechanical Engineering, and Jerry Lu MFin ’24, a former researcher at the MIT Sports Lab, visit The Curiosity Desk to discuss their work using AI to help Olympic figure skaters improve their jumps.

Then, on Thursday, Feb. 19, Professors Sangeeta Bhatia and Angela Belcher talk with Herwick about their research to improve diagnostics for ovarian cancer. We learn that about 80% of the time ovarian cancer starts in the fallopian tubes and how this points the way to a whole new approach to diagnosing and treating the disease. 

Early in the 1969 film “Midnight Cowboy,” Dustin Hoffman, playing the character of Ratso Rizzo, crosses a Manhattan street and angrily bangs on the hood of an encroaching taxi. Hoffman’s line — “I’m walking here!” — has since been repeated by thousands of New Yorkers. Where cars and people mix, tensions rise.

And yet, governments and planners across the U.S. haven’t thoroughly tracked where it is that cars and people mix. Officials have long measured vehicle traffic closely while largely ignoring pedestrian traffic. Now, an MIT research group has assembled a routable dataset of sidewalks, crosswalks, and footpaths for all of New York City — a massive mapping project and the first complete model of pedestrian activity in any U.S. city.

The model could help planners decide where to make pedestrian infrastructure and public space investments, and illuminate how development decisions could affect non-motorized travel in the city. The study also helps pinpoint locations throughout the city where there are both lots of pedestrians and high pedestrian hazards, such as traffic crashes, and where streets or intersections are most in need of upgrades.

“We now have a first view of foot traffic all over New York City and can check planning decisions against it,” says Andres Sevtsuk, an associate professor in MIT’s Department of Urban Studies and Planning (DUSP), who led the study. “New York has very high densities of foot traffic outside of its most well-known areas.”

Indeed, one upshot of the model is that while Manhattan has the most foot traffic per block, the city’s other boroughs contain plenty of pedestrian-heavy stretches of sidewalk and could probably use more investment on behalf of walkers.

“Midtown Manhattan has by far the most foot traffic, but we found there is a probably unintentional Manhattan bias when it comes to policies that support pedestrian infrastructure,” Sevtsuk says. “There are a whole lot of streets in New York with very high pedestrian volumes outside of Manhattan, whether in Queens or the Bronx or Brooklyn, and we’re able to show, based on data, that a lot of these streets have foot-traffic levels similar to many parts of Manhattan.”

And, in an advance that could help cities anywhere, the model was used to quantify vehicle crashes involving pedestrians not only as raw totals, but on a per-pedestrian basis.

“A lot of cities put real investments behind keeping pedestrians safe from vehicles by prioritizing dangerous locations,” Sevtsuk says. “But that’s not only where the most crashes occur. Here we are able to calculate accidents per pedestrian, the risk people face, and that broadens the picture in terms of where the most dangerous intersections for pedestrians really are.”

The paper, “Spatial Distribution of Foot-traffic in New York City and Applications for Urban Planning,” is published today in Nature Cities.

The authors are Sevtsuk, the Charles and Ann Spaulding Associate Professor of Urban Science and Planning in DUSP and head of the City Design and Development Group; Rounaq Basu, an assistant professor at Georgia Tech; Liu Liu, a PhD student at the City Form Lab in DUSP; Abdulaziz Alhassan, a PhD student at MIT’s Center for Complex Engineering Systems; and Justin Kollar, a PhD student at MIT’s Leventhal Center for Advanced Urbanism in DUSP.

Walking everywhere

The current study continues work Sevtsuk and his colleagues have conducted charting and modeling pedestrian traffic around the world, from Melbourne to MIT’s Kendall Square neighborhood in Cambridge, Massachusetts. Many cities collect some pedestrian count data — but not much. And while officials usually request vehicle traffic impact assessments for new development plans, they rarely study how new developments or infrastructure proposals affect pedestrians.

However, New York City does devote part of its Department of Transportation (DOT) to pedestrian issues, and about 41 percent of trips city-wide are made on foot, compared to just 28 percent by vehicle, likely the highest such ratio in any big U.S. city. To calibrate the model, the MIT team used pedestrian counts that New York City’s DOT recorded in 2018 and 2019, covering up to 1,000 city sidewalk segments on weekdays and up to roughly 450 segments on weekends.

The researchers were able to test the model — which incorporates a wide range of factors — against New York City’s pedestrian-count data. Once calibrated, the model could expand foot-traffic estimates throughout the whole city, not just the points where pedestrian counts were observed.

The results showed that in Midtown Manhattan, there are about 1,697 pedestrians, on average, per sidewalk segment per hour during the evening peak of foot traffic, the highest in the city. The financial district in lower Manhattan comes in second, at 740 pedestrians per hour, with Greenwich Village third at 656.

Other parts of Manhattan register lower levels of foot traffic, however. Morningside Heights and East Harlem register 226 and 227 pedestrians per block per hour. And that’s similar to, or lower than, some parts of other boroughs. Brooklyn Heights has 277 pedestrians per sidewalk segment per hour; University Heights in the Bronx has 263; Borough Park in Brooklyn and the Grand Concourse in the Bronx average 236; and a slice of Queens in the Corona area averages 222. Many other spots are over 200.

The model overlays many different types of pedestrian journeys for each time period and shows that people are generally headed to work and schools in the morning, but conduct more varied types of trips in mid-day and the evening, as they seek out amenities or conduct social or recreational visits.

“Because of jobs, transit stops are the biggest generators of foot traffic in the morning peak,” Liu observes. “In the evening peak, of course people need to get home too, but patterns are much more varied, and people are not just returning from work or school. More social and recreational travel happens after work, whether it’s getting together with friends or running errands for family or family care trips, and that’s what the model detects too.”

On the safety front, pedestrians face danger in many places, not just the intersections with the most total accidents. Many parts of the city are riskier than others on a per-pedestrian basis, compared to the locations with the most pedestrian-related crashes.

“Places like Times Square and Herald Square in Manhattan may have numerous crashes, but they have very high pedestrian volumes, and it’s actually relatively safe to walk there,” Basu says. “There are other parts of the city, around highway off-ramps and heavy car-infrastructure, including the relatively low-density borough of Staten Island, which turn out to have a disproportionate number of crashes per pedestrian.”

Taking the model across the U.S.

The MIT model stands a solid chance of being applied in New York City policy and planning circles, since officials there are aware of the research and have been regularly communicating with the MIT team about it.

