While working to boost economic development in Detroit in the late 2010s, Nick Allen found he was running up against a problem.

The city was trying to spur more investment after long-term industrial flight to suburbs and other states. Relying more heavily on property taxes for revenue, the city was negotiating individualized tax deals with prospective businesses. That’s hardly a scenario unique to Detroit, but such deals involved lengthy approval processes that slowed investment decisions and made smaller projects seem unrealistic. 

Moreover, while creating small pockets of growth, these individualized tax abatements were not changing the city’s broader fiscal structure. They also favored those with leverage and resources to work the system for a break.

“The thing you really don’t want to do with taxes is have very particular, highly procedural ways of adjusting the burdens,” says Allen, now a doctoral student in MIT’s Department of Urban Studies and Planning (DUSP). “You want a simple process that fits people’s ideas about what fairness looks like.”

So, after starting his PhD program at MIT, Allen kept studying urban fiscal policy. Along with a group of other scholars, he has produced research papers making the case for a land-value tax — a common tax rate on land that, combined with reduced property taxes, could raise more local revenue by encouraging more city-wide investment, even while lowering tax burdens on residents and businesses. As a bonus, it could also reduce foreclosures.

In the last few years, this has become a larger topic in urban policy circles. The mayor of Detroit has endorsed the idea. The New York Times has written about the work of Allen and his colleagues. The land-value tax is now a serious policy option.

It is unusual for a graduate student to have their work become part of a prominent policy debate. But then, Allen is an unusual student. At MIT, he has not just conducted influential research in his field, but thrown himself into campus-based work with substantial impact as well. Allen has served on task forces assessing student stipend policy, expanding campus housing, and generating ideas for dining program reform.

For all these efforts, in May, Allen received the Karl Taylor Compton Prize, MIT’s highest student honor. At the ceremony, MIT Chancellor Melissa Nobles observed that Allen’s work helped Institute stakeholders “fully understand complex issues, ensuring his recommendations are not only well-informed but also practical and impactful.”

Looking to revive growth

Allen is a Minnesota native who received his BA from Yale University. In 2015, he enrolled in graduate school at MIT, receiving his master’s in city planning from DUSP in 2017. At the time, Allen worked on the Malaysia Sustainable Cities Project, headed by Professor Lawrence Susskind. At one point Allen spent a couple of months in a small Malaysian village studying the effects of coastal development on local fishing and farming.

Malaysia may be different than Michigan, but the issues that Allen encountered in Asia were similar to the ones he wanted to keep studying back in the U.S.: finding ways to finance growth.

“The core interests I have are around real estate, the physical environment, and these fiscal policy questions of how this all gets funded and what the responsibilities are of the state and private markets,” Allen says. “And that brought me to Detroit.”

Specifically, that landed him at the Detroit Economic Growth Corporation, a city-charted development agency that works to facilitate new investment. There, Allen started grappling with the city’s revenue problems. Once heralded as the richest city in America, Detroit has seen a lot of property go vacant, and has hiked property taxes on existing structures to compensate for that. Those rates then discouraged further investment and building.

To be sure, the challenges Detroit has faced stem from far more than tax policy and relate to many macroscale socioeconomic factors, including suburban flight, the shift of manufacturing to states with nonunion employees, and much more. But changing tax policy can be one lever to pull in response.

“It’s difficult to figure out how to revive growth in a place that’s been cannibalized by its losses,” Allen says.

Tasked with underwriting real estate projects, Allen started cataloguing the problems arising from Detroit’s property tax reliance, and began looking at past economics work on optimal tax policy in search of alternatives.

“There’s a real nose-to-the-ground empiricism you start with, asking why we have a system nobody would choose,” Allen says. “There were two parts to that, for me. One was initially looking at the difficulty of making individual projects work, from affordable housing to big industrial plants, along with, secondly, this wave of tax foreclosures in the city.”

Engineering, but for policy

After two years in Detroit, Allen returned to MIT, this time as a doctoral student in DUSP and with a research program oriented around the issues he had worked on. In pursuing that, Allen has worked closely with John E. Anderson, an economist at the University of Nebraska at Lincoln. With a nationwide team of economists convened by the Lincoln Institute of Land Policy, they worked to address the city’s questions on property tax reform.

One paper used current data to show that a land-value tax should lower tax-connected foreclosures in the city. Two other papers study the use of the tax in certain parts of Pennsylvania, one of the few states where it has been deployed. There, the researchers concluded, the land-value tax both leads to greater business development and raises property values.

“What we found overall, looking at past tax reduction in Detroit and other cities, is that in reducing the rate at which people in deep tax distress go through foreclosure, it has a fairly large effect,” Allen says. “It has some effect on allowing business to reinvest in properties. We are seeing a lot more attraction of investment. And it’s got the virtue of being a rules-based system.”

Those empirical results, he notes, helped confirm the sense that a policy change could help growth in Detroit.

“That really validated the hunch we were following,” Allen says.

The widespread attention the policy proposal has garnered could not really have been predicted. The tax has not yet been implemented in Detroit, although it has been a prominent part of civic debates there. Allen has been asked to consult on tax policy by officials in numerous large cities, and is hopeful the concept will gain still more traction.

Meanwhile, at MIT, Allen has one more year to go in his doctoral program. On top of his academic research, he has been an active participant in Institute matters, helping reshape graduate-school policies on multiple fronts.

For instance, Allen was part of the Graduate Housing Working Group, whose efforts helped spur MIT to build Graduate Junction, a new housing complex for 675 graduate students on Vassar Street in Cambridge, Massachusetts. The name also refers to the Grand Junction rail line that runs nearby; the complex formally opened in 2024.

“Innovative places struggle to build housing fast enough,” Allen said at the time Graduate Junction opened, also noting that “new housing for students reduces price pressure on the rest of the Cambridge community.”

Commenting on it now, he adds, “Maybe to most people graduate housing policy doesn’t sound that fun, but to me these are very absorbing questions.”

And ultimately, Allen says, the intellectual problems in either domain can be similar, whether he is working on city policy issues or campus enhancements.

“The reason I think planning fits so well here at MIT is, a lot of what I do is like policy engineering,” Allen says. “It’s really important to understand system constraints, and think seriously about finding solutions that can be built to purpose. I think that’s why I’ve felt at home here at MIT, working on these outside public policy topics, and projects for the Institute. You need to take seriously what people say about the constraints in their lives.”

John “JJ” Joannopoulos, the Francis Wright Davis Professor of Physics at MIT and director of the MIT Institute for Soldier Nanotechnologies (ISN), passed away on Aug. 17. He was 78. 

Joannopoulos was a prolific researcher in the field of theoretical condensed-matter physics, and an early pioneer in the study and application of photonic crystals. Many of his discoveries, in the ways materials can be made to manipulate light, have led to transformative and life-saving technologies, from chip-based optical wave guides, to wireless energy transfer to health-monitoring textiles, to precision light-based surgical tools.

His remarkable career of over 50 years was spent entirely at MIT, where he was known as much for his generous and unwavering mentorship as for his contributions to science. He made a special point to keep up rich and meaningful collaborations with many of his former students and postdocs, dozens of whom have gone on to faculty positions at major universities, and to leadership roles in the public and private sectors. In his five decades at MIT, he made lasting connections across campus, both in service of science, and friendship.

“A scientific giant, inspiring leader, and a masterful communicator, John carried a generous and loving heart,” says Yoel Fink PhD ’00, an MIT professor of materials science and engineering who was Joannopoulos’ former student and a longtime collaborator. “He chose to see the good in people, keeping his mind and heart always open. Asking little for himself, he gave everything in care of others. John lived a life of deep impact and meaning — savoring the details of truth-seeking, achieving rare discoveries and mentoring generations of students to achieve excellence. With warmth, humor, and a never-ending optimism, JJ left an indelible impact on science and on all who had the privilege to know him. Above all, he was a loving husband, father, grandfather, friend, and mentor.”

“In the end, the most remarkable thing about him was his unmatched humanity, his ability to make you feel that you were the most important thing in the world that deserved his attention, no matter who you were,” says Raul Radovitzky, ISN associate director and the Jerome C. Hunsaker Professor in MIT’s Department of Aeronautics and Astronautics. “The legacy he leaves is not only in equations and innovations, but in the lives he touched, the minds he inspired, and the warmth he spread in every room he entered.”

“JJ was a very special colleague: a brilliant theorist who was also adept at identifying practical applications; a caring and inspiring mentor of younger scientists; a gifted teacher who knew every student in his class by name,” says Deepto Chakrabarty ’88, the William A. M. Burden Professor in Astrophysics and head of MIT’s Department of Physics. “He will be deeply missed.”

Layers of light

John Joannopoulos was born in 1947 in New York City, where his parents both emigrated from Greece. His father was a playwright, and his mother worked as a psychologist. From an early age, Joannopoulos knew he wanted to be a physicist — mainly because the subject was his most challenging in school. In a recent interview with MIT News, he enthusiastically shared: “You probably wouldn’t believe this, but it’s true: I wanted to be a physics professor since I was in high school! I loved the idea of being able to work with students, and being able to have ideas.”

He attended the University of California at Berkeley, where he received a bachelor’s degree in 1968, and a PhD in 1974, both in physics. That same year, he joined the faculty at MIT, where he would spend his 50-plus-year career — though at the time, the chances of gaining a long-term foothold at the Institute seemed slim, as Joannopoulos told MIT News.

“The chair of the physics department was the famous nuclear physicist, Herman Feshbach, who told me the probability that I would get tenure was something like 30 percent,” Joannopoulos recalled. “But when you’re young and just starting off, it was certainly better than zero, and I thought, that was fine — there was hope down the line.”

Starting out at MIT, Joannopoulos knew exactly what he wanted to do. He quickly set up a group to study theoretical condensed-matter physics, and specifically, ab initio physics, meaning physics “from first principles.” In this initial work, he sought to build theoretical models to predict the electronic behavior and structure of materials, based solely on the atomic numbers of the atoms in a material. Such foundational models could be applied to understand and design a huge range of materials and structures.

Then, in the early 1990s, Joannopoulos took a research turn, spurred by a paper by physicist Eli Yablonovitch at the University of California at Los Angeles, who did some preliminary work on materials that can affect the behavior of photons, or particles of light. Joannopoulos recognized a connection with his first-principles work with electrons. Along with his students, he applied that approach to predict the fundamental behavior of photons in different classes of materials. His group was one of the first to pioneer the field of photonic crystals, and the study of how materials can be manipulated at the nanoscale to control the behavior of light traveling through. In 1995, Joannopoulos co-authored the first textbook on the subject.

And in 1998, he took on a more-than-century-old assumption about how light should reflect, and turned it on its head. That assumption predicted that light, shining onto a structure made of multiple refractive layers, could reflect back, but only for a limited range of angles. But in fact, Joannopoulos and his group showed that the opposite is true: If the structure’s layers followed a particular design criteria, the structure as a whole could reflect light coming from any and all angles. This structure, was called the “perfect mirror.”

That insight led to another: If the structure were rolled into a tube, the resulting hollow fiber could act as a perfect optical conduit. Any light traveling through the fiber would reflect and bounce around within the fiber, with none scattering away. Joannopoulos and his group applied this insight to develop the first precision “optical scalpel” — a fiber that can be safely handled, while delivering a highly focused laser, precise and powerful enough to perform delicate surgical procedures. Joannopoulos helped to commercialize the new tool with a startup, Omniguide, that has since provided the optical scalpel to assist in hundreds of thousands of medical procedures around the world.

Legendary mentor

In 2006, Joannopoulos took the helm as director of MIT’s Institute for Soldier Nanotechnologies — a post he steadfastly held for almost 20 years. During his dedicated tenure, he worked with ISN members across campus and in departments outside his own, getting to know and champion their work. He has facilitated countless collaborations between MIT faculty, industry partners, and the U.S. Department of Defense. Among the many projects he raised support for were innovations in lightweight armor, hyperspectral imaging, energy-efficient batteries, and smart and responsive fabrics.

Joannopoulos helped to translate many basic science insights into practical applications. He was a cofounder of six spinoff companies based on his fundamental research, and helped to create dozens more companies, which have advanced technologies as wide-ranging as laser surgery tools, to wireless electric power transmission, transparent display technologies, and optical computing. He was awarded 126 patents for his many discoveries, and has authored over 750 peer-reviewed papers.

In recognition of his wide impact and contributions, Joannopoulos was elected to the National Academy of Sciences and the American Academy of Arts and Sciences. He was also a fellow of both the American Physical Society and the American Association for the Advancement of Science. Over his 50-plus-year career, he was the recipient of many scientific awards and honors including the Max Born Award, and the Aneesur Rahman Prize in Computational Physics. Joannopoulos was also a gifted classroom teacher, and was recognized at MIT with the Buechner Teaching Prize in Physics and the Graduate Teaching Award in Science.

This year, Joannopoulos was the recipient of MIT’s Killian Achievement Award, which recognizes the extraordinary lifetime contributions of a member of the MIT faculty. In addition to the many accomplishments Joannopoulos has made in science, the award citation emphasized his lasting impact on the generations of students he has mentored:

“Professor Joannopoulos has served as a legendary mentor to generations of students, inspiring them to achieve excellence in science while at the same time facilitating the practical benefit to society through entrepreneurship,” the citation reads. “Through all of these individuals he has impacted — not to mention their academic descendants — Professor Joannopoulos has had a vast influence on the development of science in recent decades.”

“JJ was an amazing scientist: He published hundreds of papers that have been cited close to 200,000 times. He was also a serial entrepreneur: Companies he cofounded raised hundreds of millions of dollars and employed hundreds of people,” says MIT Professor Marin Soljacic ’96, a former postdoc under Joannopoulos who with him cofounded a startup, Witricity. “He was an amazing mentor, a close friend, and like a scientific father to me. He always had time for me, any time of the day, and as much as I needed.”

Indeed, Joannopoulos strived to meaningfully support his many students. In the classroom, he “was legendary,” says friend and colleague Patrick Lee ’66, PhD ’70, who recalls that Joannopoulos would make a point of memorizing the names and faces of more than 100 students on the first day of class, and calling them each by their first name, starting on the second day, and for the rest of the term.

