To what extent can an artificial system be rational?

A new MIT course, 6.S044/24.S00 (AI and Rationality), doesn’t seek to answer this question. Instead, it challenges students to explore this and other philosophical problems through the lens of AI research. For the next generation of scholars, concepts of rationality and agency could prove integral in AI decision-making, especially when influenced by how humans understand their own cognitive limits and their constrained, subjective views of what is or isn’t rational.

This inquiry is rooted in a deep relationship between computer science and philosophy, which have long collaborated in formalizing what it is to form rational beliefs, learn from experience, and make rational decisions in pursuit of one's goals.

“You’d imagine computer science and philosophy are pretty far apart, but they’ve always intersected. The technical parts of philosophy really overlap with AI, especially early AI,” says course instructor Leslie Kaelbling, the Panasonic Professor of Computer Science and Engineering at MIT, calling to mind Alan Turing, who was both a computer scientist and a philosopher. Kaelbling herself holds an undergraduate degree in philosophy from Stanford University, noting that computer science wasn’t available as a major at the time.

Brian Hedden, a professor in the Department of Linguistics and Philosophy, holding an MIT Schwarzman College of Computing shared position with the Department of Electrical Engineering and Computer Science (EECS), who teaches the class with Kaelbling, notes that the two disciplines are more aligned than people might imagine, adding that the “differences are in emphasis and perspective.”

Tools for further theoretical thinking

Offered for the first time in fall 2025, Kaelbling and Hedden created AI and Rationality as part of the Common Ground for Computing Education, a cross-cutting initiative of the MIT Schwarzman College of Computing that brings multiple departments together to develop and teach new courses and launch new programs that blend computing with other disciplines.

With over two dozen students registered, AI and Rationality is one of two Common Ground classes with a foundation in philosophy, the other being 6.C40/24.C40 (Ethics of Computing).

While Ethics of Computing explores concerns about the societal impacts of rapidly advancing technology, AI and Rationality examines the disputed definition of rationality by considering several components: the nature of rational agency, the concept of a fully autonomous and intelligent agent, and the ascription of beliefs and desires onto these systems.

Because AI is extremely broad in its implementation and each use case raises different issues, Kaelbling and Hedden brainstormed topics that could provide fruitful discussion and engagement between the two perspectives of computer science and philosophy.

“It's important when I work with students studying machine learning or robotics that they step back a bit and examine the assumptions they’re making,” Kaelbling says. “Thinking about things from a philosophical perspective helps people back up and understand better how to situate their work in actual context.”

Both instructors stress that this isn’t a course that provides concrete answers to questions on what it means to engineer a rational agent.

Hedden says, “I see the course as building their foundations. We’re not giving them a body of doctrine to learn and memorize and then apply. We’re equipping them with tools to think about things in a critical way as they go out into their chosen careers, whether they’re in research or industry or government.”

The rapid progress of AI also presents a new set of challenges in academia. Predicting what students may need to know five years from now is something Kaelbling sees as an impossible task. “What we need to do is give them the tools at a higher level — the habits of mind, the ways of thinking — that will help them approach the stuff that we really can’t anticipate right now,” she says.

Blending disciplines and questioning assumptions

So far, the class has drawn students from a wide range of disciplines — from those firmly grounded in computing to others interested in exploring how AI intersects with their own fields of study.

Throughout the semester’s reading and discussions, students grappled with different definitions of rationality and how they pushed back against assumptions in their fields.

On what surprised her about the course, Amanda Paredes Rioboo, a senior in EECS, says, “We’re kind of taught that math and logic are this golden standard or truth. This class showed us a variety of examples that humans act inconsistently with these mathematical and logical frameworks. We opened up this whole can of worms as to whether, is it humans that are irrational? Is it the machine learning systems that we designed that are irrational? Is it math and logic itself?”

Junior Okoroafor, a PhD student in the Department of Brain and Cognitive Sciences, was appreciative of the class’s challenges and the ways in which the definition of a rational agent could change depending on the discipline. “Representing what each field means by rationality in a formal framework, makes it clear exactly which assumptions are to be shared, and which were different, across fields.”

The co-teaching, collaborative structure of the course, as with all Common Ground endeavors, gave students and the instructors opportunities to hear different perspectives in real-time.

For Paredes Rioboo, this is her third Common Ground course. She says, “I really like the interdisciplinary aspect. They’ve always felt like a nice mix of theoretical and applied from the fact that they need to cut across fields.”

According to Okoroafor, Kaelbling and Hedden demonstrated an obvious synergy between fields, saying that it felt as if they were engaging and learning along with the class. How computer science and philosophy can be used to inform each other allowed him to understand their commonality and invaluable perspectives on intersecting issues.