For his part, Sevtsuk emphasizes that, as distinct as New York City might be, the MIT model can be applied to cities and town anywhere in the U.S. As it happens, the team is working with municipal officials in two other places at the moment. One is Los Angeles, where city officials are not only trying to upgrade pedestrian and public transit mobility for regular daily trips, but making plans to handle an influx of visitors for the 2028 summer Olympics.

Meanwhile the state of Maine is working with the MIT team to evaluate pedestrian movement in over 140 of its cities and towns, to better understand the kinds of upgrades and safety improvements it could make for pedestrians across the state. Sevtsuk hopes that still other places will take notice of the New York City study and recognize that the tools are in place to analyze foot traffic more broadly in U.S. cities, to address the urgent need to decarbonize cities, and to start balancing what he views as the disproportionate focus on car travel prevalent in 20th century urban planning.

“I hope this can inspire other cities to invest in modeling foot traffic and mapping pedestrian infrastructure as well,” Sevtsuk says. “Very few cities make plans for pedestrian mobility or examine rigorously how future developments will impact foot-traffic. But they can. Our models serve as a test bed for making future changes.” 

Oxygen is a vital and constant presence on Earth today. But that hasn’t always been the case. It wasn’t until around 2.3 billion years ago that oxygen became a permanent fixture in the atmosphere, during a pivotal period known as the Great Oxidation Event (GOE), which set the evolutionary course for oxygen-breathing life as we know it today.

A new study by MIT researchers suggests some early forms of life may have evolved the ability to use oxygen hundreds of millions of years before the GOE. The findings may represent some of the earliest evidence of aerobic respiration on Earth.

In a study appearing today in the journal Palaeogeography, Palaeoclimatology, Palaeoecology, MIT geobiologists traced the evolutionary origins of a key enzyme that enables organisms to use oxygen. The enzyme is found in the vast majority of aerobic, oxygen-breathing life forms today. The team discovered that this enzyme evolved during the Mesoarchean — a geological period that predates the Great Oxidation Event by hundreds of millions of years.

The team’s results may help to explain a longstanding puzzle in Earth’s history: Why did it take so long for oxygen to build up in the atmosphere?

The very first producers of oxygen on the planet were cyanobacteria — microbes that evolved the ability to use sunlight and water to photosynthesize, releasing oxygen as a byproduct. Scientists have determined that cyanobacteria emerged around 2.9 billion years ago. The microbes, then, were presumably churning out oxygen for hundreds of millions of years before the Great Oxidation Event. So, where did all of cyanobacteria’s early oxygen go?

Scientists suspect that rocks may have drawn down a large portion of oxygen early on, through various geochemical reactions. The MIT team’s new study now suggests that biology may have also played a role.

The researchers found that some organisms may have evolved the enzyme to use oxygen hundreds of millions of years before the Great Oxidation Event. This enzyme may have enabled the organisms living near cyanobacteria to gobble up any small amounts of oxygen that the microbes produced, in turn delaying oxygen’s accumulation in the atmosphere for hundreds of millions of years.

“This does dramatically change the story of aerobic respiration,” says study co-author Fatima Husain, a postdoc in MIT’s Department of Earth, Atmospheric and Planetary Sciences (EAPS). “Our study adds to this very recently emerging story that life may have used oxygen much earlier than previously thought. It shows us how incredibly innovative life is at all periods in Earth’s history.”

The study’s other co-authors include Gregory Fournier, associate professor of geobiology at MIT, along with Haitao Shang and Stilianos Louca of the University of Oregon.

First respirers

The new study adds to a long line of work at MIT aiming to piece together oxygen’s history on Earth. This body of research has helped to pin down the timing of the Great Oxidation Event as well as the first evidence of oxygen-producing cyanobacteria. The overall understanding that has emerged is that oxygen was first produced by cyanobacteria around 2.9 billion years ago, while the Great Oxidation Event — when oxygen finally accumulated enough to persist in the atmosphere — took place much later, around 2.33 billion years ago.

For Husain and her colleagues, this apparent delay between oxygen’s first production and its eventual persistence inspired a question.

“We know that the microorganisms that produce oxygen were around well before the Great Oxidation Event,” Husain says. “So it was natural to ask, was there any life around at that time that could have been capable of using that oxygen for aerobic respiration?”

If there were in fact some life forms that were using oxygen, even in small amounts, they might have played a role in keeping oxygen from building up in the atmosphere, at least for a while.

To investigate this possibility, the MIT team looked to heme-copper oxygen reductases, which are a set of enzymes that are essential for aerobic respiration. The enzymes act to reduce oxygen to water, and they are found in the majority of aerobic, oxygen-breathing organism today, from bacteria to humans.

“We targeted the core of this enzyme for our analyses because that’s where the reaction with oxygen is actually taking place,” Husain explains.

Tree dates

The team aimed to trace the enzyme’s evolution backward in time to see when the enzyme first emerged to enable organisms to use oxygen. They first identified the enzyme’s genetic sequence and then used an automated search tool to look for this same sequence in databases containing the genomes of millions of different species of organisms.

“The hardest part of this work was that we had too much data,” Fournier says. “This enzyme is just everywhere and is present in most modern living organism. So we had to sample and filter the data down to a dataset that was representative of the diversity of modern life and also small enough to do computation with, which is not trivial.”

The team ultimately isolated the enzyme’s sequence from several thousand modern species and mapped these sequences onto an evolutionary tree of life, based on what scientists know about when each respective species has likely evolved and branched off. They then looked through this tree for specific species that might offer related information about their origins.

If, for instance, there is a fossil record for a particular organism on the tree, that record would include an estimate of when that organism appeared on Earth. The team would use that fossil’s age to “pin” a date to that organism on the tree. In a similar way, they could place pins across the tree to effectively tighten their estimates for when in time the enzyme evolved from one species to the next.

In the end, the researchers were able to trace the enzyme as far back as the Mesoarchean — a geological era that lasted from 3.2 to 2.8 billion years ago. It’s around this time that the team suspects the enzyme — and organisms’ ability to use oxygen — first emerged. This period predates the Great Oxidation Event by several hundred million years.

The new findings suggest that, shortly after cyanobacteria evolved the ability to produce oxygen, other living things evolved the enzyme to use that oxygen. Any such organism that happened to live near cyanobacteria would have been able to quickly take up the oxygen that the bacteria churned out. These early aerobic organisms may have then played some role in preventing oxygen from escaping to the atmosphere, delaying its accumulation for hundreds of millions of years.

“Considered all together, MIT research has filled in the gaps in our knowledge of how Earth’s oxygenation proceeded,” Husain says. “The puzzle pieces are fitting together and really underscore how life was able to diversify and live in this new, oxygenated world.”