What’s more, Joannopoulos encouraged graduate students and postdocs to follow their ideas, even when they ran counter to his own.

“John did not produce clones,” says Lee, who is an MIT professor emeritus of physics. “He showed them the way to do science by example, by caring and by sharing his optimism. I have never seen someone so deeply loved by his students.”

Even students who stepped off the photonics path have kept in close contact with their mentor, as former student and MIT professor Josh Winn ’94, SM ’94, PhD ’01 has done.

“Even though our work together ended more than 25 years ago, and I now work in a different field, I still feel like part of the Joannopoulos academic family,” says Winn, who is now a professor of astrophysics at Princeton University. “It's a loyal group with branches all over the world. We even had our own series of conferences, organized by former students to celebrate John's 50th, 60th, and 70th birthdays. Most professors would consider themselves fortunate to have even one such ‘festschrift’ honoring their legacy.”

MIT professor of mathematics Steven Johnson ’95, PhD ’01, a former student and frequent collaborator, has experienced personally, and seen many times over, Joannopoulos’ generous and open-door mentorship.

“In every collaboration, I’ve unfailingly observed him to cast a wide net to value multiple voices, to ensure that everyone feels included and valued, and to encourage collaborations across groups and fields and institutions,” Johnson says. “Kind, generous, and brimming with infectious enthusiasm and positivity, he set an example so many of his lucky students have striven to follow.”

Joannopoulos started at MIT around the same time as Marc Kastner, who had a nearby office on the second floor of Building 13.

“I would often hear loud arguments punctuated by boisterous laughter, coming from John’s office, where he and his students were debating physics,” recalls Kastner, who is the Donner Professor of Physics Emeritus at MIT. “I am sure this style of interaction is what made him such a great mentor.”

“He exuded such enthusiasm for science and good will to others that he was just good fun to be around,” adds friend and colleague Erich Ippen, MIT professor emeritus of physics.

“John was indeed a great man — a very special one. Everyone who ever worked with him understands this,” says Stanford University physics professor Robert Laughlin PhD ’79, one of Joannopoulos’ first graduate students, who went on to win the 1998 Nobel Prize in Physics. “He sprinkled a kind of transformative magic dust on people that induced them to dedicate every waking moment to the task of making new and wonderful things. You can find traces of it in lots of places around the world that matter, all of them the better for it. There’s quite a pile of it in my office.”

Joannopoulos is survived by his wife, Kyri Dunussi-Joannopoulos; their three daughters, Maria, Lena, and Alkisti; and their families. Details for funeral and memorial services are forthcoming.

Using machine learning, MIT chemical engineers have created a computational model that can predict how well any given molecule will dissolve in an organic solvent — a key step in the synthesis of nearly any pharmaceutical. This type of prediction could make it much easier to develop new ways to produce drugs and other useful molecules.

The new model, which predicts how much of a solute will dissolve in a particular solvent, should help chemists to choose the right solvent for any given reaction in their synthesis, the researchers say. Common organic solvents include ethanol and acetone, and there are hundreds of others that can also be used in chemical reactions.

“Predicting solubility really is a rate-limiting step in synthetic planning and manufacturing of chemicals, especially drugs, so there’s been a longstanding interest in being able to make better predictions of solubility,” says Lucas Attia, an MIT graduate student and one of the lead authors of the new study.

The researchers have made their model freely available, and many companies and labs have already started using it. The model could be particularly useful for identifying solvents that are less hazardous than some of the most commonly used industrial solvents, the researchers say.

“There are some solvents which are known to dissolve most things. They’re really useful, but they’re damaging to the environment, and they’re damaging to people, so many companies require that you have to minimize the amount of those solvents that you use,” says Jackson Burns, an MIT graduate student who is also a lead author of the paper. “Our model is extremely useful in being able to identify the next-best solvent, which is hopefully much less damaging to the environment.”

William Green, the Hoyt Hottel Professor of Chemical Engineering and director of the MIT Energy Initiative, is the senior author of the study, which appears today in Nature Communications. Patrick Doyle, the Robert T. Haslam Professor of Chemical Engineering, is also an author of the paper.

Solving solubility

The new model grew out of a project that Attia and Burns worked on together in an MIT course on applying machine learning to chemical engineering problems. Traditionally, chemists have predicted solubility with a tool known as the Abraham Solvation Model, which can be used to estimate a molecule’s overall solubility by adding up the contributions of chemical structures within the molecule. While these predictions are useful, their accuracy is limited.

In the past few years, researchers have begun using machine learning to try to make more accurate solubility predictions. Before Burns and Attia began working on their new model, the state-of-the-art model for predicting solubility was a model developed in Green’s lab in 2022.

That model, known as SolProp, works by predicting a set of related properties and combining them, using thermodynamics, to ultimately predict the solubility. However, the model has difficulty predicting solubility for solutes that it hasn’t seen before.

“For drug and chemical discovery pipelines where you’re developing a new molecule, you want to be able to predict ahead of time what its solubility looks like,” Attia says.

Part of the reason that existing solubility models haven’t worked well is because there wasn’t a comprehensive dataset to train them on. However, in 2023 a new dataset called BigSolDB was released, which compiled data from nearly 800 published papers, including information on solubility for about 800 molecules dissolved about more than 100 organic solvents that are commonly used in synthetic chemistry.

Attia and Burns decided to try training two different types of models on this data. Both of these models represent the chemical structures of molecules using numerical representations known as embeddings, which incorporate information such as the number of atoms in a molecule and which atoms are bound to which other atoms. Models can then use these representations to predict a variety of chemical properties.

One of the models used in this study, known as FastProp and developed by Burns and others in Green’s lab, incorporates “static embeddings.” This means that the model already knows the embedding for each molecule before it starts doing any kind of analysis.

The other model, ChemProp, learns an embedding for each molecule during the training, at the same time that it learns to associate the features of the embedding with a trait such as solubility. This model, developed across multiple MIT labs, has already been used for tasks such as antibiotic discovery, lipid nanoparticle design, and predicting chemical reaction rates.

The researchers trained both types of models on over 40,000 data points from BigSolDB, including information on the effects of temperature, which plays a significant role in solubility. Then, they tested the models on about 1,000 solutes that had been withheld from the training data. They found that the models’ predictions were two to three times more accurate than those of SolProp, the previous best model, and the new models were especially accurate at predicting variations in solubility due to temperature.

“Being able to accurately reproduce those small variations in solubility due to temperature, even when the overarching experimental noise is very large, was a really positive sign that the network had correctly learned an underlying solubility prediction function,” Burns says.

Accurate predictions

The researchers had expected that the model based on ChemProp, which is able to learn new representations as it goes along, would be able to make more accurate predictions. However, to their surprise, they found that the two models performed essentially the same. That suggests that the main limitation on their performance is the quality of the data, and that the models are performing as well as theoretically possible based on the data that they’re using, the researchers say.

“ChemProp should always outperform any static embedding when you have sufficient data,” Burns says. “We were blown away to see that the static and learned embeddings were statistically indistinguishable in performance across all the different subsets, which indicates to us that that the data limitations that are present in this space dominated the model performance.”

The models could become more accurate, the researchers say, if better training and testing data were available — ideally, data obtained by one person or a group of people all trained to perform the experiments the same way.

“One of the big limitations of using these kinds of compiled datasets is that different labs use different methods and experimental conditions when they perform solubility tests. That contributes to this variability between different datasets,” Attia says.

Because the model based on FastProp makes its predictions faster and has code that is easier for other users to adapt, the researchers decided to make that one, known as FastSolv, available to the public. Multiple pharmaceutical companies have already begun using it.

“There are applications throughout the drug discovery pipeline,” Burns says. “We’re also excited to see, outside of formulation and drug discovery, where people may use this model.”

The research was funded, in part, by the U.S. Department of Energy.

Within the past few years, models that can predict the structure or function of proteins have been widely used for a variety of biological applications, such as identifying drug targets and designing new therapeutic antibodies.

These models, which are based on large language models (LLMs), can make very accurate predictions of a protein’s suitability for a given application. However, there’s no way to determine how these models make their predictions or which protein features play the most important role in those decisions.

In a new study, MIT researchers have used a novel technique to open up that “black box” and allow them to determine what features a protein language model takes into account when making predictions. Understanding what is happening inside that black box could help researchers to choose better models for a particular task, helping to streamline the process of identifying new drugs or vaccine targets.

“Our work has broad implications for enhanced explainability in downstream tasks that rely on these representations,” says Bonnie Berger, the Simons Professor of Mathematics, head of the Computation and Biology group in MIT’s Computer Science and Artificial Intelligence Laboratory, and the senior author of the study. “Additionally, identifying features that protein language models track has the potential to reveal novel biological insights from these representations.”

Onkar Gujral, an MIT graduate student, is the lead author of the study, which appears this week in the Proceedings of the National Academy of Sciences. Mihir Bafna, an MIT graduate student, and Eric Alm, an MIT professor of biological engineering, are also authors of the paper.

Opening the black box

In 2018, Berger and former MIT graduate student Tristan Bepler PhD ’20 introduced the first protein language model. Their model, like subsequent protein models that accelerated the development of AlphaFold, such as ESM2 and OmegaFold, was based on LLMs. These models, which include ChatGPT, can analyze huge amounts of text and figure out which words are most likely to appear together.

Protein language models use a similar approach, but instead of analyzing words, they analyze amino acid sequences. Researchers have used these models to predict the structure and function of proteins, and for applications such as identifying proteins that might bind to particular drugs.

In a 2021 study, Berger and colleagues used a protein language model to predict which sections of viral surface proteins are less likely to mutate in a way that enables viral escape. This allowed them to identify possible targets for vaccines against influenza, HIV, and SARS-CoV-2.

However, in all of these studies, it has been impossible to know how the models were making their predictions.

“We would get out some prediction at the end, but we had absolutely no idea what was happening in the individual components of this black box,” Berger says.

In the new study, the researchers wanted to dig into how protein language models make their predictions. Just like LLMs, protein language models encode information as representations that consist of a pattern of activation of different “nodes” within a neural network. These nodes are analogous to the networks of neurons that store memories and other information within the brain.

The inner workings of LLMs are not easy to interpret, but within the past couple of years, researchers have begun using a type of algorithm known as a sparse autoencoder to help shed some light on how those models make their predictions. The new study from Berger’s lab is the first to use this algorithm on protein language models.

Sparse autoencoders work by adjusting how a protein is represented within a neural network. Typically, a given protein will be represented by a pattern of activation of a constrained number of neurons, for example, 480. A sparse autoencoder will expand that representation into a much larger number of nodes, say 20,000.

When information about a protein is encoded by only 480 neurons, each node lights up for multiple features, making it very difficult to know what features each node is encoding. However, when the neural network is expanded to 20,000 nodes, this extra space along with a sparsity constraint gives the information room to “spread out.” Now, a feature of the protein that was previously encoded by multiple nodes can occupy a single node.

“In a sparse representation, the neurons lighting up are doing so in a more meaningful manner,” Gujral says. “Before the sparse representations are created, the networks pack information so tightly together that it's hard to interpret the neurons.”

Interpretable models

Once the researchers obtained sparse representations of many proteins, they used an AI assistant called Claude (related to the popular Anthropic chatbot of the same name), to analyze the representations. In this case, they asked Claude to compare the sparse representations with the known features of each protein, such as molecular function, protein family, or location within a cell.

By analyzing thousands of representations, Claude can determine which nodes correspond to specific protein features, then describe them in plain English. For example, the algorithm might say, “This neuron appears to be detecting proteins involved in transmembrane transport of ions or amino acids, particularly those located in the plasma membrane.”

This process makes the nodes far more “interpretable,” meaning the researchers can tell what each node is encoding. They found that the features most likely to be encoded by these nodes were protein family and certain functions, including several different metabolic and biosynthetic processes.

“When you train a sparse autoencoder, you aren’t training it to be interpretable, but it turns out that by incentivizing the representation to be really sparse, that ends up resulting in interpretability,” Gujral says.

Understanding what features a particular protein model is encoding could help researchers choose the right model for a particular task, or tweak the type of input they give the model, to generate the best results. Additionally, analyzing the features that a model encodes could one day help biologists to learn more about the proteins that they are studying.

“At some point when the models get a lot more powerful, you could learn more biology than you already know, from opening up the models,” Gujral says.

The research was funded by the National Institutes of Health. 

MIT researchers have developed a reconfigurable antenna that dynamically adjusts its frequency range by changing its physical shape, making it more versatile for communications and sensing than static antennas.

A user can stretch, bend, or compress the antenna to make reversible changes to its radiation properties, enabling a device to operate in a wider frequency range without the need for complex, moving parts. With an adjustable frequency range, a reconfigurable antenna could adapt to changing environmental conditions and reduce the need for multiple antennas.

The word “antenna” may draw to mind metal rods like the “bunny ears” on top of old television sets, but the MIT team instead worked with metamaterials — engineered materials whose mechanical properties, such as stiffness and strength, depend on the geometric arrangement of the material’s components.

The result is a simplified design for a reconfigurable antenna that could be used for applications like energy transfer in wearable devices, motion tracking and sensing for augmented reality, or wireless communication across a wide range of network protocols.

In addition, the researchers developed an editing tool so users can generate customized metamaterial antennas, which can be fabricated using a laser cutter.

“Usually, when we think of antennas, we think of static antennas — they are fabricated to have specific properties and that is it. However, by using auxetic metamaterials, which can deform into three different geometric states, we can seamlessly change the properties of the antenna by changing its geometry, without fabricating a new structure. In addition, we can use changes in the antenna’s radio frequency properties, due to changes in the metamaterial geometry, as a new method of sensing for interaction design,” says lead author Marwa AlAlawi, a mechanical engineering graduate student at MIT.