He adds, “philosophy also has a way of surprising you.”

New treatments based on biological molecules like RNA give scientists unprecedented control over how cells function. But delivering those drugs to the right tissues remains one of the biggest obstacles to turning these promising yet fragile molecules into powerful new treatments.

Now Gensaic, founded by Lavi Erisson MBA ’19; Uyanga Tsedev SM ’15, PhD ’21; and Jonathan Hsu PhD ’22, is building an artificial intelligence-powered discovery engine to develop protein shuttles that can deliver therapeutic molecules like RNA to specific tissues and cells in the body. The company is using its platform to create advanced treatments for metabolic diseases and other conditions. It is also developing treatments in partnership with Novo Nordisk and exploring additional collaborations to amplify the speed and scale of its impact.

The founders believe their delivery technology — combined with advanced therapies that precisely control gene expression, like RNA interference (RNAi) and small activating RNA (saRNA) — will enable new ways of improving health and treating disease.

“I think the therapeutic space in general is going to explode with the possibilities our approach unlocks,” Erisson says. “RNA has become a clinical-grade commodity that we know is safe. It is easy to synthesize, and it has unparalleled specificity and reversibility. By taking that and combining it with our targeting and delivery, we can change the therapeutic landscape.”

Drinking from the firehose

Erisson worked on drug development at the large pharmaceutical company Teva Pharmaceuticals before coming to MIT for his Sloan Fellows MBA in 2018.

“I came to MIT in large part because I was looking to stretch the boundaries of how I apply critical thinking,” Erisson says. “At that point in my career, I had taken about 10 drug programs into clinical development, with products on the market now. But what I didn’t have were the intellectual and quantitative tools for interrogating finance strategy and other disciplines that aren’t purely scientific. I knew I’d be drinking from the firehose coming to MIT.”

Erisson met Hsu and Tsedev, then PhD students at MIT, in a class taught by professors Harvey Lodish and Andrew Lo. The group started holding weekly meetings to discuss their research and the prospect of starting a business.

After Erisson completed his MBA program in 2019, he became chief medical and business officer at the MIT spinout Iterative Health, a company using AI to improve screening for colorectal cancer and inflammatory bowel disease that has raised over $200 million to date. There, Erisson ran a 1,400-patient study and led the development and clearance of the company’s software product.

During that time, the eventual founders continued to meet at Erisson’s house to discuss promising research avenues, including Tsedev’s work in the lab of Angela Belcher, MIT’s James Mason Crafts Professor of Biological Engineering. Tsedev’s research involved using bacteriophages, which are fast-replicating protein particles, to deliver treatments into hard-to-drug places like the brain.

As Hsu and Tsedev neared completion of their PhDs, the team decided to commercialize the technology, founding Gensaic at the end of 2021. Gensaic’s approach uses a method called unbiased directed evolution to find the best protein scaffolding to reach target tissues in the body.

“Directed evolution means having a lot of different species of proteins competing together for a certain function,” Erisson says. “The proteins are competing for the ability to reach the right cell, and we are then able to look at the genetic code of the protein that has ‘won’ that competition. When we do that process repeatedly, we find extremely adaptable proteins that can achieve the function we’re looking for.”

Initially, the founders focused on developing protein scaffolds to deliver gene therapies. Gensaic has since pivoted to focus on delivering molecules like siRNA and RNAi, which have been hard to deliver outside of the liver.

Today Gensaic has screened more than 500 billion different proteins using a process called phage display and directed evolution. It calls its platform FORGE, for Functional Optimization by Recursive Genetic Evolution.

Erisson says Gensaic’s delivery vehicles can also carry multiple RNA molecules into cells at the same time, giving doctors a novel and powerful set of tools to treat and prevent diseases.

“Today FORGE is built into the idea of multifunctional medicines,” Erisson says. “We are moving into a future where we can extract multiple therapeutic mechanisms from a single molecule. We can combine proteins with multiple tissue selectivity and multiple molecules of siRNA or other therapeutic modalities, and affect complex disease system biology with a single molecule.”

A “universe of opportunity”

The founders believe their approach will enable new ways of improving health by delivering advanced therapies directly to new places in the body. Precise delivery of drugs to anywhere in the body could not only unlock new therapeutic targets but also boost the effectiveness of existing treatments and reduce side effects.

“We’ve found we can get to the brain, and we can get to specific tissues like skeletal and adipose tissue,” Erisson says. “We’re the only company, to my knowledge, that has a protein-based delivery mechanism to get to adipose tissue.”

Delivering drugs into fat and muscle cells could be used to help people lose weight, retain muscle, and prevent conditions like fatty liver disease or osteoporosis.