This research was supported, in part, by the Research Corporation for Science Advancement Scialog program.

The National Academy of Engineering (NAE) has announced T. Alan Hatton, MIT’s Ralph Landau Professor of Chemical Engineering Practice, Post-Tenure, as the recipient of the 2026 Bernard M. Gordon Prize for Innovation in Engineering and Technology Education, recognizing his transformative leadership of the Institute’s David H. Koch School of Chemical Engineering Practice. The award citation highlights his efforts to advance “an immersive, industry-integrated educational model that has produced thousands of engineering leaders, strengthening U.S. technological competitiveness and workforce readiness.”

The Gordon Prize recognizes “new modalities and experiments in education that develop effective engineering leaders.” The prize is awarded annually and carries a $500,000 cash award, half granted to the recipient and the remainder granted to their institution to support the recognized innovation.

“As engineering challenges become more complex and interdisciplinary, education must evolve alongside them,” says Paula Hammond, Institute Professor and dean of the School of Engineering. “Under Alan’s leadership, the Practice School has demonstrated how rigorous academics, real industrial problems, and student responsibility can be woven together into an educational experience that is both powerful and adaptable. His work offers a compelling blueprint for the future of engineering education.”

Hatton served as director of the Practice School for 36 years, from 1989 until his retirement in 2025. When he assumed the role, the program worked with a limited number of host companies, largely within traditional chemical industries. Over time, Hatton reshaped the program’s scope and structure, enabling it to operate across continents and sectors to offer students exposure to diverse technologies, organizational cultures, and geographic settings.

“The MIT Chemical Engineering Practice School represents a level of experiential learning that few programs anywhere can match,” says Kristala L. J. Prather, the Arthur D. Little Professor and head of the Department of Chemical Engineering. “This recognition reflects not only Alan’s extraordinary personal contributions, but also the enduring value of a program that prepares students to deliver impact from their very first day as engineers.”

Central to Hatton’s approach was a deliberate strategy of adaptability. He introduced a model in which new companies are recruited regularly as Practice School hosts, broadening participation while keeping the program aligned with emerging technologies and industry needs. He also strengthened on-campus preparation by launching an intensive project management course during MIT’s Independent Activities Period (IAP) — training that has since become foundational for students entering complex, team-based industrial environments.

This forward-looking vision is shared by current Practice School leadership. Fikile Brushett, Ralph Landau Professor of Chemical Engineering Practice and director of the program, emphasizes that Hatton’s legacy is not a static one. “Alan consistently positioned the Practice School to respond to change — whether in technology, industry expectations, or educational practice,” Brushett says. “The Gordon Prize provides an opportunity to further evolve the program while staying true to its core principles of immersion, rigor, and partnership.”

In recognition of Hatton’s service, the department established the T. Alan Hatton Fund in fall 2025 with support from Practice School alumni. The fund is dedicated to helping launch new Practice School stations, lowering barriers for emerging partners and sustaining the program’s ability to engage with a broad and diverse set of industries.

Learning that delivers value on both sides

The Practice School’s impact extends well beyond the classroom. Student teams are embedded directly within host organizations — often in manufacturing plants or research and development centers — where they tackle open-ended technical problems under real operational constraints. Sponsors routinely cite tangible outcomes from these projects, including improved processes, reduced costs, and new technical directions informed by MIT-level analysis.

For students, the experience offers something difficult to replicate in traditional academic settings: sustained responsibility for complex work, direct interaction with industry professionals, and repeated opportunities to present, defend, and refine their ideas. The result is a training environment that closely mirrors professional engineering practice, while retaining the reflective depth of an academic program.

A program shaped by history — and by change

The Practice School was established in 1916 to complement classroom instruction with hands-on industrial experience, an idea that was unconventional at the time. More than a century later, the program has not only endured but continually reinvented itself, expanding far beyond its early focus on regional chemical manufacturing.

Today, Practice School students work with companies around the world in fields that include pharmaceuticals, food production, energy, advanced materials, software, and finance. The program remains a defining feature of graduate education in MIT’s Department of Chemical Engineering, linking research strengths with the practical demands of industry.

Participation in the Practice School is a required component of the department’s Master of Science in Chemical Engineering Practice (MSCEP) and PhD/ScD Chemical Engineering Practice (CEP) programs. After completing coursework, students attend two off-campus stations, spending two months at each site. Teams of two or three students work on month-long projects, culminating in formal presentations and written reports delivered to host organizations. Recent stations have included placements with Evonik in Germany, AstraZeneca in Maryland, EGA in the United Arab Emirates, AspenTech in Massachusetts, and Shell Technology Center and Dimensional Energy in Texas.

“I’m deeply honored by this recognition,” Hatton says. “The Practice School has always been about learning through responsibility — placing students in situations where their work matters. This award will help MIT build on that foundation and explore ways to extend the model so it can serve even more students and partners in the years ahead.”

Hatton obtained his BS and MS degrees in chemical engineering at the University of Natal in Durban, South Africa, before spending three years as a researcher at the Council for Scientific and Industrial Research in Pretoria. He later earned his PhD at the University of Wisconsin at Madison and joined the MIT faculty in 1982 as an assistant professor.

Over the course of his career at MIT, Hatton helped extend the Practice School model beyond campus through his involvement in the Singapore–MIT Alliance for Research and Technology and the Cambridge–MIT Institute, contributing to the development of practice-based engineering education in international settings. He also served as co-director of the MIT Energy Initiative’s Low-Carbon Energy Center focused on carbon capture, utilization, and storage.

Hatton has long been recognized for his commitment to education and service. From 1983 to 1986, he served as a junior faculty housemaster (now known as an associate head of house) in MacGregor House and received MIT’s Everett Moore Baker Teaching Award in 1983. His professional honors include being named a founding fellow of the American Institute of Medical and Biological Engineering and an honorary professorial fellow at the University of Melbourne in Australia.

In addition to his educational leadership, Hatton has made substantial contributions to the broader engineering community, chairing multiple national and international conferences in the areas of colloids and separation processes and delivering numerous plenary, keynote, and invited lectures worldwide.

Hatton will formally receive the Bernard M. Gordon Prize at a ceremony hosted by the National Academy of Engineering at MIT on April 30.

The ability to use language to communicate is one of things that makes us human. At MIT’s McGovern Institute for Brain Research, scientists led by Evelina Fedorenko have defined an entire network of areas within the brain dedicated to this ability, which work together when we speak, listen, read, write, or sign.