Her co-authors include Regina Zheng and Katherine Yan, both MIT undergraduate students; Ticha Sethapakdi, an MIT graduate student in electrical engineering and computer science; Soo Yeon Ahn of the Gwangju Institute of Science and Technology in Korea; and co-senior authors Junyi Zhu, assistant professor at the University of Michigan; and Stefanie Mueller, the TIBCO Career Development Associate Professor in MIT’s departments of Electrical Engineering and Computer Science and Mechanical Engineering and leader of the Human-Computer Interaction Group at the Computer Science and Artificial Intelligence Lab. The research will be presented at the ACM Symposium on User Interface Software and Technology.

Making sense of antennas

While traditional antennas radiate and receive radio signals, in this work, the researchers looked at how the devices can act as sensors. The team’s goal was to develop a mechanical element that can also be used as an antenna for sensing.

To do this, they leveraged the antenna’s “resonance frequency,” which is the frequency at which the antenna is most efficient.

An antenna’s resonance frequency will shift due to changes in its shape. (Think about extending the left “bunny ear” to reduce TV static.) Researchers can capture these shifts for sensing. For instance, a reconfigurable antenna could be used in this way to detect the expansion of a person’s chest, to monitor their respiration.

To design a versatile reconfigurable antenna, the researchers used metamaterials. These engineered materials, which can be programmed to adopt different shapes, are composed of a periodic arrangement of unit cells that can be rotated, compressed, stretched, or bent.

By deforming the metamaterial structure, one can shift the antenna’s resonance frequency.

“In order to trigger changes in resonance frequency, we either need to change the antenna’s effective length or introduce slits and holes into it. Metamaterials allow us to get those different states from only one structure,” AlAlawi says.

The device, dubbed the meta-antenna, is composed of a dielectric layer of material sandwiched between two conductive layers.

To fabricate a meta-antenna, the researchers cut the dielectric laser out of a rubber sheet with a laser cutter. Then they added a patch on top of the dielectric layer using conductive spray paint, creating a resonating “patch antenna.”

But they found that even the most flexible conductive material couldn’t withstand the amount of deformation the antenna would experience.

“We did a lot of trial and error to determine that, if we coat the structure with flexible acrylic paint, it protects the hinges so they don’t break prematurely,” AlAlawi explains.

A means for makers

With the fabrication problem solved, the researchers built a tool that enables users to design and produce metamaterial antennas for specific applications.

The user can define the size of the antenna patch, choose a thickness for the dielectric layer, and set the length to width ratio of the metamaterial unit cells. Then the system automatically simulates the antenna’s resonance frequency range.

“The beauty of metamaterials is that, because it is an interconnected system of linkages, the geometric structure allows us to reduce the complexity of a mechanical system,” AlAlawi says.

Using the design tool, the researchers incorporated meta-antennas into several smart devices, including a curtain that dynamically adjusts household lighting and headphones that seamlessly transition between noise-cancelling and transparent modes.

For the smart headphone, for instance, when the meta-antenna expands and bends, it shifts the resonance frequency by 2.6 percent, which switches the headphone mode. The team’s experiments also showed that meta-antenna structures are durable enough to withstand more than 10,000 compressions.

Because the antenna patch can be patterned onto any surface, it could be used with more complex structures. For instance, the antenna could be incorporated into smart textiles that perform noninvasive biomedical sensing or temperature monitoring.

In the future, the researchers want to design three-dimensional meta-antennas for a wider range of applications. They also want to add more functions to the design tool, improve the durability and flexibility of the metamaterial structure, experiment with different symmetric metamaterial patterns, and streamline some manual fabrication steps.

This research was funded, in part, by the Bahrain Crown Prince International Scholarship and the Gwangju Institute of Science and Technology.

Using artificial intelligence, MIT researchers have come up with a new way to design nanoparticles that can more efficiently deliver RNA vaccines and other types of RNA therapies.

After training a machine-learning model to analyze thousands of existing delivery particles, the researchers used it to predict new materials that would work even better. The model also enabled the researchers to identify particles that would work well in different types of cells, and to discover ways to incorporate new types of materials into the particles.

“What we did was apply machine-learning tools to help accelerate the identification of optimal ingredient mixtures in lipid nanoparticles to help target a different cell type or help incorporate different materials, much faster than previously was possible,” says Giovanni Traverso, an associate professor of mechanical engineering at MIT, a gastroenterologist at Brigham and Women’s Hospital, and the senior author of the study.

This approach could dramatically speed the process of developing new RNA vaccines, as well as therapies that could be used to treat obesity, diabetes, and other metabolic disorders, the researchers say.

Alvin Chan, a former MIT postdoc who is now an assistant professor at Nanyang Technological University, and Ameya Kirtane, a former MIT postdoc who is now an assistant professor at the University of Minnesota, are the lead authors of the new open-access study, which appears today in Nature Nanotechnology.

Particle predictions

RNA vaccines, such as the vaccines for SARS-CoV-2, are usually packaged in lipid nanoparticles (LNPs) for delivery. These particles protect mRNA from being broken down in the body and help it to enter cells once injected.

Creating particles that handle these jobs more efficiently could help researchers to develop even more effective vaccines. Better delivery vehicles could also make it easier to develop mRNA therapies that encode genes for proteins that could help to treat a variety of diseases.

In 2024, Traverso’s lab launched a multiyear research program, funded by the U.S. Advanced Research Projects Agency for Health (ARPA-H), to develop new ingestible devices that could achieve oral delivery of RNA treatments and vaccines.

“Part of what we’re trying to do is develop ways of producing more protein, for example, for therapeutic applications. Maximizing the efficiency is important to be able to boost how much we can have the cells produce,” Traverso says.

A typical LNP consists of four components — a cholesterol, a helper lipid, an ionizable lipid, and a lipid that is attached to polyethylene glycol (PEG). Different variants of each of these components can be swapped in to create a huge number of possible combinations. Changing up these formulations and testing each one individually is very time-consuming, so Traverso, Chan, and their colleagues decided to turn to artificial intelligence to help speed up the process.

“Most AI models in drug discovery focus on optimizing a single compound at a time, but that approach doesn’t work for lipid nanoparticles, which are made of multiple interacting components,” Chan says. “To tackle this, we developed a new model called COMET, inspired by the same transformer architecture that powers large language models like ChatGPT. Just as those models understand how words combine to form meaning, COMET learns how different chemical components come together in a nanoparticle to influence its properties — like how well it can deliver RNA into cells.”

To generate training data for their machine-learning model, the researchers created a library of about 3,000 different LNP formulations. The team tested each of these 3,000 particles in the lab to see how efficiently they could deliver their payload to cells, then fed all of this data into a machine-learning model.

After the model was trained, the researchers asked it to predict new formulations that would work better than existing LNPs. They tested those predictions by using the new formulations to deliver mRNA encoding a fluorescent protein to mouse skin cells grown in a lab dish. They found that the LNPs predicted by the model did indeed work better than the particles in the training data, and in some cases better than LNP formulations that are used commercially.

Accelerated development

Once the researchers showed that the model could accurately predict particles that would efficiently deliver mRNA, they began asking additional questions. First, they wondered if they could train the model on nanoparticles that incorporate a fifth component: a type of polymer known as branched poly beta amino esters (PBAEs).

Research by Traverso and his colleagues has shown that these polymers can effectively deliver nucleic acids on their own, so they wanted to explore whether adding them to LNPs could improve LNP performance. The MIT team created a set of about 300 LNPs that also include these polymers, which they used to train the model. The resulting model could then predict additional formulations with PBAEs that would work better.

Next, the researchers set out to train the model to make predictions about LNPs that would work best in different types of cells, including a type of cell called Caco-2, which is derived from colorectal cancer cells. Again, the model was able to predict LNPs that would efficiently deliver mRNA to these cells.

Lastly, the researchers used the model to predict which LNPs could best withstand lyophilization — a freeze-drying process often used to extend the shelf-life of medicines.

“This is a tool that allows us to adapt it to a whole different set of questions and help accelerate development. We did a large training set that went into the model, but then you can do much more focused experiments and get outputs that are helpful on very different kinds of questions,” Traverso says.

He and his colleagues are now working on incorporating some of these particles into potential treatments for diabetes and obesity, which are two of the primary targets of the ARPA-H funded project. Therapeutics that could be delivered using this approach include GLP-1 mimics with similar effects to Ozempic.

This research was funded by the GO Nano Marble Center at the Koch Institute, the Karl van Tassel Career Development Professorship, the MIT Department of Mechanical Engineering, Brigham and Women’s Hospital, and ARPA-H.

Graphite is a key structural component in some of the world’s oldest nuclear reactors and many of the next-generation designs being built today. But it also condenses and swells in response to radiation — and the mechanism behind those changes has proven difficult to study.

Now, MIT researchers and collaborators have uncovered a link between properties of graphite and how the material behaves in response to radiation. The findings could lead to more accurate, less destructive ways of predicting the lifespan of graphite materials used in reactors around the world.

“We did some basic science to understand what leads to swelling and, eventually, failure in graphite structures,” says MIT Research Scientist Boris Khaykovich, senior author of the new study. “More research will be needed to put this into practice, but the paper proposes an attractive idea for industry: that you might not need to break hundreds of irradiated samples to understand their failure point.”

Specifically, the study shows a connection between the size of the pores within graphite and the way the material swells and shrinks in volume, leading to degradation.

“The lifetime of nuclear graphite is limited by irradiation-induced swelling,” says co-author and MIT Research Scientist Lance Snead. “Porosity is a controlling factor in this swelling, and while graphite has been extensively studied for nuclear applications since the Manhattan Project, we still do not have a clear understanding of the porosity in both mechanical properties and swelling. This work addresses that.”

The open-access paper appears this week in Interdisciplinary Materials. It is co-authored by Khaykovich, Snead, MIT Research Scientist Sean Fayfar, former MIT research fellow Durgesh Rai, Stony Brook University Assistant Professor David Sprouster, Oak Ridge National Laboratory Staff Scientist Anne Campbell, and Argonne National Laboratory Physicist Jan Ilavsky.

A long-studied, complex material

Ever since 1942, when physicists and engineers built the world’s first nuclear reactor on a converted squash court at the University of Chicago, graphite has played a central role in the generation of nuclear energy. That first reactor, dubbed the Chicago Pile, was constructed from about 40,000 graphite blocks, many of which contained nuggets of uranium.

Today graphite is a vital component of many operating nuclear reactors and is expected to play a central role in next-generation reactor designs like molten-salt and high-temperature gas reactors. That’s because graphite is a good neutron moderator, slowing down the neutrons released by nuclear fission so they are more likely to create fissions themselves and sustain a chain reaction.

“The simplicity of graphite makes it valuable,” Khaykovich explains. “It’s made of carbon, and it’s relatively well-known how to make it cleanly. Graphite is a very mature technology. It’s simple, stable, and we know it works.”

But graphite also has its complexities.

“We call graphite a composite even though it’s made up of only carbon atoms,” Khaykovich says. “It includes ‘filler particles’ that are more crystalline, then there is a matrix called a ‘binder’ that is less crystalline, then there are pores that span in length from nanometers to many microns.”

Each graphite grade has its own composite structure, but they all contain fractals, or shapes that look the same at different scales.

Those complexities have made it hard to predict how graphite will respond to radiation in microscopic detail, although it’s been known for decades that when graphite is irradiated, it first densifies, reducing its volume by up to 10 percent, before swelling and cracking. The volume fluctuation is caused by changes to graphite’s porosity and lattice stress.

“Graphite deteriorates under radiation, as any material does,” Khaykovich says. “So, on the one hand we have a material that’s extremely well-known, and on the other hand, we have a material that is immensely complicated, with a behavior that’s impossible to predict through computer simulations.”

For the study, the researchers received irradiated graphite samples from Oak Ridge National Laboratory. Co-authors Campbell and Snead were involved in irradiating the samples some 20 years ago. The samples are a grade of graphite known as G347A.

The research team used an analysis technique known as X-ray scattering, which uses the scattered intensity of an X-ray beam to analyze the properties of material. Specifically, they looked at the distribution of sizes and surface areas of the sample’s pores, or what are known as the material’s fractal dimensions.

“When you look at the scattering intensity, you see a large range of porosity,” Fayfar says. “Graphite has porosity over such large scales, and you have this fractal self-similarity: The pores in very small sizes look similar to pores spanning microns, so we used fractal models to relate different morphologies across length scales.”

Fractal models had been used on graphite samples before, but not on irradiated samples to see how the material’s pore structures changed. The researchers found that when graphite is first exposed to radiation, its pores get filled as the material degrades.

“But what was quite surprising to us is the [size distribution of the pores] turned back around,” Fayfar says. “We had this recovery process that matched our overall volume plots, which was quite odd. It seems like after graphite is irradiated for so long, it starts recovering. It’s sort of an annealing process where you create some new pores, then the pores smooth out and get slightly bigger. That was a big surprise.”

The researchers found that the size distribution of the pores closely follows the volume change caused by radiation damage.

“Finding a strong correlation between the [size distribution of pores] and the graphite’s volume changes is a new finding, and it helps connect to the failure of the material under irradiation,” Khaykovich says. “It’s important for people to know how graphite parts will fail when they are under stress and how failure probability changes under irradiation.”

From research to reactors

The researchers plan to study other graphite grades and explore further how pore sizes in irradiated graphite correlate with the probability of failure. They speculate that a statistical technique known as the Weibull Distribution could be used to predict graphite’s time until failure. The Weibull Distribution is already used to describe the probability of failure in ceramics and other porous materials like metal alloys.

Khaykovich also speculated that the findings could contribute to our understanding of why materials densify and swell under irradiation.

“There’s no quantitative model of densification that takes into account what’s happening at these tiny scales in graphite,” Khaykovich says. “Graphite irradiation densification reminds me of sand or sugar, where when you crush big pieces into smaller grains, they densify. For nuclear graphite, the crushing force is the energy that neutrons bring in, causing large pores to get filled with smaller, crushed pieces. But more energy and agitation create still more pores, and so graphite swells again. It’s not a perfect analogy, but I believe analogies bring progress for understanding these materials.”

The researchers describe the paper as an important step toward informing graphite production and use in nuclear reactors of the future.

“Graphite has been studied for a very long time, and we’ve developed a lot of strong intuitions about how it will respond in different environments, but when you’re building a nuclear reactor, details matter,” Khaykovich says. “People want numbers. They need to know how much thermal conductivity will change, how much cracking and volume change will happen. If components are changing volume, at some point you need to take that into account.”