Erisson says combining RNA therapeutics is another differentiator for Gensaic.

“The idea of multiplexed medicines is just emerging,” Erisson says. “There are no clinically approved drugs using dual-targeted siRNAs, especially ones that have multi-tissue targeting. We are focused on metabolic indications that have two targets at the same time and can take on unique tissues or combinations of tissues.”

Gensaic’s collaboration with Novo Nordisk, announced last year, targets cardiometabolic diseases and includes up to $354 million in upfront and milestone payments per disease target.

“We already know we can deliver multiple types of payloads, and Novo Nordisk is not limited to siRNA, so we can go after diseases in ways that aren’t available to other companies,” Erisson says. “We are too small to try to swallow this universe of opportunity on our own, but the potential of this platform is incredibly large. Patients deserve safer medicines and better outcomes than what are available now.”

The modern world runs on chemicals and fuels that require a huge amount of energy to produce: Industrial chemical separation accounts for 10 to 15 percent of the world’s total energy consumption. That’s because most separations today rely on heat to boil off unwanted materials and isolate compounds.

The MIT spinout Osmoses is making industrial chemical separations more efficient by reducing the need for all that heat. The company, founded by former MIT postdoc Francesco Maria Benedetti; Katherine Mizrahi Rodriguez ’17, PhD ’22; Professor Zachary Smith; and Holden Lai, has developed a polymer technology capable of filtering gases with unprecedented selectivity.

Gases — consisting of some of the smallest molecules in the world — have historically been the hardest to separate. Osmoses says its membranes enable industrial customers to increase production, use less energy, and operate in a smaller footprint than is possible using conventional heat-based separation processes.

Osmoses has already begun working with partners to demonstrate its technology’s performance, including its ability to upgrade biogas, which involves separating CO2 and methane. The company also has projects in the works to recover hydrogen from large chemical facilities and, in a partnership with the U.S. Department of Energy, to pull helium from underground hydrogen wells.

“Chemical separations really matter, and they are a bottleneck to innovation and progress in an industry where innovation is challenging, yet an existential need,” Benedetti says. “We want to make it easier for our customers to reach their revenue targets, their decarbonization goals, and expand their markets to move the industry forward.”

Better separations

Benedetti joined Smith’s lab in MIT’s Department of Chemical Engineering in 2017. He was joined by Mizrahi Rodriguez the following year, and the pair spent the next few years conducting fundamental research into membrane materials for gas separations, collaborating with chemists at MIT and beyond, including Lai as he conducted his PhD at Stanford University with Professor Yan Xia.

“I was fascinated by the projects [Smith] was thinking about,” Benedetti says. “It was high-risk, high-reward, and that’s something I love. I had the opportunity to work with talented chemists, and they were synthesizing amazing polymers. The idea was for us chemical engineers at MIT to study those polymers, support chemists in taking next steps, and find an application in the separations world.”

The researchers slowly iterated on the membranes, gradually achieving better performance until, in 2020, a group including Lai, Benedetti, Xia, and Smith broke records for gas separation selectivity with a class of three-dimensional polymers whose structural backbone could be tuned to optimize performance. They filed patents with Stanford and MIT over the next two years, publishing their results in the journal Science in 2022.

“We were facing a decision of what to do with this incredible innovation,” Benedetti recalls. “By then, we’d published a lot of papers where, as the introduction, we described the huge energy footprint of thermal gas separations and the potential of membranes to solve that. We thought rather than wait for somebody to pick up the paper and do something with it, we wanted to lead the effort to commercialize the technology.”

Benedetti joined forces with Mizrahi Rodriguez, Lai, and industrial advisor Xinjin Zhao PhD ’92 to go through the National Science Foundation’s I-Corps Program, which challenges researchers to speak to potential customers in industry. The researchers interviewed more than 100 people, which confirmed for them the huge impact their technology could have.

Benedetti received grants from the MIT Deshpande Center for Technological Innovation, MIT Sandbox, and was a fellow with the MIT Energy Initiative. Osmoses also won the MIT $100K Entrepreneurship Competition in 2021, the same year they founded the company.

“I spent a lot of time talking to stakeholders of companies, and it was a window into the challenges the industry is facing,” Benedetti says. “It helped me determine this was a problem they were facing, and showed me the problem was massive. We realized if we could solve the problem, we could change the world.”

Today, Benedetti says more than 90 percent of energy in the chemicals industry is used to thermally separate gases. One study in Nature found that replacing thermal distillation could reduce annual U.S. energy costs by $4 billion and save 100 million tons of carbon dioxide emissions.