Much of the language network lies within the brain’s neocortex, where many of our most sophisticated cognitive functions are carried out. Now, Fedorenko’s lab, which is part of MIT's Department of Brain and Cognitive Sciences, has identified language-processing regions within the cerebellum, extending the language network to a part of the brain better known for helping to coordinate the body’s movements. Their findings are reported Jan. 21 in the journal Neuron.

“It’s like there’s this region in the cerebellum that we’ve been forgetting about for a long time,” says Colton Casto, a graduate student at Harvard and MIT who works in Fedorenko’s lab. “If you’re a language researcher, you should be paying attention to the cerebellum.”

Imaging the language network

There have been hints that the cerebellum makes important contributions to language. Some functional imaging studies detected activity in this area during language use, and people who suffer damage to the cerebellum sometimes experience language impairments. But no one had been able to pin down exactly which parts of the cerebellum were involved, or tease out their roles in language processing.

To get some answers, Fedorenko’s lab took a systematic approach, using methods they have used to map the language network in the neocortex. For 15 years, the lab has captured functional brain imaging data as volunteers carried out various tasks inside an MRI scanner. By monitoring brain activity as people engaged in different kinds of language tasks, like reading sentences or listening to spoken words, as well as non-linguistic tasks, like listening to noise or memorizing spatial patterns, the team has been able identify parts of the brain that are exclusively dedicated to language processing.

Their work shows that everyone’s language network uses the same neocortical regions. The precise anatomical location of these regions varies, however, so to study the language network in any individual, Fedorenko and her team must map that person’s network inside an MRI scanner using their language-localizer tasks.

Satellite language network

While the Fedorenko lab has largely focused on how the neocortex contributes to language processing, their brain scans also capture activity in the cerebellum. So Casto revisited those scans, analyzing cerebellar activity from more than 800 people to look for regions involved in language processing. Fedorenko points out that teasing out the individual anatomy of the language network turned out to particularly vital in the cerebellum, where neurons are densely packed and areas with different functional specializations sit very close to one another. Ultimately, Casto was able to identify four cerebellar areas that consistently got involved during language use.

Three of these regions were clearly involved in language use, but also reliably became engaged during certain kinds of non-linguistic tasks. Casto says this was a surprise, because all the core language areas in the neocortex are dedicated exclusively to language processing. The researchers speculate that the cerebellum may be integrating information from different parts of the cortex — a function that could be important for many cognitive tasks.

“We’ve found that language is distinct from many, many other things — but at some point, complex cognition requires everything to work together,” Fedorenko says. “How do these different kinds of information get connected? Maybe parts of the cerebellum serve that function.”

The researchers also found a spot in the right posterior cerebellum with activity patterns that more closely echoed those of the language network in the neocortex. This region stayed silent during non-linguistic tasks, but became active during language use. For all of the linguistic activities that Casto analyzed, this region exhibited patterns of activity that were very similar to what the lab has seen in neocortical components of the language network. “Its contribution to language seems pretty similar,” Casto says. The team describes this area as a “cerebellar satellite” of the language network.

Still, the researchers think it’s unlikely that neurons in the cerebellum, which are organized very differently than those in the neocortex, replicate the precise function of other parts of the language network. Fedorenko’s team plans to explore the function of this satellite region more deeply, investigating whether it may participate in different kinds of tasks.

The researchers are also exploring the possibility that the cerebellum is particularly important for language learning — playing an outsized role during development, or when people learn languages later in life.

Fedorenko says the discovery may also have implications for treating language impairments caused when an injury or disease damages the brain’s neocortical language network. “This area may provide a very interesting potential target to help recovery from aphasia,” Fedorenko says.

Currently, researchers are exploring the possibility that non-invasively stimulating language-associated parts of the brain might promote language recovery. “This right cerebellar region may be just the right thing to potentially stimulate to up-regulate some of that function that’s lost,” Fedorenko says.

Whether you’re a scientist brainstorming research ideas or a CEO hoping to automate a task in human resources or finance, you’ll find that artificial intelligence tools are becoming the assistants you didn’t know you needed. In particular, many professionals are tapping into the talents of semi-autonomous software systems called AI agents, which can call on AI at specific points to solve problems and complete tasks.

AI agents are particularly effective when they use large language models (LLMs) because those systems are powerful, efficient, and adaptable. One way to program such technology is by describing in code what you want your system to do (the “workflow”), including when it should use an LLM. If you were a software company trying to revamp your old codebase to use a more modern programming language for better optimizations and safety, you might build a system that uses an LLM to translate the codebase one file at a time, testing each file as you go.

But what happens when LLMs make mistakes? You’ll want the agent to backtrack to make another attempt, incorporating lessons it learned from previous mistakes. Coding this up can take as much effort as implementing the original agent; if your system for translating a codebase contained thousands of lines of code, then you’d be making thousands of lines of code changes or additions to support the logic for backtracking when LLMs make mistakes. 

To save programmers time and effort, researchers with MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and Asari AI have developed a framework called “EnCompass.” 

With EnCompass, you no longer have to make these changes yourself. Instead, when EnCompass runs your program, it automatically backtracks if LLMs make mistakes. EnCompass can also make clones of the program runtime to make multiple attempts in parallel in search of the best solution. In full generality, EnCompass searches over the different possible paths your agent could take as a result of the different possible outputs of all the LLM calls, looking for the path where the LLM finds the best solution.

Then, all you have to do is to annotate the locations where you may want to backtrack or clone the program runtime, as well as record any information that may be useful to the strategy used to search over the different possible execution paths of your agent (the search strategy). You can then separately specify the search strategy — you could either use one that EnCompass provides out of the box or, if desired, implement your own custom search strategy.

“With EnCompass, we’ve separated the search strategy from the underlying workflow of an AI agent,” says lead author Zhening Li ’25, MEng ’25, who is an MIT electrical engineering and computer science (EECS) PhD student, CSAIL researcher, and research consultant at Asari AI. “Our framework lets programmers easily experiment with different search strategies to find the one that makes the AI agent perform the best.” 

EnCompass was used for agents implemented as Python programs that call LLMs, where it demonstrated noticeable code savings. EnCompass reduced coding effort for implementing search by up to 80 percent across agents, such as an agent for translating code repositories and for discovering transformation rules of digital grids. In the future, EnCompass could enable agents to tackle large-scale tasks, including managing massive code libraries, designing and carrying out science experiments, and creating blueprints for rockets and other hardware.