This work was supported, in part, by the U.S. Department of Energy.

With help from artificial intelligence, MIT researchers have designed novel antibiotics that can combat two hard-to-treat infections: drug-resistant Neisseria gonorrhoeae and multi-drug-resistant Staphylococcus aureus (MRSA).

Using generative AI algorithms, the research team designed more than 36 million possible compounds and computationally screened them for antimicrobial properties. The top candidates they discovered are structurally distinct from any existing antibiotics, and they appear to work by novel mechanisms that disrupt bacterial cell membranes.

This approach allowed the researchers to generate and evaluate theoretical compounds that have never been seen before — a strategy that they now hope to apply to identify and design compounds with activity against other species of bacteria.

“We’re excited about the new possibilities that this project opens up for antibiotics development. Our work shows the power of AI from a drug design standpoint, and enables us to exploit much larger chemical spaces that were previously inaccessible,” says James Collins, the Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering.

Collins is the senior author of the study, which appears today in Cell. The paper’s lead authors are MIT postdoc Aarti Krishnan, former postdoc Melis Anahtar ’08, and Jacqueline Valeri PhD ’23.

Exploring chemical space

Over the past 45 years, a few dozen new antibiotics have been approved by the FDA, but most of these are variants of existing antibiotics. At the same time, bacterial resistance to many of these drugs has been growing. Globally, it is estimated that drug-resistant bacterial infections cause nearly 5 million deaths per year.

In hopes of finding new antibiotics to fight this growing problem, Collins and others at MIT’s Antibiotics-AI Project have harnessed the power of AI to screen huge libraries of existing chemical compounds. This work has yielded several promising drug candidates, including halicin and abaucin.

To build on that progress, Collins and his colleagues decided to expand their search into molecules that can’t be found in any chemical libraries. By using AI to generate hypothetically possible molecules that don’t exist or haven’t been discovered, they realized that it should be possible to explore a much greater diversity of potential drug compounds.

In their new study, the researchers employed two different approaches: First, they directed generative AI algorithms to design molecules based on a specific chemical fragment that showed antimicrobial activity, and second, they let the algorithms freely generate molecules, without having to include a specific fragment.

For the fragment-based approach, the researchers sought to identify molecules that could kill N. gonorrhoeae, a Gram-negative bacterium that causes gonorrhea. They began by assembling a library of about 45 million known chemical fragments, consisting of all possible combinations of 11 atoms of carbon, nitrogen, oxygen, fluorine, chlorine, and sulfur, along with fragments from Enamine’s REadily AccessibLe (REAL) space.

Then, they screened the library using machine-learning models that Collins’ lab has previously trained to predict antibacterial activity against N. gonorrhoeae. This resulted in nearly 4 million fragments. They narrowed down that pool by removing any fragments predicted to be cytotoxic to human cells, displayed chemical liabilities, and were known to be similar to existing antibiotics. This left them with about 1 million candidates.

“We wanted to get rid of anything that would look like an existing antibiotic, to help address the antimicrobial resistance crisis in a fundamentally different way. By venturing into underexplored areas of chemical space, our goal was to uncover novel mechanisms of action,” Krishnan says.

Through several rounds of additional experiments and computational analysis, the researchers identified a fragment they called F1 that appeared to have promising activity against N. gonorrhoeae. They used this fragment as the basis for generating additional compounds, using two different generative AI algorithms.

One of those algorithms, known as chemically reasonable mutations (CReM), works by starting with a particular molecule containing F1 and then generating new molecules by adding, replacing, or deleting atoms and chemical groups. The second algorithm, F-VAE (fragment-based variational autoencoder), takes a chemical fragment and builds it into a complete molecule. It does so by learning patterns of how fragments are commonly modified, based on its pretraining on more than 1 million molecules from the ChEMBL database.

Those two algorithms generated about 7 million candidates containing F1, which the researchers then computationally screened for activity against N. gonorrhoeae. This screen yielded about 1,000 compounds, and the researchers selected 80 of those to see if they could be produced by chemical synthesis vendors. Only two of these could be synthesized, and one of them, named NG1, was very effective at killing N. gonorrhoeae in a lab dish and in a mouse model of drug-resistant gonorrhea infection.

Additional experiments revealed that NG1 interacts with a protein called LptA, a novel drug target involved in the synthesis of the bacterial outer membrane. It appears that the drug works by interfering with membrane synthesis, which is fatal to cells.

Unconstrained design

In a second round of studies, the researchers explored the potential of using generative AI to freely design molecules, using Gram-positive bacteria, S. aureus as their target.

Again, the researchers used CReM and VAE to generate molecules, but this time with no constraints other than the general rules of how atoms can join to form chemically plausible molecules. Together, the models generated more than 29 million compounds. The researchers then applied the same filters that they did to the N. gonorrhoeae candidates, but focusing on S. aureus, eventually narrowing the pool down to about 90 compounds.

They were able to synthesize and test 22 of these molecules, and six of them showed strong antibacterial activity against multi-drug-resistant S. aureus grown in a lab dish. They also found that the top candidate, named DN1, was able to clear a methicillin-resistant S. aureus (MRSA) skin infection in a mouse model. These molecules also appear to interfere with bacterial cell membranes, but with broader effects not limited to interaction with one specific protein.

Phare Bio, a nonprofit that is also part of the Antibiotics-AI Project, is now working on further modifying NG1 and DN1 to make them suitable for additional testing.

“In a collaboration with Phare Bio, we are exploring analogs, as well as working on advancing the best candidates preclinically, through medicinal chemistry work,” Collins says. “We are also excited about applying the platforms that Aarti and the team have developed toward other bacterial pathogens of interest, notably Mycobacterium tuberculosis and Pseudomonas aeruginosa.”

The research was funded, in part, by the U.S. Defense Threat Reduction Agency, the National Institutes of Health, the Audacious Project, Flu Lab, the Sea Grape Foundation, Rosamund Zander and Hansjorg Wyss for the Wyss Foundation, and an anonymous donor.

Is this movie review a rave or a pan? Is this news story about business or technology? Is this online chatbot conversation veering off into giving financial advice? Is this online medical information site giving out misinformation?

These kinds of automated conversations, whether they involve seeking a movie or restaurant review or getting information about your bank account or health records, are becoming increasingly prevalent. More than ever, such evaluations are being made by highly sophisticated algorithms, known as text classifiers, rather than by human beings. But how can we tell how accurate these classifications really are?

Now, a team at MIT’s Laboratory for Information and Decision Systems (LIDS) has come up with an innovative approach to not only measure how well these classifiers are doing their job, but then go one step further and show how to make them more accurate.

The new evaluation and remediation software was developed by Kalyan Veeramachaneni, a principal research scientist at LIDS, his students Lei Xu and Sarah Alnegheimish, and two others. The software package is being made freely available for download by anyone who wants to use it.

A standard method for testing these classification systems is to create what are known as synthetic examples — sentences that closely resemble ones that have already been classified. For example, researchers might take a sentence that has already been tagged by a classifier program as being a rave review, and see if changing a word or a few words while retaining the same meaning could fool the classifier into deeming it a pan. Or a sentence that was determined to be misinformation might get misclassified as accurate. This ability to fool the classifiers makes these adversarial examples.

People have tried various ways to find the vulnerabilities in these classifiers, Veeramachaneni says. But existing methods of finding these vulnerabilities have a hard time with this task and miss many examples that they should catch, he says.

Increasingly, companies are trying to use such evaluation tools in real time, monitoring the output of chatbots used for various purposes to try to make sure they are not putting out improper responses. For example, a bank might use a chatbot to respond to routine customer queries such as checking account balances or applying for a credit card, but it wants to ensure that its responses could never be interpreted as financial advice, which could expose the company to liability. “Before showing the chatbot’s response to the end user, they want to use the text classifier to detect whether it’s giving financial advice or not,” Veeramachaneni says. But then it’s important to test that classifier to see how reliable its evaluations are.

“These chatbots, or summarization engines or whatnot are being set up across the board,” he says, to deal with external customers and within an organization as well, for example providing information about HR issues. It’s important to put these text classifiers into the loop to detect things that they are not supposed to say, and filter those out before the output gets transmitted to the user.

That’s where the use of adversarial examples comes in — those sentences that have already been classified but then produce a different response when they are slightly modified while retaining the same meaning. How can people confirm that the meaning is the same? By using another large language model (LLM) that interprets and compares meanings. So, if the LLM says the two sentences mean the same thing, but the classifier labels them differently, “that is a sentence that is adversarial — it can fool the classifier,” Veeramachaneni says. And when the researchers examined these adversarial sentences, “we found that most of the time, this was just a one-word change,” although the people using LLMs to generate these alternate sentences often didn’t realize that.

Further investigation, using LLMs to analyze many thousands of examples, showed that certain specific words had an outsized influence in changing the classifications, and therefore the testing of a classifier’s accuracy could focus on this small subset of words that seem to make the most difference. They found that one-tenth of 1 percent of all the 30,000 words in the system’s vocabulary could account for almost half of all these reversals of classification, in some specific applications.

Lei Xu PhD ’23, a recent graduate from LIDS who performed much of the analysis as part of his thesis work, “used a lot of interesting estimation techniques to figure out what are the most powerful words that can change the overall classification, that can fool the classifier,” Veeramachaneni says. The goal is to make it possible to do much more narrowly targeted searches, rather than combing through all possible word substitutions, thus making the computational task of generating adversarial examples much more manageable. “He’s using large language models, interestingly enough, as a way to understand the power of a single word.”

Then, also using LLMs, he searches for other words that are closely related to these powerful words, and so on, allowing for an overall ranking of words according to their influence on the outcomes. Once these adversarial sentences have been found, they can be used in turn to retrain the classifier to take them into account, increasing the robustness of the classifier against those mistakes.

Making classifiers more accurate may not sound like a big deal if it’s just a matter of classifying news articles into categories, or deciding whether reviews of anything from movies to restaurants are positive or negative. But increasingly, classifiers are being used in settings where the outcomes really do matter, whether preventing the inadvertent release of sensitive medical, financial, or security information, or helping to guide important research, such as into properties of chemical compounds or the folding of proteins for biomedical applications, or in identifying and blocking hate speech or known misinformation.

As a result of this research, the team introduced a new metric, which they call p, which provides a measure of how robust a given classifier is against single-word attacks. And because of the importance of such misclassifications, the research team has made its products available as open access for anyone to use. The package consists of two components: SP-Attack, which generates adversarial sentences to test classifiers in any particular application, and SP-Defense, which aims to improve the robustness of the classifier by generating and using adversarial sentences to retrain the model.

In some tests, where competing methods of testing classifier outputs allowed a 66 percent success rate by adversarial attacks, this team’s system cut that attack success rate almost in half, to 33.7 percent. In other applications, the improvement was as little as a 2 percent difference, but even that can be quite important, Veeramachaneni says, since these systems are being used for so many billions of interactions that even a small percentage can affect millions of transactions.

The team’s results were published on July 7 in the journal Expert Systems in a paper by Xu, Veeramachaneni, and Alnegheimish of LIDS, along with Laure Berti-Equille at IRD in Marseille, France, and Alfredo Cuesta-Infante at the Universidad Rey Juan Carlos, in Spain. 

“Manufacturing is the engine of society, and it is the backbone of robust, resilient economies,” says John Hart, head of MIT’s Department of Mechanical Engineering (MechE) and faculty co-director of the MIT Initiative for New Manufacturing (INM). “With manufacturing a lively topic in today’s news, there’s a renewed appreciation and understanding of the importance of manufacturing to innovation, to economic and national security, and to daily lives.”

Launched this May, INM will “help create a transformation of manufacturing through new technology, through development of talent, and through an understanding of how to scale manufacturing in a way that enables imparts higher productivity and resilience, drives adoption of new technologies, and creates good jobs,” Hart says.

INM is one of MIT’s strategic initiatives and builds on the successful three-year-old Manufacturing@MIT program. “It’s a recognition by MIT that manufacturing is an Institute-wide theme and an Institute-wide priority, and that manufacturing connects faculty and students across campus,” says Hart. Alongside Hart, INM’s faculty co-directors are Institute Professor Suzanne Berger and Chris Love, professor of chemical engineering.

The initiative is pursuing four main themes: reimagining manufacturing technologies and systems, elevating the productivity and human experience of manufacturing, scaling up new manufacturing, and transforming the manufacturing base.

Breaking manufacturing barriers for corporations

Amgen, Autodesk, Flex, GE Vernova, PTC, Sanofi, and Siemens are founding members of INM’s industry consortium. These industry partners will work closely with MIT faculty, researchers, and students across many aspects of manufacturing-related research, both in broad-scale initiatives and in particular areas of shared interests. Membership requires a minimum three-year commitment of $500,000 a year to manufacturing-related activities at MIT, including the INM membership fee of $275,000 per year, which supports several core activities that engage the industry members.

One major thrust for INM industry collaboration is the deployment and adoption of AI and automation in manufacturing. This effort will include seed research projects at MIT, collaborative case studies, and shared strategy development.

INM also offers companies participation in the MIT-wide New Manufacturing Research effort, which is studying the trajectories of specific manufacturing industries and examining cross-cutting themes such as technology and financing.

Additionally, INM will concentrate on education for all professions in manufacturing, with alliances bringing together corporations, community colleges, government agencies, and other partners. “We'll scale our curriculum to broader audiences, from aspiring manufacturing workers and aspiring production line supervisors all the way up to engineers and executives,” says Hart.

In workforce training, INM will collaborate with companies broadly to help understand the challenges and frame its overall workforce agenda, and with individual firms on specific challenges, such as acquiring suitably prepared employees for a new factory.

Importantly, industry partners will also engage directly with students. Founding member Flex, for instance, hosted MIT researchers and students at the Flex Institute of Technology in Sorocaba, Brazil, developing new solutions for electronics manufacturing.