Made up of a class of molecules with tunable structures called hydrocarbon ladder polymers, Osmoses’ membranes are capable of filtering gas molecules with high levels of selectivity, at scale. The technology reduces the size of separation systems, making it easier to add to existing spaces and lowering upfront costs for customers.

“This technology is a paradigm shift with respect to how most separations are happening in industry today,” Benedetti says. “It doesn’t require any thermal processes, which is the reason why the chemical and petrochemical industries have such high energy consumption. There are huge inefficiencies in how separations are done today because of the traditional systems used.”

From the lab to the world

In the lab, the founders were making single grams of their membrane polymers for experiments. Since then, they’ve scaled up production dramatically, reducing the cost of the material with an eye toward producing potentially hundreds of kilograms in the future.

The company is currently working toward its first pilot project upgrading biogas at a landfill operated by a large utility in North America. It is also planning a pilot at a dairy farm in North America. Mizrahi Rodriguez says waste gas from landfills and agricultural make up over 80 percent of the biogas upgrading market overall and represent a promising alternative source of renewable methane for customers.

“In the near term, our goal is to validate this technology at scale,” Benedetti says, noting Osmoses aims to scale up its pilot projects. “It has been a big accomplishment to secure funded pilots in all of the verticals that will serve as a springboard for our next commercial phase.”

Osmoses’ other two pilot projects focus on recovering valuable gas, including helium with the Department of Energy.

“Helium is a scarce resource that we need for a variety of applications, like MRIs, and our membranes’ high performance can be used to extract small amounts of it from underground wells,” Mizrahi Rodriguez explains. “Helium is very important in the semiconductor industry to build chips and graphical processing units that are powering the AI revolution. It’s a strategic resource that the U.S. has a growing interest to produce domestically.”

Benedetti says further down the line, Osmoses’ technology could be used in carbon capture, gas “sweetening” to remove acid gases from natural gas, to separate oxygen and nitrogen, to reuse refrigerants, and more.

“There will be a progressive expansion of our capabilities and markets to deliver on our mission of redefining the backbone of the chemical, petrochemical, and energy industries,” Benedetti says. “Separations should not be a bottleneck to innovation and progress anymore.”

For decades, MIT Professor Ted Gibson has taught the meaning of language to first-year graduate students in the Department of Brain and Cognitive Sciences (BCS). A new book, Gibson’s first, brings together his years of teaching and research to detail the rules of how words combine.

“Syntax: A Cognitive Approach,” released by MIT Press on Dec. 16, lays out the grammar of a language from the perspective of a cognitive scientist, outlining the components of language structure and the model of syntax that Gibson advocates: dependency grammar.

It was his research collaborator and wife, associate professor of BCS and McGovern Institute for Brain Research investigator Ev Fedorenko, who encouraged him to put pen to paper. Here, Gibson takes some time to discuss the book.

Q: Where did the process for “Syntax” begin?

A: I think it started with my teaching. Course 9.012 (Cognitive Science), which I teach with Josh Tenenbaum and Pawan Sinha, divides language into three components: sound, structure, and meaning. I work on the structure and meaning parts of language: words and how they get put together. That’s called syntax.

I’ve spent a lot of time over the last 30 years trying to understand the compositional rules of syntax, and even though there are many grammar rules in any language, I actually don’t think the form for grammar rules is that complicated. I’ve taught it in a very simple way for many years, but I’ve never written it all down in one place. My wife, Ev, is a longtime collaborator, and she suggested I write a paper. It turned into a book.

Q: How do you like to explain syntax?

A: For any sentence, for any utterance in any human language, there’s always going to be a word that serves as the head of that sentence, and every other other word will somehow depend on that headword, maybe as an immediate dependent, or further away, through some other dependent words. This is called dependency grammar; it means there’s a root word in each sentence, and dependents of that root, on down, for all the words in the sentence, form a simple tree structure. I have cognitive reasons to suggest that this model is correct, but it isn’t my model; it was first proposed in the 1950s. I adopted it because it aligns with human cognitive phenomena.

That very simple framework gives you the following observation: that longer-distance connections between words are harder to produce and understand than shorter-distance ones. This is because of limitations in human memory. The closer the words are together, the easier it is for me to produce them in a sentence, and the easier it is for you to understand them. If they’re far apart, then it’s a complicated memory problem to produce and understand them.

This gives rise to a cool observation: Languages optimize their rules in order to keep the words close together. We can have very different orders of the same elements across languages, such as the difference in word orders for English versus Japanese, where the order of the words in the English sentence “Mary eats an apple” is “Mary apple eats” in Japanese. But then the ordering rules in English and Japanese are aligned within themselves in order to minimize dependency lengths on average for the language.