Branching out

When programming your agent, you mark particular operations — such as calls to an LLM — where results may vary. These annotations are called “branchpoints.” If you imagine your agent program as generating a single plot line of a story, then adding branchpoints turns the story into a choose-your-own-adventure story game, where branchpoints are locations where the plot branches into multiple future plot lines. 

You can then specify the strategy that EnCompass uses to navigate that story game, in search of the best possible ending to the story. This can include launching parallel threads of execution or backtracking to a previous branchpoint when you get stuck in a dead end.

Users can also plug-and-play a few common search strategies provided by EnCompass out of the box, or define their own custom strategy. For example, you could opt for Monte Carlo tree search, which builds a search tree by balancing exploration and exploitation, or beam search, which keeps the best few outputs from every step. EnCompass makes it easy to experiment with different approaches to find the best strategy to maximize the likelihood of successfully completing your task.

The coding efficiency of EnCompass

So just how code-efficient is EnCompass for adding search to agent programs? According to researchers’ findings, the framework drastically cut down how much programmers needed to add to their agent programs to add search, helping them experiment with different strategies to find the one that performs the best.

For example, the researchers applied EnCompass to an agent that translates a repository of code from the Java programming language, which is commonly used to program apps and enterprise software, to Python. They found that implementing search with EnCompass — mainly involving adding branchpoint annotations and annotations that record how well each step did — required 348 fewer lines of code (about 82 percent) than implementing it by hand. They also demonstrated how EnCompass enabled them to easily try out different search strategies, identifying the best strategy to be a two-level beam search algorithm, achieving an accuracy boost of 15 to 40 percent across five different repositories at a search budget of 16 times the LLM calls made by the agent without search.

“As LLMs become a more integral part of everyday software, it becomes more important to understand how to efficiently build software that leverages their strengths and works around their limitations,” says co-author Armando Solar-Lezama, who is an MIT professor of EECS and CSAIL principal investigator. “EnCompass is an important step in that direction.”

The researchers add that EnCompass targets agents where a program specifies the steps of the high-level workflow; the current iteration of their framework is less applicable to agents that are entirely controlled by an LLM. “In those agents, instead of having a program that specifies the steps and then using an LLM to carry out those steps, the LLM itself decides everything,” says Li. “There is no underlying programmatic workflow, so you can execute inference-time search on whatever the LLM invents on the fly. In this case, there’s less need for a tool like EnCompass that modifies how a program executes with search and backtracking.”

Li and his colleagues plan to extend EnCompass to more general search frameworks for AI agents. They also plan to test their system on more complex tasks to refine it for real-world uses, including at companies. What’s more, they’re evaluating how well EnCompass helps agents work with humans on tasks like brainstorming hardware designs or translating much larger code libraries. For now, EnCompass is a powerful building block that enables humans to tinker with AI agents more easily, improving their performance.

“EnCompass arrives at a timely moment, as AI-driven agents and search-based techniques are beginning to reshape workflows in software engineering,” says Carnegie Mellon University Professor Yiming Yang, who wasn’t involved in the research. “By cleanly separating an agent’s programming logic from its inference-time search strategy, the framework offers a principled way to explore how structured search can enhance code generation, translation, and analysis. This abstraction provides a solid foundation for more systematic and reliable search-driven approaches to software development.”  

Li and Solar-Lezama wrote the paper with two Asari AI researchers: Caltech Professor Yisong Yue, an advisor at the company; and senior author Stephan Zheng, who is the founder and CEO. Their work was supported by Asari AI.

The team’s work was presented at the Conference on Neural Information Processing Systems (NeurIPS) in December.

A longstanding goal of immunotherapies and vaccine research is to induce antibodies in humans that neutralize deadly viruses such as HIV and influenza. Of particular interest are antibodies that are “broadly neutralizing,” meaning they can in principle eliminate multiple strains of a virus such as HIV, which mutates rapidly to evade the human immune system.

Researchers at MIT and the Scripps Research Institute have now developed a vaccine that generates a significant population of rare precursor B cells that are capable of evolving to produce broadly neutralizing antibodies. Expanding these cells is the first step toward a successful HIV vaccine.

The researchers’ vaccine design uses DNA instead of protein as a scaffold to fabricate a virus-like particle (VLP) displaying numerous copies of an engineered HIV immunogen called eOD-GT8, which was developed at Scripps. This vaccine generated substantially more precursor B cells in a humanized mouse model compared to a protein-based virus-like particle that has shown significant success in human clinical trials.

Preclinical studies showed that the DNA-VLP generated eight times more of the desired, or “on-target,” B cells than the clinical product, which was already shown to be highly potent.

“We were all surprised that this already outstanding VLP from Scripps was significantly outperformed by the DNA-based VLP,” says Mark Bathe, an MIT professor of biological engineering and an associate member of the Broad Institute of MIT and Harvard. “These early preclinical results suggest a potential breakthrough as an entirely new, first-in-class VLP that could transform the way we think about active immunotherapies, and vaccine design, across a variety of indications.”

The researchers also showed that the DNA scaffold doesn’t induce an immune response when applied to the engineered HIV antigen. This means the DNA VLP might be used to deliver multiple antigens when boosting strategies are needed, such as for challenging diseases such as HIV.

“The DNA-VLP allowed us for the first time to assess whether B cells targeting the VLP itself limit the development of ‘on target’ B cell responses — a longstanding question in vaccine immunology,” says Darrell Irvine, a professor of immunology and microbiology at the Scripps Research Institute and a Howard Hughes Medical Institute Investigator.

Bathe and Irvine are the senior authors of the study, which appears today in Science. The paper’s lead author is Anna Romanov PhD ’25.

Priming B cells

The new study is part of a major ongoing global effort to develop active immunotherapies and vaccines that expand specific lineages of B cells. All humans have the necessary genes to produce the right B cells that can neutralize HIV, but they are exceptionally rare and require many mutations to become broadly neutralizing. If exposed to the right series of antigens, however, these cells can in principle evolve to eventually produce the requisite broadly neutralizing antibodies.

In the case of HIV, one such target antibody, called VRC01, was discovered by National Institutes of Health researchers in 2010 when they studied humans living with HIV who did not develop AIDS. This set off a major worldwide effort to develop an HIV vaccine that would induce this target antibody, but this remains an outstanding challenge.

Generating HIV-neutralizing antibodies is believed to require three stages of vaccination, each one initiated by a different antigen that helps guide B cell evolution toward the correct target, the native HIV envelope protein gp120.