“History shows that you need to innovate in manufacturing alongside the innovation in products,” Hart comments. “At MIT, as more students take classes in manufacturing, they’ll think more about key manufacturing issues as they decide what research problems they want to solve, or what choices they make as they prototype their devices. The same is true for industry — companies that operate at the frontier of manufacturing, whether through internal capabilities or their supply chains, are positioned to be on the frontier of product innovation and overall growth.”

“We’ll have an opportunity to bring manufacturing upstream to the early stage of research, designing new processes and new devices with scalability in mind,” he says.

Additionally, MIT expects to open new manufacturing-related labs and to further broaden cooperation with industry at existing shared facilities, such as MIT.nano. Hart says that facilities will also invite tighter collaborations with corporations — not just providing advanced equipment, but working jointly on, say, new technologies for weaving textiles, or speeding up battery manufacturing.

Homing in on the United States

INM is a global project that brings a particular focus on the United States, which remains the world’s second-largest manufacturing economy, but has suffered a significant decline in manufacturing employment and innovation.

One key to reversing this trend and reinvigorating the U.S. manufacturing base is advocacy for manufacturing’s critical role in society and the career opportunities it offers.

“No one really disputes the importance of manufacturing,” Hart says. “But we need to elevate interest in manufacturing as a rewarding career, from the production workers to manufacturing engineers and leaders, through advocacy, education programs, and buy-in from industry, government, and academia.”

MIT is in a unique position to convene industry, academic, and government stakeholders in manufacturing to work together on this vital issue, he points out.

Moreover, in times of radical and rapid changes in manufacturing, “we need to focus on deploying new technologies into factories and supply chains,” Hart says. “Technology is not all of the solution, but for the U.S. to expand our manufacturing base, we need to do it with technology as a key enabler, embracing companies of all sizes, including small and medium enterprises.”

“As AI becomes more capable, and automation becomes more flexible and more available, these are key building blocks upon which you can address manufacturing challenges,” he says. “AI and automation offer new accelerated ways to develop, deploy, and monitor production processes, which present a huge opportunity and, in some cases, a necessity.”

“While manufacturing is always a combination of old technology, new technology, established practice, and new ways of thinking, digital technology gives manufacturers an opportunity to leapfrog competitors,” Hart says. “That’s very, very powerful for the U.S. and any company, or country, that aims to create differentiated capabilities.”

Fortunately, in recent years, investors have increasingly bought into new manufacturing in the United States. “They see the opportunity to re-industrialize, to build the factories and production systems of the future,” Hart says.

“That said, building new manufacturing is capital-intensive, and takes time,” he adds. “So that’s another area where it’s important to convene stakeholders and to think about how startups and growth-stage companies build their capital portfolios, how large industry can support an ecosystem of small businesses and young companies, and how to develop talent to support those growing companies.”

All these concerns and opportunities in the manufacturing ecosystem play to MIT’s strengths. “MIT’s DNA of cross-disciplinary collaboration and working with industry can let us create a lot of impact,” Hart emphasizes. “We can understand the practical challenges. We can also explore breakthrough ideas in research and cultivate successful outcomes, all the way to new companies and partnerships. Sometimes those are seen as disparate approaches, but we like to bring them together.”

“It’s probably the hardest thing I’ve ever done at MIT,” says Haley Nakamura, a second-year MEng student in the MIT Department of Electrical Engineering and Computer Science (EECS). She’s not reflecting on a class, final exam, or research paper. Nakamura is talking about the experience of being a teaching assistant (TA). “It’s really an art form, in that there is no formula for being a good teacher. It’s a skill, and something you have to continuously work at and adapt to different people.”

Nakamura, like approximately 16 percent of her EECS MEng peers, balances her own coursework with teaching responsibilities. The TA role is complex, nuanced, and at MIT, can involve much more planning and logistics than you might imagine. Nakamura works on a central computer science (CS) course, 6.3900 (Introduction to Machine Learning), which registers around 400-500 students per semester. For that enrollment, the course requires eight instructors at the lecturer/professor level; 15 TAs, between the undergraduate and graduate level; and about 50 lab assistants (LAs). Students are split across eight sections corresponding to each senior instructor, with a group of TAs and LAs for each section of 60-70 students.

To keep everyone moving forward at the same pace, coordination and organization are key. “A lot of the reason I got my initial TA-ship was because I was pretty organized,” Nakamura explains. “Everyone here at MIT can be so busy that it can be difficult to be on top of things, and students will be the first to point out logistical confusion and inconsistencies. If they’re worried about some quirk on the website, or wondering how their grades are being calculated, those things can prevent them from focusing on content.” 

Nakamura's organizational skills made her a good candidate to spot and deal with potential wrinkles before they derailed a course section. “When I joined the course, we wanted someone on the TA side to be more specifically responsible for underlying administrative tasks, so I became the first head TA for the course. Since then, we’ve built that role up more and more. There is now a head TA, a head undergraduate TA, and section leads working on internal documentation such as instructions for how to improve content and how to manage office hours.” The result of this administrative work is consistency across sections and semesters.

The other side of a TA-ship is, of course, teaching. “I was eager to engage with students in a meaningful way,” says Soroush Araei, a sixth-year graduate student who had already fulfilled the teaching requirement for his degree in electrical engineering, but who jumped at the chance to teach alongside his PhD advisor. “I enjoy teaching, and have always found that explaining concepts to others deepens my own understanding.” He was recently awarded the ​MIT School of Engineering’s 2025 Graduate Student Teaching and Mentoring Award, which honors “a graduate student in the School of Engineering who has demonstrated extraordinary teaching and mentoring as a teaching or research assistant.” Araei’s dedication comes at the price of sleep. “Juggling my own research with my TA duties was no small feat. I often found myself in the lab for long hours, helping students troubleshoot their circuits. While their design simulations looked perfect, the circuits they implemented on protoboards didn’t always perform as expected. I had to dive deep into the issues alongside the students, which often required considerable time and effort.”

The rewards for Araei’s work are often intrinsic. “Teaching has shown me that there are always deeper layers to understanding. There are concepts I thought I had mastered, but I realized gaps in my own knowledge when trying to explain them,” he says. Another challenge: the variety of background knowledge between students in a single class. “Some had never encountered transistors, while others had tape-out experience. Designing problem sets and selecting questions for office hours required careful planning to keep all students engaged.” For Araei, some of the best moments have come during office hours. “Witnessing the ‘aha’ moment on a student’s face when a complex concept finally clicked was incredibly rewarding.”

The pursuit of the “aha” moment is a common thread between TAs. “I still struggle with the feeling that you’re responsible for someone’s understanding in a given topic, and, if you’re not doing a good job, that could affect that person for the rest of their life,” says Nakamura. “But the flip side of that moment of confusion is when someone has the ‘aha!’ moment as you’re talking to them, when you’re able to explain something that wasn’t conveyed in the other materials. It was your help that broke through and gave understanding. And that reward really overruns the fear of causing confusion.”

Hope Dargan ’21, MEng ’23, a second-year PhD student in EECS, uses her role as a graduate instructor to try to reach students who may not fit into the stereotype of the scientist. She started her career at MIT planning to major in CS and become a software engineer, but a missionary trip to Sweden in 2016-17 (when refugees from the Syrian civil war were resettling in the region) sparked a broader interest in both the Middle East and in how groups of people contextualized their own narratives. When Dargan returned to MIT, she took on a history degree, writing her thesis on the experiences of queer Mormon women. Additionally, she taught for MEET (the Middle East Entrepreneurs of Tomorrow), an educational initiative for Israeli and Palestinian high school students. “I realized I loved teaching, and this experience set me on a trajectory to teaching as a career.” 

Dargan gained her teaching license as an undergrad through the MIT Scheller Teacher Education Program (STEP), then joined the MEng program, in which she designed an educational intervention for students who were struggling in class 6.101 (Fundamentals of Programming). The next step was a PhD. “Teaching is so context-dependent,” says Dargan, who was awarded the Goodwin Medal for her teaching efforts in 2023. “When I taught students for MEET, it was very different from when I was teaching eighth graders at Josiah Quincy Upper School for my teaching license, and very different now when I teach students in 6.101, versus when I teach the LGO [Leaders for Global Operations] students Python in the summers. Each student has their own unique perspective on what’s motivating them, how they learn, and what they connect to … So even if I’ve taught the material for five years (as I have for 6.101, because I was an LA, then a TA, and now an instructor), improving my teaching is always challenging. Getting better at adapting my teaching to the context of the students and their stories, which are ever-evolving, is always interesting.”

Although Dargan considers teaching one of her greatest passions, she is clear-eyed about the cost of the profession. “I think the things that we’re passionate about tell us a lot about ourselves, both our strengths and our weaknesses, and teaching has taught me a lot about my weaknesses,” she says. “Teaching is a tough career, because it tends to take people who care a lot and are perfectionists, and it can lead to a lot of burnout.”

Dargan's students have also expressed enthusiasm and gratitude for her work. “Hope is objectively the most helpful instructor I’ve ever had,” said one anonymous reviewer. Another wrote, “I never felt judged when I asked her questions, and she was great at guiding me through problems by asking motivating questions … I truly felt like she cared about me as a student and person.” Dargan herself is modest about her role, saying, “For me, the trade-off between teaching and research is that teaching has an immediate day-to-day impact, while research has this unknown potential for long-term impact.” 

With the responsibility to instruct an ever-growing percentage of the Institute’s students, the Department of Electrical Engineering and Computer Science relies heavily on dedicated and passionate students like Nakamura, Araei, and Dargan. As their caring and humane influence ripples outward through thousands of new electrical engineers and computer scientists, the day-to-day impact of their work is clear; but the long-term impact may be greater than any of them know.

Around the world, about 2 billion people suffer from iron deficiency, which can lead to anemia, impaired brain development in children, and increased infant mortality.

To combat that problem, MIT researchers have come up with a new way to fortify foods and beverages with iron, using small crystalline particles. These particles, known as metal-organic frameworks, could be sprinkled on food, added to staple foods such as bread, or incorporated into drinks like coffee and tea.

“We’re creating a solution that can be seamlessly added to staple foods across different regions,” says Ana Jaklenec, a principal investigator at MIT’s Koch Institute for Integrative Cancer Research. “What’s considered a staple in Senegal isn’t the same as in India or the U.S., so our goal was to develop something that doesn’t react with the food itself. That way, we don’t have to reformulate for every context — it can be incorporated into a wide range of foods and beverages without compromise.”

The particles designed in this study can also carry iodine, another critical nutrient. The particles could also be adapted to carry important minerals such as zinc, calcium, or magnesium.

“We are very excited about this new approach and what we believe is a novel application of metal-organic frameworks to potentially advance nutrition, particularly in the developing world,” says Robert Langer, the David H. Koch Institute Professor at MIT and a member of the Koch Institute.

Jaklenec and Langer are the senior authors of the study, which appears today in the journal Matter. MIT postdoc Xin Yang and Linzixuan (Rhoda) Zhang PhD ’24 are the lead authors of the paper.

Iron stabilization

Food fortification can be a successful way to combat nutrient deficiencies, but this approach is often challenging because many nutrients are fragile and break down during storage or cooking. When iron is added to foods, it can react with other molecules in the food, giving the food a metallic taste.

In previous work, Jaklenec’s lab has shown that encapsulating nutrients in polymers can protect them from breaking down or reacting with other molecules. In a small clinical trial, the researchers found that women who ate bread fortified with encapsulated iron were able to absorb the iron from the food.

However, one drawback to this approach is that the polymer adds a lot of bulk to the material, limiting the amount of iron or other nutrients that end up in the food.

“Encapsulating iron in polymers significantly improves its stability and reactivity, making it easier to add to food,” Jaklenec says. “But to be effective, it requires a substantial amount of polymer. That limits how much iron you can deliver in a typical serving, making it difficult to meet daily nutritional targets through fortified foods alone.”

To overcome that challenge, Yang came up with a new idea: Instead of encapsulating iron in a polymer, they could use iron itself as a building block for a crystalline particle known as a metal-organic framework, or MOF (pronounced “moff”).

MOFs consist of metal atoms joined by organic molecules called ligands to create a rigid, cage-like structure. Depending on the combination of metals and ligands chosen, they can be used for a wide variety of applications.

“We thought maybe we could synthesize a metal-organic framework with food-grade ligands and food-grade micronutrients,” Yang says. “Metal-organic frameworks have very high porosity, so they can load a lot of cargo. That’s why we thought we could leverage this platform to make a new metal-organic framework that could be used in the food industry.”

In this case, the researchers designed a MOF consisting of iron bound to a ligand called fumaric acid, which is often used as a food additive to enhance flavor or help preserve food.

This structure prevents iron from reacting with polyphenols — compounds commonly found in foods such as whole grains and nuts, as well as coffee and tea. When iron does react with those compounds, it forms a metal polyphenol complex that cannot be absorbed by the body.

The MOFs’ structure also allows them to remain stable until they reach an acidic environment, such as the stomach, where they break down and release their iron payload.

Double-fortified salts

The researchers also decided to include iodine in their MOF particle, which they call NuMOF. Iodized salt has been very successful at preventing iodine deficiency, and many efforts are now underway to create “double-fortified salts” that would also contain iron.

Delivering these nutrients together has proven difficult because iron and iodine can react with each other, making each one less likely to be absorbed by the body. In this study, the MIT team showed that once they formed their iron-containing MOF particles, they could load them with iodine, in a way that the iron and iodine do not react with each other.

In tests of the particles’ stability, the researchers found that the NuMOFs could withstand long-term storage, high heat and humidity, and boiling water.

Throughout these tests, the particles maintained their structure. When the researchers then fed the particles to mice, they found that both iron and iodine became available in the bloodstream within several hours of the NuMOF consumption.

The researchers are now working on launching a company that is developing coffee and other beverages fortified with iron and iodine. They also hope to continue working toward a double-fortified salt that could be consumed on its own or incorporated into staple food products.

The research was partially supported by J-WAFS Fellowships for Water and Food Solutions.

Other authors of the paper include Fangzheng Chen, Wenhao Gao, Zhiling Zheng, Tian Wang, Erika Yan Wang, Behnaz Eshaghi, and Sydney MacDonald.