Q: How does the book challenge some longstanding ideas in the field of linguistics?

A: In 1957, a book called “Syntactic Structures” by Noam Chomsky was published. It is a wonderful book that provides mathematical approaches to describe what human language is. It is very influential in the field of linguistics, and for good reason.

One of the key components of the theory that Chomsky proposed was the “transformation,” such that words and phrases can move from a deep structure to the structure that we produce. He thought it was self-evident from examples in English that transformations must be part of a human language. But then this concept of transformations eventually led him to conclude that grammar is unlearnable, that it has to be built into the human mind.  

In my view of grammar, there are no transformations. Instead, there are just two different versions of some words, or they can be underspecified for their grammar usage. The different usages may be related in meaning, and they can point to a similar meaning, but they have different dependency structures.

I think the advent of large language models suggests that language is learnable and that syntax isn’t as complicated as we used to think it was, because LLMs are successful at producing language. A large language model is almost the same as an adult speaker of a language in what it can produce. There are subtle ways in which they differ, but on the surface, they look the same in many ways, which suggests that these models do very well with learning language, even with human-like quantities of data.

I get pushback from some people who say, well, researchers can still use transformations to account for some phenomena. My reaction is: Unless you can show me that transformations are necessary, then I don’t think we need them.

Q: This book is open access. Why did you decide to publish it that way?

A: I am all for free knowledge for everyone. I am one of the editors of “Open Mind,” a journal established several years ago that is completely free and open access. I felt my book should be the same way, and MIT Press is a fantastic university press that is nonprofit and supportive of open-access publishing. It means I make less money, but it also means it can reach more people. For me, it is really about trying to get the information out there. I want more people to read it, to learn things. I think that’s how science is supposed to be.

Boston recently got its own good luck charm, “Amulet,” a 19-foot-tall tangle of organic spires installed in City Hall Plaza and embedded with the wishes, hopes, and prayers of residents from across the city.

The public artwork, by artist Rhea Vedro — also a lecturer and metals artist-in-residence in MIT’s Department of Materials Science and Engineering (DMSE) — was installed on the north side of City Hall, in a newly renovated stretch of the plaza along Congress Street, in October and dedicated with a ribbon cutting on Dec. 19.

“I’m really interested in this idea of protective objects worn on the skin by humans across cultures, across time,” said Vedro at the event in the Civic Pavilion, across the plaza from the sculpture. “And then, how do you take those ideas off the body and turn them into a blown-up version — a stand-in for the body?”

Vedro started exploring that question in 2021, when she was awarded a Boston Triennial Public Art Accelerator fellowship and later commissioned by the city to create the piece — the first artwork installed in the refurbished section of the plaza. She invited people to workshops and community centers to create hundreds of “wishmarks” — steel panels with hammered indentations and words, each representing a personal wish or reflection.

The plates were later used to form the metal skin of the sculpture — three bird-like forms designed to be, in Vedro’s words, a “protective amulet for the landscape.”

“I didn’t ask anyone to share what their actual wishes were, but I met people going into surgery, people who were homeless and looking for housing, people who had just lost a loved one, people dealing with immigration issues,” Vedro said. She asked participants to meditate on the idea of a journey and safe passage. “That could be a literal journey with ideas around immigration and migration,” she said, “or it could be your own internal journey.”

Large-scale art, fine-scale detail

Vedro, who has several public artworks to her name, said in a video about making “Amulet” that the project was “the biggest thing I’ve ever done.” While artworks of this scale are often handed off to fabrication teams, she handled the construction herself, starting on her driveway until zoning rules forced her to move to her father-in-law’s warehouse. Sections were also welded at Artisans Asylum, a community workshop in Boston, where she was an artist in residence, and then moved to a large industrial studio in Rhode Island.

At the ribbon-cutting event, Vedro thanked friends, family members, and city officials who helped bring the project to life. The celebration ended with a concert by musician Veronica Robles and her mariachi band. Robles runs the Veronica Robles Cultural Center in East Boston, which served as the main site for wishmark workshops. The sculpture is expected to remain in City Hall Plaza for up to five years.

Vedro’s background is in fine arts metalsmithing, a discipline that involves shaping and manipulating metals like silver, gold, and copper through forging, casting, and soldering. She began working at a very different scale, making jewelry, and then later moved primarily to welded steel sculpture — both techniques she now teaches at MIT. When working with steel, Vedro applies the same sensitivity a jeweler brings to small objects, paying close attention to small undulations and surface texture.

She loves working with steel, Vedro says — “shaping and forming and texturing and fighting with it” — because it allows her to engage physically with the material, with her hands involved in every millimeter.