In 2013, William Schief, a professor of immunology and microbiology at Scripps, reported an engineered antigen called eOD-GT6 that could be used for the first step in this process, known as priming. His team subsequently upgraded the antigen to eOD-GT8. Vaccination with eOD-GT8 arrayed on a protein VLP generated early antibody precursors to VRC01 both in mice and more recently in humans, a key first step toward an HIV vaccine.

However, the protein VLP also generated substantial “off-target” antibodies that bound the irrelevant, and potentially highly distracting, protein VLP itself. This could have unknown consequences on propagating target B cells of interest for HIV, as well as other challenging immunotherapy applications.

The Bathe and Irvine labs set out to test if they could use a particle made from DNA, instead of protein, to deliver the priming antigen. These nanoscale particles are made using DNA origami, a method that offers precise control over the structure of synthetic DNA and allows researchers to attach viral antigens at specific locations.

In 2024, Bathe and Daniel Lingwood, an associate professor at Harvard Medical School and a principal investigator at the Ragon Institute, showed this DNA VLP could be used to deliver a SARS-CoV-2 vaccine in mice to generate neutralizing antibodies. From that study, the researchers learned that the DNA scaffold does not induce antibodies to the VLP itself, unlike proteins. They wondered whether this might also enable a more focused antibody response.

Building on these results, Romanov, co-advised by Bathe and Irvine, set off to apply the DNA VLP to the Scripps HIV priming vaccine, based on eOD-GT8.

“Our earlier work with SARS-CoV-2 antigens on DNA-VLPs showed that DNA-VLPs can be used to focus the immune response on an antigen of interest. This property seemed especially useful for a case like HIV, where the B cells of interest are exceptionally rare. Thus, we hypothesized that reducing the competition among other irrelevant B cells (by delivering the vaccine on a silent DNA nanoparticle) may help these rare cells have a better chance to survive,”  Romanov says.

Initial studies in mice, however, showed the vaccine did not induce sufficient early B cell response to the first, priming dose.

After redesigning the DNA VLPs, Romanov and colleagues found that a smaller diameter version with 60 instead of 30 copies of the engineered antigen dramatically out-performed the clinical protein VLP construct, both in overall number of antigen-specific B cells and the fraction of B cells that were on-target to the specific HIV domain of interest. This was a result of improved retention of the particles in B cell follicles in lymph nodes and better collaboration with helper T cells, which promote B cell survival.

Overall, these improvements enabled the particles to generate eightfold more on-target B cells than the vaccine consisting of eOD-GT8 carried by a protein scaffold. Another key finding, elucidated by the Lingwood lab, was that the DNA particles promoted VRC01 precursor B cells toward the VRC01 antibody more efficiently than the protein VLP.

“In the field of vaccine immunology, the question of whether B cell responses to a targeted protective epitope on a vaccine antigen might be hindered by responses to neighboring off-target epitopes on the same antigen has been under intense investigation,” says Schief, who is also vice president for protein design at Moderna. “There are some data from other studies suggesting that off-target responses might not have much impact, but this study shows quite convincingly that reducing off-target responses by using a DNA VLP can improve desired on-target responses.”

“While nanoparticle formulations have been great at boosting antibody responses to various antigens, there is always this nagging question of whether competition from B cells specific for the particle’s own structural antigens won’t get in the way of antibody responses to targeted epitopes,” says Gabriel Victora, a professor of immunology, virology, and microbiology at Rockefeller University, who was not involved in the study. “DNA-based particles that leverage B cells’ natural tolerance to nucleic acids are a clever idea to circumvent this problem, and the research team’s elegant experiments clearly show that this strategy can be used to make difficult epitopes easier to target.”

A “silent” scaffold

The fact that the DNA-VLP scaffold doesn’t induce scaffold-specific antibodies means that it could be used to carry second and potentially third antigens needed in the vaccine series, as the researchers are currently investigating. It also might offer significantly improved on-target antibodies for numerous antigens that are outcompeted and dominated by off-target, irrelevant protein VLP scaffolds in this or other applications.

“A breakthrough of this paper is the rigorous, mechanistic quantification of how DNA-VLPs can ‘focus’ antibody responses on target antigens of interest, which is a consequence of the silent nature of this DNA-based scaffold we’ve previously shown is stealth to the immune system,” Bathe says.

More broadly, this new type of VLP could be used to generate other kinds of protective antibody responses against pandemic threats such as flu, or potentially against chemical warfare agents, the researchers suggest. Alternatively, it might be used as an active immunotherapy to generate antibodies that target amyloid beta or tau protein to treat degenerative diseases such as Alzheimer’s, or to generate antibodies that target noxious chemicals such as opioids or nicotine to help people suffering from addiction.

The research was funded by the National Institutes of Health; the Ragon Institute of MGH, MIT, and Harvard; the Howard Hughes Medical Institute; the National Science Foundation; the Novo Nordisk Foundation; a Koch Institute Support (core) Grant from the National Cancer Institute; the National Institute of Environmental Health Sciences; the Gates Foundation Collaboration for AIDS Vaccine Discovery; the IAVI Neutralizing Antibody Center; the National Institute of Allergy and Infectious Diseases; and the U.S. Army Research Office through MIT’s Institute for Soldier Nanotechnologies.

Before the thrill of victory; before the agony of defeat; before the gold medalist’s national anthem plays, there is the Olympic torch. A symbol of unity, friendship, and the spirit of competition, the torch links today’s Olympic Games to its heritage in ancient Greece.

The torch for the 2026 Milano Cortina Olympic Games and Paralympic Games was designed by Carlo Ratti, a professor of the practice for the MIT Department of Urban Studies and Planning and the director of the Senseable City Lab in the MIT School of Architecture and Planning.

A native of Turin, Italy, and a respected designer and architect worldwide, Ratti’s work and that of his firm, Carlo Ratti Associati, has been featured at various international expositions such as the French Pavilion at the Osaka Expo (World’s Fair) in 2025 and the Italian Pavilion at the Dubai Expo in 2020. Their design for The Cloud, a 400-foot tall spherical structure that would serve as a unique observation deck, was a finalist for the 2012 Olympic Games in London, but ultimately not built.

Ratti relishes the opportunity to participate in these events.

“You can push the boundaries more at these [venues] because you are building something that is temporary,” says Ratti. “They allow for more creativity, so it’s a good moment to experiment.”

Based on his previous work, Ratti was invited to design the torch by the Olympic organizers. He approached the project much as he instructs his students working in his lab.

“It is about what the object or the design is to convey,” Ratti says. “How it can touch people, how it can relate to people, how it can transmit emotions. That’s the most important thing.”