Jessika Trancik, a professor in MIT’s Institute for Data, Systems, and Society, has been named the new director of the Sociotechnical Systems Research Center (SSRC), effective July 1. The SSRC convenes and supports researchers focused on problems and solutions at the intersection of technology and its societal impacts.

Trancik conducts research on technology innovation and energy systems. At the Trancik Lab, she and her team develop methods drawing on engineering knowledge, data science, and policy analysis. Their work examines the pace and drivers of technological change, helping identify where innovation is occurring most rapidly, how emerging technologies stack up against existing systems, and which performance thresholds matter most for real-world impact. Her models have been used to inform government innovation policy and have been applied across a wide range of industries.

“Professor Trancik’s deep expertise in the societal implications of technology, and her commitment to developing impactful solutions across industries, make her an excellent fit to lead SSRC,” says Maria C. Yang, interim dean of engineering and William E. Leonhard (1940) Professor of Mechanical Engineering.

Much of Trancik’s research focuses on the domain of energy systems, and establishing methods for energy technology evaluation, including of their costs, performance, and environmental impacts. She covers a wide range of energy services — including electricity, transportation, heating, and industrial processes. Her research has applications in solar and wind energy, energy storage, low-carbon fuels, electric vehicles, and nuclear fission. Trancik is also known for her research on extreme events in renewable energy availability.

A prolific researcher, Trancik has helped measure progress and inform the development of solar photovoltaics, batteries, electric vehicle charging infrastructure, and other low-carbon technologies — and anticipate future trends. One of her widely cited contributions includes quantifying learning rates and identifying where targeted investments can most effectively accelerate innovation. These tools have been used by U.S. federal agencies, international organizations, and the private sector to shape energy R&D portfolios, climate policy, and infrastructure planning.

Trancik is committed to engaging and informing the public on energy consumption. She and her team developed the app carboncounter.com, which helps users choose cars with low costs and low environmental impacts.

As an educator, Trancik teaches courses for students across MIT’s five schools and the MIT Schwarzman College of Computing.

“The question guiding my teaching and research is how do we solve big societal challenges with technology, and how can we be more deliberate in developing and supporting technologies to get us there?” Trancik said in an article about course IDS.521/IDS.065 (Energy Systems for Climate Change Mitigation).

Trancik received her undergraduate degree in materials science and engineering from Cornell University. As a Rhodes Scholar, she completed her PhD in materials science at the University of Oxford. She subsequently worked for the United Nations in Geneva, Switzerland, and the Earth Institute at Columbia University. After serving as an Omidyar Research Fellow at the Santa Fe Institute, she joined MIT in 2010 as a faculty member.

Trancik succeeds Fotini Christia, the Ford International Professor of Social Sciences in the Department of Political Science and director of IDSS, who previously served as director of SSRC.

Harvey Kent Bowen PhD ’71, a longtime MIT professor celebrated for his pioneering work in manufacturing education, innovative ceramics research, and generous mentorship, died July 17 in Belmont, Massachusetts. He was 83.

At MIT, he was the founding engineering faculty leader of Leaders for Manufacturing (LFM) — now Leaders for Global Operations (LGO) — a program that continues to shape engineering and management education nearly four decades later.

Bowen spent 22 years on the MIT faculty, returning to his alma mater after earning both a master’s degree in materials science and a PhD in materials science and ceramics processing there. He held the Ford Professorship of Engineering, with appointments in the departments of Materials Science and Engineering (DMSE) and Electrical Engineering and Computer Science, before transitioning to Harvard Business School, where he bridged the worlds of engineering, manufacturing, and management. 

Bowen’s prodigious research output spans 190 articles, 45 Harvard case studies, and two books. In addition to his scholarly contributions, those who knew him best say his visionary understanding of the connection between management and engineering, coupled with his intellect and warm leadership style, set him apart at a time of rapid growth at MIT.  

A pioneering physical ceramics researcher

Bowen was born on Nov. 21, 1941, in Salt Lake City, Utah. As an MIT graduate student in the 1970s, he helped to redefine the study of ceramics — transforming it into the scientific field now known as physical ceramics, which focuses on the structure, properties, and behavior of ceramic materials.

“Prior to that, it was the art of ceramic composition,” says Michael Cima, the David H. Koch Professor of Engineering in DMSE. “What Kent and a small group of more-senior DMSE faculty were doing was trying to turn that art into science.”

Bowen advanced the field by applying scientific rigor to how ceramic materials were processed. He applied concepts from the developing field of colloid science — the study of particles evenly distributed in another material — to the manufacturing of ceramics, forever changing how such objects were made.

“That sparked a whole new generation of people taking a different look at how ceramic objects are manufactured,” Cima recalls. “It was an opportunity to make a big change. Despite the fact that physical ceramics — composition, crystal structure and so forth — had turned into a science, there still was this big gap: how do you make these things? Kent thought this was the opportunity for science to have an impact on the field of ceramics.”

One of his greatest scholarly accomplishments was “Introduction to Ceramics, 2nd edition,” with David Kingery and Donald Uhlmann, a foundational textbook he helped write early in his career. The book, published in 1976, helped maintain DMSE’s leading position in ceramics research and education.

“Every PhD student in ceramics studied that book, all 1,000 pages, from beginning to end, to prepare for the PhD qualifying exams,” says Yet-Ming Chiang, Kyocera Professor of Ceramics in DMSE. “It covered almost every aspect of the science and engineering of ceramics known at that time. That was why it was both an outstanding teaching text as well as a reference textbook for data.”

In ceramics processing, Bowen was also known for his control of particle size, shape, and size distribution, and how those factors influence sintering, the process of forming solid materials from powders.

Over time, Bowen’s interest in ceramics processing broadened into a larger focus on manufacturing. As such, Bowen was also deeply connected to industry and traveled frequently, especially to Japan, a leader in ceramics manufacturing.

“One time, he came back from Japan and told all of us graduate students that the students there worked so hard they were sleeping in the labs at night — as a way to prod us,” Chiang recalls.

While Bowen’s work in manufacturing began in ceramics, he also became a consultant to major companies, including automakers, and he worked with Lee Iacocca, the Ford executive behind the Mustang. Those experiences also helped spark LFM, which evolved into LGO. Bowen co-founded LFM with former MIT dean of engineering Tom Magnanti.

“I’m still in awe of Kent’s audacity and vision in starting the LFM program. The scale and scope of the program were, even for MIT standards, highly ambitious. Thirty-seven successful years later, we all owe a great sense of gratitude to Kent,” says LGO Executive Director Thomas Roemer, a senior lecturer at the MIT Sloan School of Management.

Bowen as mentor, teacher

Bowen’s scientific leadership was matched by his personal influence. Colleagues recall him as a patient, thoughtful mentor who valued creativity and experimentation.

“He had a lot of patience, and I think students benefited from that patience. He let them go in the directions they wanted to — and then helped them out of the hole when their experiments didn’t work. He was good at that,” Cima says.

His discipline was another hallmark of his character. Chiang was an undergraduate and graduate student when Bowen was a faculty member. He fondly recalls his tendency to get up early, a source of amusement for his 3.01 (Kinetics of Materials) class.

“One time, some students played a joke on him. They got to class before him, set up an electric griddle, and cooked breakfast in the classroom before he arrived,” says Chiang. “When we all arrived, it smelled like breakfast.”

Bowen took a personal interest in Chiang’s career trajectory, arranging for him to spend a summer in Bowen’s lab through the Undergraduate Research Opportunities Program. Funded by the Department of Energy, the project explored magnetohydrodynamics: shooting a high-temperature plasma made from coal fly ash into a magnetic field between ceramic electrodes to generate electricity.

“My job was just to sift the fly ash, but it opened my eyes to energy research,” Chiang recalls.

Later, when Chiang was an assistant professor at MIT, Bowen served on his career development committee. He was both encouraging and pragmatic.

“He pushed me to get things done — to submit and publish papers at a time when I really needed the push,” Chiang says. “After all the happy talk, he would say, ‘OK, by what date are you going to submit these papers?’ And that was what I needed.”

After leaving MIT, Bowen joined Harvard Business School (HBS), where he wrote numerous detailed case studies, including one on A123 Systems, a battery company Chiang co-founded in 2001. 

“He was very supportive of our work to commercialize battery technology, and starting new companies in energy and materials,” Chiang says.

Bowen was also a devoted mentor for LFM/LGO students, even while at HBS. Greg Dibb MBA ’04, SM ’04 recalls that Bowen agreed to oversee his work on the management philosophy known as the Toyota Production System (TPS) — a manufacturing system developed by the Japanese automaker — responding kindly to the young student’s outreach and inspiring him with methodical, real-world advice.

“By some miracle, he agreed and made the time to guide me on my thesis work. In the process, he became a mentor and a lifelong friend,” Dibb says. “He inspired me in his way of working and collaborating. He was a master thinker and listener, and he taught me by example through his Socratic style, asking me simple but difficult questions that required rigor of thought.

“I remember he asked me about my plan to learn about manufacturing and TPS. I came to him enthusiastically with a list of books I planned to read. He responded, ‘Do you think a world expert would read those books?’”   

In trying to answer that question, Dibb realized the best way to learn was to go to the factory floor.

“He had a passion for the continuous improvement of manufacturing and operations, and he taught me how to do it by being an observer and a listener just like him — all the time being inspired by his optimism, faith, and charity toward others.”

Faith was a cornerstone of Bowen’s life outside of academia. He served a mission for The Church of Jesus Christ of Latter-day Saints in the Central Germany Mission and held several leadership roles, including bishop of the Cambridge, Massachusetts Ward, stake president of the Cambridge Stake, mission president of the Tacoma, Washington Mission, and temple president of the Boston, Massachusetts Temple. 

An enthusiastic role model who inspired excellence

During early-morning conversations, Cima learned about Bowen’s growing interest in manufacturing, which would spur what is now LGO. Bowen eventually became recognized as an expert in the Toyota Production System, the company’s operational culture and practice which was a major influence on the LGO program’s curriculum design.

“I got to hear it from him — I was exposed to his early insights,” Cima says. “The fact that he would take the time every morning to talk to me — it was a huge influence.”

Bowen was a natural leader and set an example for others, Cima says.

“What is a leader? A leader is somebody who has the kind of infectious enthusiasm to convince others to work with them. Kent was really good at that,” Cima says. “What’s the way you learn leadership? Well, you’d look at how leaders behave. And really good leaders behave like Kent Bowen.”

MIT Sloan School of Management professor of the practice Zeynep Ton praises Bowen’s people skills and work ethic: “When you combine his belief in people with his ability to think big, something magical happens through the people Kent mentored. He always pushed us to do more,” Ton recalls. “Whenever I shared with Kent my research making an impact on a company, or my teaching making an impact on a student, his response was never just ‘good job.’ His next question was: ‘How can you make a bigger impact? Do you have the resources at MIT to do it? Who else can help you?’” 

A legacy of encouragement and drive

With this drive to do more, Bowen embodied MIT’s ethos, colleagues say.

“Kent Bowen embodies the MIT 'mens et manus' ['mind and hand'] motto professionally and personally as an inveterate experimenter in the lab, in the classroom, as an advisor, and in larger society,” says MIT Sloan senior lecturer Steve Spear. “Kent’s consistency was in creating opportunities to help people become their fullest selves, not only finding expression for their humanity greater than they could have achieved on their own, but greater than they might have even imagined on their own. An extraordinary number of people are directly in his debt because of this personal ethos — and even more have benefited from the ripple effect.”

Gregory Dibb, now a leader in the autonomous vehicle industry, is just one of them.

“Upon hearing of his passing, I immediately felt that I now have even more responsibility to step up and try to fill his shoes in sacrificing and helping others as he did — even if that means helping an unprepared and overwhelmed LGO grad student like me,” Dibb says.

Bowen is survived by his wife, Kathy Jones; his children, Natalie, Jennifer Patraiko, Melissa, Kirsten, and Jonathan; his sister, Kathlene Bowen; and six grandchildren. 

Jason Sparapani contributed to this article.

Water is essential for life on Earth. So, the liquid must be a requirement for life on other worlds. For decades, scientists’ definition of habitability on other planets has rested on this assumption.

But what makes some planets habitable might have very little to do with water. In fact, an entirely different type of liquid could conceivably support life in worlds where water can barely exist. That’s a possibility that MIT scientists raise in a study appearing this week in the Proceedings of the National Academy of Sciences.

From lab experiments, the researchers found that a type of fluid known as an ionic liquid can readily form from chemical ingredients that are also expected to be found on the surface of some rocky planets and moons. Ionic liquids are salts that exist in liquid form below about 100 degrees Celsius. The team’s experiments showed that a mixture of sulfuric acid and certain nitrogen-containing organic compounds produced such a liquid. On rocky planets, sulfuric acid may be a byproduct of volcanic activity, while nitrogen-containing compounds have been detected on several asteroids and planets in our solar system, suggesting the compounds may be present in other planetary systems.

Ionic liquids have extremely low vapor pressure and do not evaporate; they can form and persist at higher temperatures and lower pressures than what liquid water can tolerate. The researchers note that ionic liquid can be a hospitable environment for some biomolecules, such as certain proteins that can remain stable in the fluid.

The scientists propose that, even on planets that are too warm or that have atmospheres are too low-pressure to support liquid water, there could still be pockets of ionic liquid. And where there is liquid, there may be potential for life, though likely not anything that resembles Earth’s water-based beings.

“We consider water to be required for life because that is what’s needed for Earth life. But if we look at a more general definition, we see that what we need is a liquid in which metabolism for life can take place,” says Rachana Agrawal, who led the study as a postdoc in MIT’s Department of Earth, Atmospheric and Planetary Sciences. “Now if we include ionic liquid as a possibility, this can dramatically increase the habitability zone for all rocky worlds.”

The study’s MIT co-authors are Sara Seager, the Class of 1941 Professor of Planetary Sciences in the Department of Earth, Atmospheric and Planetary Sciences and a professor in the departments of Physics and of Aeronautics and Astronautics, along with Iaroslav Iakubivskyi, Weston Buchanan, Ana Glidden, and Jingcheng Huang. Co-authors also include Maxwell Seager of Worcester Polytechnic Institute, William Bains of Cardiff University, and Janusz Petkowski of Wroclaw University of Science and Technology, in Poland.