The sculpture’s fluid design began with loose, free-form bird drawings on a cement floor and rubber panels with soapstone, oil pastels, and paint sticks. Vedro then built the forms in metal, welding three-dimensional armatures from round steel bars. The organic shapes and flourishes emerged through a responsive, intuitive process.

“I’m someone who works in real-time, changing my mind and responding to the material,” Vedro says. She likens her process to making a patchwork quilt of steel pieces: forming patterns in a shapeable material like tar paper, transferring them to steel sheets, cutting and shaping and texturing the pieces, and welding them together. “So I can get lots of curvatures that way that are not at all modular.”

From steel plates to soaring form

The sculpture’s outer skin is made from thin, 20-gauge mild steel — a low-carbon steel that’s relatively soft and easy to work with — used for the wishmarks. Those plates were fitted over an internal armature constructed from heavier structural steel.

Because there were more wishmark panels than surface area, Vedro slipped some of them into the hollow space inside the sculpture before welding the piece closed. She compares them to treasures in a locket, “loose, rattling around, which freaked out the team when they were installing.” Any written text on the panels was burned off when the pieces were welded together.

“I believe the stuff’s all alchemized up into smoke, which to me is wonderful because it traverses realms just like a bird,” she says.

The surface of the sculpture is coated with a sealant — necessary because the outer skin material is prone to rust — along with spray paints, patinas, and accents including gold leaf. Its appearance will change over time, something Vedro embraces.

“The idea of transformation is actually integral to my work,” she says.

Standing outside the warmth of the Civic Pavilion on a windy, rainy day, artist Matt Bajor described the sculpture as “gorgeous,” attributing its impact in part to Vedro’s fluency in working across vastly different scales.

“The attention to detail — paying attention to the smaller things so that as it comes together as a whole, you have that fineness throughout the whole sculpture,” he said. “To do that at such a large scale is just crazy. It takes a lot of skill, a lot of effort, and a lot of time.”

Suveena Sreenilayam, a DMSE graduate student who has worked closely with Vedro, said her understanding of the relationship between art and craft brings a unique dimension to her work.

“Metal is hard to work with — and to build that on such small and large scales indicates real versatility,” Sreenilayam said. “To make something so artistic at this scale reflects her physical talent, and also her eye for detail and expression.”

Bajor said “Amulet” is a striking addition to the plaza, where the clean lines of City Hall’s Brutalist architecture contrast with the sculpture’s sinuous curves — and to Boston itself.

“I’m looking forward to seeing it in different conditions — in snow and bright sun — as the metal changes over time and as the patina develops,” he said. “It’s just a really great addition to the city.”

Through the MITx MicroMasters Program in Data, Economics, and Design of Policy, Munip Utama strengthened the skills he was already applying in his work with Baitul Enza, a nonprofit helping students in need via policy-shaping research and hands-on assistance. 

Utama’s commitment to advancing education for underprivileged students stems from his own background. His father is an elementary school teacher in a remote area and his mother has passed away. While financial hardship has always been a defining challenge, he says it has also been the driving force behind his pursuit of education. With the assistance of special programs for high-achieving students, Utama attended top schools and completed his bachelor’s degree in economics at UIN Jakarta — becoming the second person in his family to earn a university degree.

Utama joined Baitul Enza two months before graduation, through a faculty-led research project, and later became its manager, leading its programs and future development. In this interview, he describes how his experiences with the MicroMasters Program in Data, Economics, and Design of Policy (DEDP), offered by the Abdul Latif Jameel Poverty Action Lab (J-PAL) and MIT Open Learning, are shaping his education, career, and personal mission.

Q: What motivated you to pursue the MITx MicroMasters Program in Data, Economics, and Design of Policy?

A: I was seeking high-quality, evidence-based courses in economics and development. I needed rigorous training in data analysis, economic reasoning, and policy design to strengthen our interventions at Baitul Enza. The MITx MicroMasters Program in Data, Economics, and Design of Policy offered exactly that: a curriculum grounded in real-world problem-solving, aligned with the challenges I face in Indonesia.

I deeply admire MIT’s commitment to transforming teaching and learning not only through innovation, but also through empathy. The DEDP program exemplifies this mission: It connects theory with practice, allowing learners like me to apply analytical tools directly to real development challenges. This approach has inspired me to adopt the same philosophy in my own teaching and mentoring, encouraging students to use data and critical thinking to solve problems in their communities.

Q: What have you gained from the MITx DEDP program? 

A: The DEDP courses have provided me with rigorous analytical and quantitative training in data analysis, economics, and policy design. They have strengthened both my research and mentorship abilities by teaching me to approach poverty and inequality through evidence-based frameworks. My experience conducting independent and collaborative research projects has informed how I mentor students, guiding them to carry out their own evidence-based research projects. I continue to seek further academic dialogue to broaden my understanding and prepare for future graduate studies.