To Ratti, the fundamental aspect of the torch is the flame. A few months before the games begin, the torch is lit in Olympia, Greece, using a parabolic mirror reflecting the sun’s rays. In ancient Greece, the flame was considered “sacred,” and was to remain lit throughout the competition. Ratti, familiar with the history of the Olympic torch, is less impressed with designs that he deems overwrought. Many torches added superfluous ornamentation to its exterior much like cars are designed around their engines, he says. Instead, he decided to strip away everything that wasn’t essential to the flame itself.

What is “essential”

“Essential” — the official name for the 2026 Winter Olympic torch — was designed to perform regardless of the weather, wind, or altitude it would encounter on its journey from Olympia to Milan. The process took three years with many designs created, considered, and discussed with the local and global Olympic committees and Olympic sponsor Versalis. And, as with Ratti’s work at MIT, researchers and engineers collaborated in the effort.

“Each design pushed the boundaries in different directions, but all of them with the key principle to put the flame at the center,” says Ratti who wanted the torch to embody “an ethos of frugality.”

At the core of Ratti’s torch is a high-performance burner powered by bio-GPL produced by energy company ENI from 100 percent renewable feedstocks. Furthermore, the torch can be recharged 10 times. In previous years, torches were used only once. This allowed for a 10-fold reduction in the number of torches created.

Also unique to this torch is its internal mechanism, which is visible via a vertical opening along its side, allowing audiences to see the burner in action. This reinforces the desire to keep the emphasis on the flame instead of the object.

In keeping with the requisite for minimalism and sustainability, the torch is primarily composed of recycled aluminum. It is the lightest torch created for the Olympics, weighing just under 2.5 pounds. The body is finished with a PVD coating that is heat resistant, letting it shift colors by reflecting the environments — such as the mountains and the city lights — through which it is carried. The Olympic torch is a blue-green shade, while the Paralympic torch is gold.

The torch won an honorable mention in Italy’s most prestigious industrial design award, the Compasso d’Oro.

The Olympic Relay

The torch relay is considered an event itself, drawing thousands as it is carried to the host city by hundreds of volunteers. Its journey for the 2026 Olympics started in late November and, after visiting cities across Greece, will have covered all 110 Italian provinces before arriving in Milan for the opening ceremony on Feb. 6.

Ratti carried the torch for a portion of its journey through Turin in mid-January — another joyful invitation to this quadrennial event. He says winter sports are his favorite; he grew up skiing where these games are being held, and has since skied around the world — from Utah to the Himalayas.

In addition to a highly sustainable torch, there was another statement Ratti wanted to make: He wanted to showcase the Italy of today and of the future. It is the same issue he confronted as the curator of the 2025 Biennale Architettura in Venice titled “Intelligens. Natural. Artificial. Collective: an architecture exhibition, but infused with technology for the future.”

“When people think about Italy, they often think about the past, from ancient Romans to the Renaissance or Baroque period,” he says. “Italy does indeed have a significant past. But the reality is that it is also the second-largest industrial powerhouse in Europe and is leading in innovation and tech in many fields. So, the 2026 torch aims to combine both past and future. It draws on Italian design from the past, but also on future-forward technologies.”

“There should be some kind of architectural design always translating into form some kind of ethical principles or ideals. It’s not just about a physical thing. Ultimately, it’s about the human dimension. That applies to the work we do at MIT or the Olympic torch.”

Brian Hedden PhD ’12 has been appointed co-associate dean of the Social and Ethical Responsibilities of Computing (SERC) at MIT, a cross-cutting initiative in the MIT Schwarzman College of Computing, effective Jan. 16.

Hedden is a professor in the Department of Linguistics and Philosophy, holding an MIT Schwarzman College of Computing shared position with the Department of Electrical Engineering and Computer Science (EECS). He joined the MIT faculty last fall from the Australian National University and the University of Sydney, where he previously served as a faculty member. He earned his BA from Princeton University and his PhD from MIT, both in philosophy.

“Brian is a natural and compelling choice for SERC, as a philosopher whose work speaks directly to the intellectual challenges facing education and research today, particularly in computing and AI. His expertise in epistemology, decision theory, and ethics addresses questions that have become increasingly urgent in an era defined by information abundance and artificial intelligence. His scholarship exemplifies the kind of interdisciplinary inquiry that SERC exists to advance,” says Dan Huttenlocher, dean of the MIT Schwarzman College of Computing and the Henry Ellis Warren Professor of Electrical Engineering and Computer Science.

Hedden’s research focuses on how we ought to form beliefs and make decisions, and it explores how philosophical thinking about rationality can yield insights into contemporary ethical issues, including ethics of AI. He is the author of “Reasons without Persons: Rationality, Identity, and Time” (Oxford University Press, 2015) and articles on topics such as collective action problems, legal standards of proof, algorithmic fairness, and political polarization.

Joining co-associate dean Nikos Trichakis, the J.C. Penney Professor of Management at the MIT Sloan School of Management, Hedden will help lead SERC and advance the initiative’s ongoing research, teaching, and engagement efforts. He succeeds professor of philosophy Caspar Hare, who stepped down at the conclusion of his three-year term on Sept. 1, 2025.

Since its inception in 2020, SERC has launched a range of programs and activities designed to cultivate responsible “habits of mind and action” among those who create and deploy computing technologies, while fostering the development of technologies in the public interest.

The SERC Scholars Program invites undergraduate and graduate students to work alongside postdoctoral mentors to explore interdisciplinary ethical challenges in computing. The initiative also hosts an annual prize competition that challenges MIT students to envision the future of computing, publishes a twice-yearly series of case studies, and collaborates on coordinated curricular materials, including active-learning projects, homework assignments, and in-class demonstrations. In 2024, SERC introduced a new seed grant program to support MIT researchers investigating ethical technology development; to date, two rounds of grants have been awarded to 24 projects.

Antonio Torralba, Delta Electronics Professor of Electrical Engineering and Computer Science and faculty head of artificial intelligence and decision-making at MIT, has been named to the 2025 cohort of Association for Computing Machinery (ACM) Fellows. He shares the honor of an ACM Fellowship with three MIT alumni: Eytan Adar ’97, MEng ’98; George Candea ’97, MEng ’98; and Gookwon Edward Suh SM ’01, PhD ’05.

A principal investigator within both the Computer Science and Artificial Intelligence Laboratory and the Center for Brains, Minds, and Machines, Torralba received his BS in telecommunications engineering from Telecom BCN, Spain, in 1994, and a PhD in signal, image, and speech processing from the Institut National Polytechnique de Grenoble, France, in 2000. At different points in his MIT career, he has been director of both the MIT Quest for Intelligence (now the MIT Siegel Family Quest for Intelligence) and the MIT-IBM Watson AI Lab. 