A liquid leap

The team’s work with ionic liquid grew out of an effort to search for signs of life on Venus, where clouds of sulfuric acid envelope the planet in a noxious haze. Despite its toxicity, Venus’ clouds may contain signs of life — a notion that scientists plan to test with upcoming missions to the planet’s atmosphere.

Agrawal and Seager, who is leading the Morning Star Missions to Venus, were investigating ways to collect and evaporate sulfuric acid. If a mission collects samples from Venus’ clouds, sulfuric acid would have to be evaporated away in order to reveal any residual organic compounds that could then be analyzed for signs of life.

The researchers were using their custom, low-pressure system designed to evaporate away excess sulfuric acid, to test evaporation of a solution of the acid and an organic compound, glycine. They found that in every case, while most of the liquid sulfuric acid evaporated, a stubborn layer of liquid always remained. They soon realized that sulfuric acid was chemically reacting with glycine, resulting in an exchange of hydrogen atoms from the acid to the organic compound. The result was a fluid mixture of salts, or ions, known as an ionic liquid, that persists as a liquid across a wide range of temperatures and pressures.

This accidental finding kickstarted an idea: Could ionic liquid form on planets that are too warm and host atmospheres too thin for water to exist?

“From there, we took the leap of imagination of what this could mean,” Agrawal says. “Sulfuric acid is found on Earth from volcanoes, and organic compounds have been found on asteroids and other planetary bodies. So, this led us to wonder if ionic liquids could potentially form and exist naturally on exoplanets.”

Rocky oases

On Earth, ionic liquids are mainly synthesized for industrial purposes. They do not occur naturally, except for in one specific case, in which the liquid is generated from the mixing of venoms produced by two rival species of ants.

The team set out to investigate what conditions ionic liquid could be naturally produced in, and over what range of temperatures and pressures. In the lab, they mixed sulfuric acid with various nitrogen-containing organic compounds. In previous work, Seager’s team had found that the compounds, some of which can be considered ingredients associated with life, are surprisingly stable in sulfuric acid.

“In high school, you learn that an acid wants to donate a proton,” Seager says. “And oddly enough, we knew from our past work with sulfuric acid (the main component of Venus’ clouds) and nitrogen-containing compounds, that a nitrogen wants to receive a hydrogen. It’s like one person’s trash is another person’s treasure.”

The reaction could produce a bit of ionic liquid if the sulfuric acid and nitrogen-containing organics were in a one-to-one ratio — a ratio that was not a focus of the prior work. For their new study, Seager and Agrawal mixed sulfuric acid with over 30 different nitrogen-containing organic compounds, across a range of temperatures and pressures, then observed whether ionic liquid formed when they evaporated away the sulfuric acid in various vials. They also mixed the ingredients onto basalt rocks, which are known to exist on the surface of many rocky planets.

“We were just astonished that the ionic liquid forms under so many different conditions,” Seager says. “If you put the sulfuric acid and the organic on a rock, the excess sulfuric acid seeps into the rock pores, but you’re still left with a drop of ionic liquid on the rock. Whatever we tried, ionic liquid still formed.”

The team found that the reactions produced ionic liquid at temperatures up to 180 degrees Celsius and at extremely low pressures — much lower than that of the Earth’s atmosphere. Their results suggest that ionic liquid could naturally form on other planets where liquid water cannot exist, under the right conditions.

“We’re envisioning a planet warmer than Earth, that doesn’t have water, and at some point in its past or currently, it has to have had sulfuric acid, formed from volcanic outgassing,” Seager says. “This sulfuric acid has to flow over a little pocket of organics. And organic deposits are extremely common in the solar system.”

Then, she says, the resulting pockets of liquid could stay on the planet’s surface, potentially for years or millenia, where they could theoretically serve as small oases for simple forms of ionic-liquid-based life. Going forward, Seager’s team plans to investigate further, to see what biomolecules, and ingredients for life, might survive, and thrive, in ionic liquid.

“We just opened up a Pandora’s box of new research,” Seager says. “It’s been a real journey.”

This research was supported, in part, by the Sloan Foundation and the Volkswagen Foundation.

The cost of solar panels has dropped by more than 99 percent since the 1970s, enabling widespread adoption of photovoltaic systems that convert sunlight into electricity.

A new MIT study drills down on specific innovations that enabled such dramatic cost reductions, revealing that technical advances across a web of diverse research efforts and industries played a pivotal role.

The findings could help renewable energy companies make more effective R&D investment decisions and aid policymakers in identifying areas to prioritize to spur growth in manufacturing and deployment.

The researchers’ modeling approach shows that key innovations often originated outside the solar sector, including advances in semiconductor fabrication, metallurgy, glass manufacturing, oil and gas drilling, construction processes, and even legal domains.

“Our results show just how intricate the process of cost improvement is, and how much scientific and engineering advances, often at a very basic level, are at the heart of these cost reductions. A lot of knowledge was drawn from different domains and industries, and this network of knowledge is what makes these technologies improve,” says study senior author Jessika Trancik, a professor in MIT’s Institute for Data, Systems, and Society.

Trancik is joined on the paper by co-lead authors Goksin Kavlak, a former IDSS graduate student and postdoc who is now a senior energy associate at the Brattle Group; Magdalena Klemun, a former IDSS graduate student and postdoc who is now an assistant professor at Johns Hopkins University; former MIT postdoc Ajinkya Kamat; as well as Brittany Smith and Robert Margolis of the National Renewable Energy Laboratory. The research appears today in PLOS ONE.

Identifying innovations

This work builds on mathematical models that the researchers previously developed that tease out the effects of engineering technologies on the cost of photovoltaic (PV) modules and systems.

In this study, the researchers aimed to dig even deeper into the scientific advances that drove those cost declines.

They combined their quantitative cost model with a detailed, qualitative analysis of innovations that affected the costs of PV system materials, manufacturing steps, and deployment processes.

“Our quantitative cost model guided the qualitative analysis, allowing us to look closely at innovations in areas that are hard to measure due to a lack of quantitative data,” Kavlak says.

Building on earlier work identifying key cost drivers — such as the number of solar cells per module, wiring efficiency, and silicon wafer area — the researchers conducted a structured scan of the literature for innovations likely to affect these drivers. Next, they grouped these innovations to identify patterns, revealing clusters that reduced costs by improving materials or prefabricating components to streamline manufacturing and installation. Finally, the team tracked industry origins and timing for each innovation, and consulted domain experts to zero in on the most significant innovations.

All told, they identified 81 unique innovations that affected PV system costs since 1970, from improvements in antireflective coated glass to the implementation of fully online permitting interfaces.

“With innovations, you can always go to a deeper level, down to things like raw materials processing techniques, so it was challenging to know when to stop. Having that quantitative model to ground our qualitative analysis really helped,” Trancik says.

They chose to separate PV module costs from so-called balance-of-system (BOS) costs, which cover things like mounting systems, inverters, and wiring.

PV modules, which are wired together to form solar panels, are mass-produced and can be exported, while many BOS components are designed, built, and sold at the local level.

“By examining innovations both at the BOS level and within the modules, we identify the different types of innovations that have emerged in these two parts of PV technology,” Kavlak says.

BOS costs depend more on soft technologies, nonphysical elements such as permitting procedures, which have contributed significantly less to PV’s past cost improvement compared to hardware innovations.

“Often, it comes down to delays. Time is money, and if you have delays on construction sites and unpredictable processes, that affects these balance-of-system costs,” Trancik says.

Innovations such as automated permitting software, which flags code-compliant systems for fast-track approval, show promise. Though not yet quantified in this study, the team’s framework could support future analysis of their economic impact and similar innovations that streamline deployment processes.

Interconnected industries

The researchers found that innovations from the semiconductor, electronics, metallurgy, and petroleum industries played a major role in reducing both PV and BOS costs, but BOS costs were also impacted by innovations in software engineering and electric utilities.

Noninnovation factors, like efficiency gains from bulk purchasing and the accumulation of knowledge in the solar power industry, also reduced some cost variables.

In addition, while most PV panel innovations originated in research organizations or industry, many BOS innovations were developed by city governments, U.S. states, or professional associations.

“I knew there was a lot going on with this technology, but the diversity of all these fields and how closely linked they are, and the fact that we can clearly see that network through this analysis, was interesting,” Trancik says.

“PV was very well-positioned to absorb innovations from other industries — thanks to the right timing, physical compatibility, and supportive policies to adapt innovations for PV applications,” Klemun adds.

The analysis also reveals the role greater computing power could play in reducing BOS costs through advances like automated engineering review systems and remote site assessment software.

“In terms of knowledge spillovers, what we've seen so far in PV may really just be the beginning,” Klemun says, pointing to the expanding role of robotics and AI-driven digital tools in driving future cost reductions and quality improvements.

In addition to their qualitative analysis, the researchers demonstrated how this methodology could be used to estimate the quantitative impact of a particular innovation if one has the numerical data to plug into the cost equation.

For instance, using information about material prices and manufacturing procedures, they estimate that wire sawing, a technique which was introduced in the 1980s, led to an overall PV system cost decrease of $5 per watt by reducing silicon losses and increasing throughput during fabrication.

“Through this retrospective analysis, you learn something valuable for future strategy because you can see what worked and what didn’t work, and the models can also be applied prospectively. It is also useful to know what adjacent sectors may help support improvement in a particular technology,” Trancik says.

Moving forward, the researchers plan to apply this methodology to a wide range of technologies, including other renewable energy systems. They also want to further study soft technology to identify innovations or processes that could accelerate cost reductions.

“Although the process of technological innovation may seem like a black box, we’ve shown that you can study it just like any other phenomena,” Trancik says.

This research is funded, in part, by the U.S. Department of Energy Solar Energies Technology Office.

Davi Augusto Oliveira Pinto’s career in Brazil’s foreign service took him all over the world. His work as a diplomat for more than two decades exposed him to the realities of life for all kinds of people, which informed his interest in economics and public policy. 

Oliveira Pinto is now focused on strengthening his diplomatic work through his MIT education. He completed the MITx MicroMasters program in Data, Economics, and Design of Policy (DEDP), which is jointly administered by MIT Open Learning and the Abdul Latif Jameel Poverty Action Lab (J-PAL), and then applied and was accepted to the DEDP master’s program within MIT’s Department of Economics. 

“I think governments should be able to provide data-driven, research-supported services to their constituents,” he says. “Returning to my role as a diplomat, I hope to use the tools I acquired in the DEDP program to enhance my contributions as a public servant.”

Oliveira Pinto was one of Brazil’s representatives to the World Trade Organization (WTO), helped Brazilian citizens and companies abroad, and worked to improve relationships with governments in South Africa, Argentina, Italy, Spain, and Uruguay. He observed firsthand how economic disparities could influence laws and lives. He believes in a nonpartisan approach to public service, producing and sharing policy based on peer-reviewed data and research that can help as many people as possible. 

“We need public policy informed by evidence and science, rather than by politics and ideology,” he says. “My experience at MIT reinforced my conviction that diplomacy should be used to gather people from different backgrounds and develop joint solutions to our collective challenges.”

As someone responsible for dealing with international trade issues and who understands the potential negative, far-reaching impacts of poorly researched and instituted policies, Oliveira Pinto saw MIT and its world-class economics programs as potentially world-altering tools to help him advance his work. 

Advocacy and economics

Growing up in Minas Gerais, Brazil, Oliveira Pinto learned about the country’s past of economic cycles driven by exporting commodities like minerals and coffee. He also witnessed what he described as Brazil’s “eternal state of development,” one in which broad swaths of the population suffered, and very soon became aware of the impact that issues like inflation and unemployment had on the country. 

“I thought studying economics could help solve issues I observed when growing up,” he says.

Oliveira Pinto earned an undergraduate degree in economics from Universidade Federal de Minas Gerais and a master’s degree in public policy from Escola Nacional de Administração Pública.

Oliveira Pinto’s personal experiences and his commitment to understanding and improving the lives of his fellow Brazilians led him to enroll in the Instituto Rio Branco, Brazil’s diplomatic academy, where he was trained in a variety of disciplines. “I was drawn to investigate inequality between countries, which led to my diplomatic career,” he says. “I worked to help Brazilian migrants abroad, promoted Brazilian companies’ exports, represented Brazil at the WTO, and helped pandemic-era assistance efforts for people in Brazil’s poor border towns.”

During the pandemic, Oliveira Pinto found himself drawn to the DEDP MicroMasters program. He was able to review foundational economics concepts, improve his ability to synthesize and interpret data, and refine his analytical skills. “My favorite course, Data Analysis for Social Scientists, reinforced the critical importance of interpreting data correctly in a world where information is increasingly abundant,” he recalls. 

The online program also offered an opportunity for him to apply to study in person. Now at MIT, Oliveira Pinto is finishing his degree with a capstone project focused on how J-PAL works with governments to support the scaling of evidence-informed policies.  

J-PAL’s research center and network have built long-term partnerships with government agencies around the world to generate evidence from randomized evaluations and incorporate the findings into policy decisions. They work closely with policymakers to inform anti-poverty programs to improve their effectiveness, an area of particular interest to the Brazilian diplomat. 

“I’m trying to understand how J-PAL’s partnerships in these places are working, any lessons we can learn from successes, challenges faced, and how we can most effectively scale the successful programs,” he says.

Inside and beyond MIT

Oliveira Pinto was welcomed into a thriving, diverse community in Cambridge, a journey that was both edifying and challenging. “My family and I found a home,” he notes, observing that many Brazilians live in the area, “and it’s sobering to see so many people from my country working hard to build their lives in the U.S.”

Oliveira Pinto says working closely with members of the MIT community was one of the DEDP master’s program’s big draws. “The ability to forge connections with students and faculty while learning from Nobel laureates and accomplished researchers and practitioners is amazing,” he says. Collaborating with people from a variety of professional, experiential, and backgrounds, he notes, was especially satisfying. 