Another key component has been the program’s financial assistance offers. Even with DEDP’s personalized income-based course pricing, financial constraints remain a significant challenge for me, and Baitul Enza operates entirely on donations and volunteer support. The scholarships administered by DEDP have been crucial in enabling me to continue my studies. It has allowed me to focus on learning without the constant burden of financial insecurity, while staying committed to my mission of breaking cycles of poverty through education. 

Q: How are you applying what you’ve learned from MIT Open Learning’s MITx programs, and how will you use what you’ve learned going forward?

A: The DEDP program has transformed how I lead Baitul Enza. I now apply data-driven and evidence-based approaches to program design, monitoring, and evaluation — enhancing cost-effectiveness and long-term impact. The program has enabled me to design case-based learning modules for students, where they analyze real-world data on poverty and education; mentor youth researchers to conduct small-scale projects using evidence-based methods; and improve program cost-effectiveness and outcome measurement to attract collaborators and government support.

Coming from a lower-middle-class family with limited access to education, MIT Open Learning has opened doors I never imagined possible. It has reaffirmed my belief that education, grounded in data and empathy, can break the cycle of poverty. The DEDP program continues to inspire me to mentor young researchers, empower disadvantaged students, and build a community rooted in evidence-based decision-making.

With the foundation built by MITx, I aim to produce policy-relevant research and scale up Baitul Enza’s impact. My long-term vision is to generate experimental evidence in Indonesia on scalable education interventions, inform national policy, and empower marginalized youth to thrive. MITx has not only prepared me academically, but has also strengthened my resolve to lead with clarity, design with evidence, and act with purpose. Beyond my own growth, MITx has multiplied its impact by empowering the next generation of students to use data and evidence in solving local development challenges.

MIT researchers have designed silicon structures that can perform calculations in an electronic device using excess heat instead of electricity. These tiny structures could someday enable more energy-efficient computation.

In this computing method, input data are encoded as a set of temperatures using the waste heat already present in a device. The flow and distribution of heat through a specially designed material forms the basis of the calculation. Then the output is represented by the power collected at the other end, which is thermostat at a fixed temperature.      

The researchers used these structures to perform matrix vector multiplication with more than 99 percent accuracy. Matrix multiplication is the fundamental mathematical technique machine-learning models like LLMs utilize to process information and make predictions.

While the researchers still have to overcome many challenges to scale up this computing method for modern deep-learning models, the technique could be applied to detect heat sources and measure temperature changes in electronics without consuming extra energy. This would also eliminate the need for multiple temperature sensors that take up space on a chip.

“Most of the time, when you are performing computations in an electronic device, heat is the waste product. You often want to get rid of as much heat as you can. But here, we’ve taken the opposite approach by using heat as a form of information itself and showing that computing with heat is possible,” says Caio Silva, an undergraduate student in the Department of Physics and lead author of a paper on the new computing paradigm.

Silva is joined on the paper by senior author Giuseppe Romano, a research scientist at MIT’s Institute for Soldier Nanotechnologies and a member of the MIT-IBM Watson AI Lab. The research appears today in Physical Review Applied.

Turning up the heat

This work was enabled by a software system the researchers previously developed that allows them to automatically design a material that can conduct heat in a specific manner.

Using a technique called inverse design, this system flips the traditional engineering approach on its head. The researchers define the functionality they want first, then the system uses powerful algorithms to iteratively design the best geometry for the task.

They used this system to design complex silicon structures, each roughly the same size as a dust particle, that can perform computations using heat conduction. This is a form of analog computing, in which data are encoded and signals are processed using continuous values, rather than digital bits that are either 0s or 1s.

The researchers feed their software system the specifications of a matrix of numbers that represents a particular calculation. Using a grid, the system designs a set of rectangular silicon structures filled with tiny pores. The system continually adjusts each pixel in the grid until it arrives at the desired mathematical function.

Heat diffuses through the silicon in a way that performs the matrix multiplication, with the geometry of the structure encoding the coefficients.

“These structures are far too complicated for us to come up with just through our own intuition. We need to teach a computer to design them for us. That is what makes inverse design a very powerful technique,” Romano says.

But the researchers ran into a problem. Due to the laws of heat conduction, which impose that heat goes from hot to cold regions, these structures can only encode positive coefficients. 

They overcame this problem by splitting the target matrix into its positive and negative components and representing them with separately optimized silicon structures that encode positive entries. Subtracting the outputs at a later stage allows them to compute negative matrix values.