Torralba’s research focuses on computer vision, machine learning, and human visual perception; as he puts it, “I am interested in building systems that can perceive the world like humans do.” Alongside Phillip Isola and William Freeman, he recently co-authored “Foundations of Computer Vision,” an 800-plus page textbook exploring the foundations and core principles of the field. 

Among other awards and recognitions, he is the recipient of the 2008 National Science Foundation Career award; the 2010 J. K. Aggarwal Prize from the International Association for Pattern Recognition; the 2017 Frank Quick Faculty Research Innovation Fellowship; the Louis D. Smullin (’39) Award for Teaching Excellence; and the 2020 PAMI Mark Everingham Prize. In 2021, he was awarded the inaugural Thomas Huang Memorial Prize by the Pattern Analysis and Machine Intelligence Technical Committee and was named a fellow of the Association for the Advancement of Artificial Intelligence. In 2022, he received an honorary doctoral degree from the Universitat Politècnica de Catalunya — BarcelonaTech (UPC). 

ACM fellows, the highest honor bestowed by the professional organization, are registered members of the society selected by their peers for outstanding accomplishments in computing and information technology and/or outstanding service to ACM and the larger computing community.

In the pursuit of solutions to complex global challenges including disease, energy demands, and climate change, scientific researchers, including at MIT, have turned to artificial intelligence, and to quantitative analysis and modeling, to design and construct engineered cells with novel properties. The engineered cells can be programmed to become new therapeutics — battling, and perhaps eradicating, diseases.

James J. Collins is one of the founders of the field of synthetic biology, and is also a leading researcher in systems biology, the interdisciplinary approach that uses mathematical analysis and modeling of complex systems to better understand biological systems. His research has led to the development of new classes of diagnostics and therapeutics, including in the detection and treatment of pathogens like Ebola, Zika, SARS-CoV-2, and antibiotic-resistant bacteria. Collins, the Termeer Professor of Medical Engineering and Science and professor of biological engineering at MIT, is a core faculty member of the Institute for Medical Engineering and Science (IMES), the director of the MIT Abdul Latif Jameel Clinic for Machine Learning in Health, as well as an institute member of the Broad Institute of MIT and Harvard, and core founding faculty at the Wyss Institute for Biologically Inspired Engineering, Harvard.

In this Q&A, Collins speaks about his latest work and goals for this research.

Q.  You’re known for collaborating with colleagues across MIT, and at other institutions. How have these collaborations and affiliations helped you with your research? 

A: Collaboration has been central to the work in my lab. At the MIT Jameel Clinic for Machine Learning in Health, I formed a collaboration with Regina Barzilay [the Delta Electronics Professor in the MIT Department of Electrical Engineering and Computer Science and affiliate faculty member at IMES] and Tommi Jaakkola [the Thomas Siebel Professor of Electrical Engineering and Computer Science and the Institute for Data, Systems, and Society] to use deep learning to discover new antibiotics. This effort combined our expertise in artificial intelligence, network biology, and systems microbiology, leading to the discovery of halicin, a potent new antibiotic effective against a broad range of multidrug-resistant bacterial pathogens. Our results were published in Cell in 2020 and showcased the power of bringing together complementary skill sets to tackle a global health challenge.

At the Wyss Institute, I’ve worked closely with Donald Ingber [the Judah Folkman Professor of Vascular Biology at Harvard Medical School and the Vascular Biology Program at Boston Children’s Hospital, and Hansjörg Wyss Professor of Biologically Inspired Engineering at Harvard], leveraging his organs-on-chips technology to test the efficacy of AI-discovered and AI-generated antibiotics. These platforms allow us to study how drugs behave in human tissue-like environments, complementing traditional animal experiments and providing a more nuanced view of their therapeutic potential.

The common thread across our many collaborations is the ability to combine computational predictions with cutting-edge experimental platforms, accelerating the path from ideas to validated new therapies.

Q. Your research has led to many advances in designing novel antibiotics, using generative AI and deep learning. Can you talk about some of the advances you’ve been a part of in the development of drugs that can battle multi-drug-resistant pathogens, and what you see on the horizon for breakthroughs in this arena?

A: In 2025, our lab published a study in Cell demonstrating how generative AI can be used to design completely new antibiotics from scratch. We used genetic algorithms and variational autoencoders to generate millions of candidate molecules, exploring both fragment-based designs and entirely unconstrained chemical space. After computational filtering, retrosynthetic modeling, and medicinal chemistry review, we synthesized 24 compounds and tested them experimentally. Seven showed selective antibacterial activity. One lead, NG1, was highly narrow-spectrum, eradicating multi-drug-resistant Neisseria gonorrhoeae, including strains resistant to first-line therapies, while sparing commensal species. Another, DN1, targeted methicillin-resistant Staphylococcus aureus (MRSA) and cleared infections in mice through broad membrane disruption. Both were non-toxic and showed low rates of resistance.

Looking ahead, we are using deep learning to design antibiotics with drug-like properties that make them stronger candidates for clinical development. By integrating AI with high-throughput biological testing, we aim to accelerate the discovery and design of antibiotics that are novel, safe, and effective, ready for real-world therapeutic use. This approach could transform how we respond to drug-resistant bacterial pathogens, moving from a reactive to a proactive strategy in antibiotic development.

Q. You’re a co-founder of Phare Bio, a nonprofit organization that uses AI to discover new antibiotics, and the Collins Lab has helped to launch the Antibiotics-AI Project in collaboration with Phare Bio. Can you tell us more about what you hope to accomplish with these collaborations, and how they tie back to your research goals?

A: We founded Phare Bio as a nonprofit to take the most promising antibiotic candidates emerging from the Antibiotics-AI Project at MIT and advance them toward the clinic. The idea is to bridge the gap between discovery and development by collaborating with biotech companies, pharmaceutical partners, AI companies, philanthropies, other nonprofits, and even nation states. Akhila Kosaraju has been doing a brilliant job leading Phare Bio, coordinating these efforts and moving candidates forward efficiently.

Recently, we received a grant from ARPA-H to use generative AI to design 15 new antibiotics and develop them as pre-clinical candidates. This project builds directly on our lab’s research, combining computational design with experimental testing to create novel antibiotics that are ready for further development. By integrating generative AI, biology, and translational partnerships, we hope to create a pipeline that can respond more rapidly to the global threat of antibiotic resistance, ultimately delivering new therapies to patients who need them most.