Oliveira Pinto offered special praise for MIT’s support for his family, describing it as “particularly rewarding.” “MIT offers so many different activities for families,” he says. “My wife and three daughters benefited from the support the Institute provides.” While taking advantage of his time in the States to visit Canada and Washington, D.C., they also made the most of their time in Cambridge. The family enjoyed sailing, swimming, yoga, sports, pottery, lectures, and more while Davi pursued his studies. “The facilities are awesome,” he continues.

Assessing and quantifying impact

Oliveira Pinto’s investigations have yielded some fascinating findings. “Data can be misused,” he notes. “I learned how easily data can tell all kinds of stories, so it’s important to be careful and rigorous when assessing different claims.” He recalls how, during an econometrics class, he learned about parties on opposite sides of a health insurance divide pursuing radically different ends using the same data, each side promoting different views. 

Oliveira Pinto believes his studies have improved his abilities as a diplomat, one of the reasons he’s excited about his eventual return to the public service. “I’ll return to government service armed with the skills the DEDP program and the research conducted during my capstone project have provided,” he says. “My job as a diplomat is to seek opportunities to connect with different people, investigate carefully, and find common ground,” work for which his DEDP MicroMasters and master’s studies have helped prepare him.

Completing his capstone, Oliveira Pinto hopes to draw lessons from J-PAL’s work with governments to improve constituents' quality of life. He’s helping generate case studies that may foster future collaborations between researchers and the public sector. 

“Work like this can be a good opportunity for governments interested in a research-supported, data-driven approach to policymaking,” he says. 

There are 63 million people caring for family members with an illness or disability in the U.S. That translates to one in four adults devoting their time to helping loved ones with things like transportation, meals, prescriptions, and medical appointments.

Caregiving exacts a huge toll on the people responsible, and ianacare is seeking to lessen the burden. The company, founded by Steven Lee ’97, MEng ’98 and Jessica Kim, has built a platform that helps caregivers navigate available tools and local resources, build a network of friends and family to assist with everyday tasks, and coordinate meals, rides, and care shifts.

The name ianacare is short for “I am not alone care.” The company’s mission is to equip and empower the millions of people who perform a difficult and underappreciated role in our society.

“Family caregivers are the invisible backbone of the health care system,” Lee says. “Without them, the health care system would literally collapse, but they are still largely unrecognized. Ianacare acts as the front door for family caregivers. These caregivers are often thrust into this role untrained and unguided. But the moment they start, they have to become experts. Ianacare fills that gap.”

The company has partnered with employers and health care providers to serve more than 50,000 caregivers to date. And thanks to a partnerships with organizations like Elevance Health, the American Association of Retired Persons (AARP), and Medicare providers, its coordination and support tools are available to family caregivers across the country.

“Ultimately we want to make the biggest impact possible,” Lee says. “From a business standpoint, the 50,000 caregivers we’ve served is a huge number. But from the overall universe of caregivers that could use our help, it’s relatively small. We’re on a mission to help all 63 million caregivers.”

From ad tech to ianacare

As an electrical engineering and computer science student at MIT in the 1990s, Lee conducted research on early speech-recognition technology as part of the Spoken Language Systems group in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL).

Following graduation, Lee started a company with Waikit Lau ’97 that optimized video advertising placement within streams. The company has gone through several mergers and acquisitions, but is now part of the public company Magnite, which places the ads on platforms like Netflix, Hulu, and Disney+.

Lee left the company in 2016 and began advising startups through programs including MIT’s Venture Mentoring Service as he looked to work on something he would find more meaningful.

“Over the years, the MIT network has been invaluable for connecting with customers, recruiting top talent, and engaging investors,” Lee says. “So much innovation flows out of MIT, and I’ve loved giving back, especially working alongside [VMS Venture Mentor] Paul Bosco ’95 and the rest of the VMS team. It’s deeply rewarding to share the best practices I’ve learned with the next generation of innovators.”

In 2017, Lee met Kim, who was caregiving for her mother with pancreatic cancer. Hearing about her experience brought him back to his own family’s challenges caring for his grandfather with Parkinson’s disease when Lee was a child.

“We realized the gaps that existed in caregiving support three decades ago still exist,” Lee says. “Nothing has changed.”

Officially launched in 2018, ianacare may seem far-removed from speech recognition or ad technologies, but Lee sees the work as an extension of his previous experiences.

“In my mind, AI got its start in speech recognition, and the intelligence we use to surface recommendations and create care plans for family caregivers uses a lot of the same statistical modeling techniques I used in speech recognition and ad placement,” Lee says. “It all goes back to the foundation I got at MIT.”

The founders first launched a free solution that allowed caregivers to connect with friends and family members to coordinate caregiving tasks.

“In our app, you can coordinate with anyone who’s interested in helping,” Lee says. “When you share a struggle with a friend or co-worker, they always say, ‘How can I help?’ But caregivers rarely go back to them and actually ask. In our platform, you can add those people to your informal care team and ask the team for help with something instead of having to text someone directly, which you’re less likely to do.”

Next, the founders built an enterprise solution so businesses could help employee caregivers, adding features like resource directories and ways to find and select various caregiving tools.

“An immense amount of local resources are available, but nobody knows about them,” Lee says. “For instance, every county in the country has an Area Agency on Aging, but these agencies aren’t marketing experts, and caregivers don’t know where to get guidance.”

Last year, ianacare began working with AARP and health care providers participating in the nationwide GUIDE model (for “Guiding an Improved Dementia Experience”) to improve the quality of life for dementia patients and their caregivers. Through the voluntary program, participants can use ianacare’s platform to coordinate care, access educational resources, and access free respite care up to $2,500 each year.

Lee says the CMS partnership gives ianacare a pathway to reach millions of people caring for dementia patients across the country.

“This is already a crisis, and it will get worse because we have an aging population and a capacity-constraint in our health care system,” Lee says. “The population above 65 is set to double between 2000 and 2040. We aren’t going to have three times the hospitals or three times the doctors or nurse practitioners. So, we can either make clinicians more efficient or move more health care into the home. That’s why we have empower family caregivers.”

Aging with dignity

Lee recalls one family who used ianacare after their son was born with a severe disease. The child only lived eight months, but for those eight months, the parents had meals delivered to them in the hospital by friends and family.

“It was not something they had to worry about the entire time their son was alive,” Lee says. “It’s been rewarding to help these people in so much need.”

Other ianacare users say the platform has helped them keep their parents out of the hospital and lessen their depression and anxiety around caregiving.

“Nobody wants to die in a hospital, so we’ve worked hard to honor the wishes of loved ones who want to age in the home,” Lee says. “We have a lot of examples of folks who, if our support was not there, their loved one would have had to enter a nursing home or institution. Ianacare is there to ensure the home is safe and that the caregiver can manage the care burden. It’s a win-win for everybody because it’s also less costly for the health care system.”

Each year, faculty and researchers across the MIT School of Engineering are recognized with prestigious awards for their contributions to research, technology, society, and education. To celebrate these achievements, the school periodically highlights select honors received by members of its departments, labs, and centers. The following individuals were recognized in spring 2025:

Markus Buehler, the Jerry McAfee (1940) Professor in Engineering in the Department of Civil and Environmental Engineering, received the Washington Award. The award honors engineers whose professional attainments have preeminently advanced the welfare of humankind.

Sili Deng, an associate professor in the Department of Mechanical Engineering, received the 2025 Hiroshi Tsuji Early Career Researcher Award. The award recognizes excellence in fundamental or applied combustion science research. Deng was honored for her work on energy conversion and storage, including combustion fundamentals, data-driven modeling of reacting flows, carbon-neutral energetic materials, and flame synthesis of materials for catalysis and energy storage.

Jonathan How, the Richard Cockburn Maclaurin Professor in Aeronautics and Astronautics, received the IEEE Transactions on Robotics King-Sun Fu Memorial Best Paper Award. The award recognizes the best paper published annually in the IEEE Transactions on Robotics for technical merit, originality, potential impact, clarity, and practical significance.

Richard Linares, the Rockwell International Career Development Professor in the Department of Aeronautics and Astronautics, received the 2024 American Astronautical Society Emerging Astrodynamicist Award. The award honors junior researchers making significant contributions to the field of astrodynamics.

Youssef Marzouk, the Breene M. Kerr (1951) Professor in the Department of Aeronautics and Astronautics, was named a fellow of the Society for Industrial and Applied Mathematics. He was honored for influential contributions to multiple aspects of uncertainty quantification, particularly Bayesian computation and measure transport.

Dava Newman, the director of the MIT Media Lab and the Apollo Program Professor in the Department of Aeronautics and Astronautics, received the Carolyn “Bo” Aldigé Visionary Award. The award was presented in recognition of the MIT Media Lab's women’s health program, WHx, for groundbreaking research in advancing women’s health.

Martin Rinard, a professor in the Department of Electrical Engineering and Computer Science, received the 2025 SIGSOFT Outstanding Research Award. The award recognizes his fundamental contributions in pioneering the new fields of program repair and approximate computing.

Franz-Josef Ulm, the Class of 1922 Professor in the Department of Civil and Environmental Engineering, was named an ASCE Distinguished Member. He was recognized for contributions to the nano- and micromechanics of heterogeneous materials, including cement, concrete, rock, and bone, with applications in sustainable infrastructure, underground energy harvesting, and human health.

Enter the basement in one of MIT’s iconic buildings and you’ll find students hammering on anvils and forging red-hot metal into blades. This hands-on lesson in metallurgy is captured in the documentary “That Creative Spark,” which won an Emmy Award for the Education/Schools category at the 48th annual Boston/New England Emmy Awards Ceremony held in Boston in June.

“It’s wonderful to be recognized for the work that we do,” says Clayton Hainsworth, director of MIT Video Productions at MIT Open Learning. “We’re lucky to have incredible people who have decided to bring their outstanding talents here in order to tell MIT’s stories.”

The National Academy of Television Arts and Sciences Boston/New England Chapter recently honored Hainsworth, the documentary’s executive producer; Joe McMaster, director/producer; and Wesley Richardson, cinematographer.

“That Creative Spark” spotlights a series of 2024 Independent Activities Period (IAP) classes about bladesmithing, guest-taught by Bob Kramer, a world-renowned maker of hand-forged knives. In just one week, students learned how to grind, forge, and temper blocks of steel into knives sharp enough to slice through a sheet of paper without resistance.

“It’s an incredibly physical task of making something out of metal,” says McMaster, senior producer for MIT Video Productions. He says this tangible example of hands-on learning “epitomized the MIT motto of ‘mens et manus’ [‘mind and hand’].”

The IAP Bladesmithing with Bob Kramer course allowed students to see concepts and techniques like conductivity and pattern welding in action. Abhi Ratna Sharda, a PhD student at the Department of Materials Science and Engineering (DMSE), still recalls the feeling of metal changing as he worked on it.

“Those are things that you can be informed about through readings and textbooks, but the actual experience of doing them leaves an intuition you’re not quick to forget,” Sharda says.

Filming in the forge — the Merton C. Flemings Materials Processing Laboratory — is not an experience the MIT Video Productions team will be quick to forget, either. Richardson, field production videographer at MIT Video Productions, held the camera just six feet away from red-hot blades being dipped into tubs of oil, creating minor fireballs and plumes of smoke.

“It’s intriguing to see the dexterity that the students have around working with their hands with very dangerous objects in close proximity to each other,” says Richardson. “Students were able to get down to these really precise knives at the end of the class.”

Some people may be surprised to learn that MIT has a working forge, but metalworking is a long tradition at the Institute. In the documentary, Yet-Ming Chiang, Kyocera Professor of Ceramics at DMSE, points out a clue hidden in plain sight: “If you look at the MIT logo, there’s a blacksmith, and ‘mens et manus’ — ‘mind and hand,’” says Chiang, referring to the Institute’s official seal, adopted in 1894. “So the teaching and the practice of working with metals has been an important part of our department for a long time.”

Chiang invited Kramer to be a guest instructor and lecturer for two reasons: Kramer is an industry expert, and he achieved success through hands-on learning — an integral part of an MIT education. After dropping out of college and joining the circus, Kramer later gained practical experience in service-industry kitchens and eventually became one of just 120 Master Bladesmiths in the United States today.

“This nontraditional journey of Bob’s inspires students to think about projects and problems in different ways,” Hainsworth says.

Sharda, for example, is applying the pattern welding process he learned from Kramer in both his PhD program and his recreational jewelry making. The effect creates striking visuals — from starbursts to swirls looking like agate geodes, and more — that extend all the way through the steel, not just the surface of the blade.

“A lot of my research has to do with bonding metals and bonding dissimilar metals, which is the foundation for pattern welding,” Sharda says, adding how this technique has many potential industrial applications. He compares it to the mokume-gane technique used with precious metals, a practice he encountered while researching solid-state welding methods.

“Seeing that executed in a space where it’s very difficult to achieve that level of precision — it inspired me to polish all the tightest nooks and crannies of the pieces I make, and make sure everything is as flawless as possible,” Sharda adds.

In the documentary, Kramer reflects on his month of teaching experience: “When you give someone the opportunity and guide them to actually make something with their hands, there’s very few things that are as satisfying as that.”

In addition to highlighting MIT’s hands-on approach to teaching, “That Creative Spark” showcases the depth of its unique learning experiences.

“There are many sides to MIT in terms of what the students are actually given access to and able to do,” says Richardson. “There is no one face of MIT, because they're highly gifted, highly talented, and often those talents and gifts extend beyond their courses of study.”

That message resonates with Chiang, who says the class underscores the importance of hands-on, experimental research in higher education.

“What I think is a real benefit in experimental research is the physical understanding of how objects and forces relate to each other,” he says. “This kind of class helps students — especially students who’ve never had that experience, never had a job that requires real hands-on work — gain an understanding of those relationships.”

Hainsworth says it’s wonderful to collaborate with his team to tell stories about the spirit and generosity of Institute faculty, guest speakers, and students. The documentary was made possible, in part, thanks to the generous support of A. Neil Pappalardo ’64 and Jane Pappalardo.

“It really is a joy to come in every day and collaborate with people who care deeply about the work they do,” Hainsworth says. “And to be recognized with an Emmy, that is very rewarding.”

Jason Sparapani contributed to this story.