They can also tune the thickness of the structures, which allows them to realize a greater variety of matrices. Thicker structures have greater heat conduction.

“Finding the right topology for a given matrix is challenging. We beat this problem by developing an optimization algorithm that ensures the topology being developed is as close as possible to the desired matrix without having any weird parts,” Silva explains.

Microelectronic applications

The researchers used simulations to test the structures on simple matrices with two or three columns. While simple, these small matrices are relevant for important applications, such as fusion sensing and diagnostics in microelectronics.     

The structures performed computations with more than 99 percent accuracy in many cases.

However, there is still a long way to go before this technique could be used for large-scale applications such as deep learning, since millions of structures would need to be tiled together. As the matrices become more complicated, the structures become less accurate, especially when there is a large distance between the input and output terminals. In addition, the devices have limited bandwidth, which would need to be greatly expanded if they were to be used for deep learning.

But because the structures rely on excess heat, they could be directly applied for tasks like thermal management, as well as heat source or temperature gradient detection in microelectronics.

“This information is critical. Temperature gradients can cause thermal expansion and damage a circuit or even cause an entire device to fail. If we have a localized  heat source where we don’t want a heat source, it means we have a problem. We could directly detect such heat sources with these structures, and we can just plug them in without needing any digital components,” Romano says.

Building on this proof-of-concept, the researchers want to design structures that can perform sequential operations, where the output of one structure becomes an input for the next. This is how machine-learning models perform computations. They also plan to develop programmable structures, enabling them to encode different matrices without starting from scratch with a new structure each time.

Keeril Makan has been appointed vice provost for the arts at MIT, effective Feb. 1. In this role, Makan, who is the Michael (1949) and Sonja Koerner Music Composition Professor at MIT, will provide leadership and strategic direction for the arts across the Institute.

Provost Anantha Chandrakasan announced Makan’s appointment in an email to the MIT community today.

“Keeril’s record of accomplishment both as an artist and an administrative leader makes him exceedingly qualified to take on this important role,” Chandrakasan wrote, noting that Makan “has repeatedly taken on new leadership assignments with skill and enthusiasm.”

Makan’s appointment follows the publication last September of the final report of the Future of the Arts at MIT Committee. At MIT, the report noted, “the arts thrive as a constellation of recognized disciplines while penetrating and illuminating countless aspects of the Institute’s scientific and technological enterprise.” Makan will build on this foundation as MIT continues to strengthen the role of the arts in research, education, and community life.

As vice provost for the arts, Makan will provide Institute-wide leadership and strategic direction for the arts, working in close partnership with academic leaders, arts units, and administrative colleagues across MIT, including the Office of the Arts; the MIT Center for Art, Science and Technology; the MIT Museum; the List Visual Arts Center; and the Council for the Arts at MIT. His role will focus on strengthening connections between artistic practice, research, education, and community life, and on supporting public engagement and interdisciplinary collaboration.

“At MIT, the arts are a vital way of thinking, making, and convening,” Makan says. “As vice provost, my priority is to support and strengthen the extraordinary artistic work already happening across the Institute, while listening carefully to faculty, students, and staff as we shape what comes next. I’m excited to build on MIT’s distinctive, only-at-MIT approach to the arts and to help ensure that artistic practice remains central to MIT’s intellectual and community life.”

Makan says he will begin his new role with a period of listening and learning across MIT’s arts ecosystem, informed by the Future of the Arts at MIT report. His initial focus will be on understanding how artistic practice intersects with research, education, and community life, and on identifying opportunities to strengthen connections, visibility, and coordination across MIT’s many arts activities.

Over time, Makan says he will work with the arts community to advance MIT’s long-standing commitment to artistic excellence and experimentation, while supporting student participation and public engagement in the arts. He said his approach will “emphasize collaboration, clarity, and sustainability, reflecting MIT’s distinctive integration of the arts with science and technology.”

Makan came to MIT in 2006 as an assistant professor of music. From 2018 to 2024, he served as head of the Music and Theater Arts (MTA) Section in the School of Humanities, Arts, and Social Sciences (SHASS). In 2023, he was appointed associate dean for strategic initiatives in SHASS, where he helped guide the school’s response to recent fiscal pressures and led Institute-wide strategic initiatives.

With colleagues from MTA and the School of Engineering, Makan helped launch a new, multidisciplinary graduate program in music technology and computation. He was intimately involved in the project to develop the new Edward and Joyce Linde Music Building (Building 18), a state-of-the-art facility that opened in 2025. 

Makan was a member of the Future of the Arts at MIT Committee and chaired a working group on the creation of a center for the humanities, which ultimately became the MIT Human Insight Collaborative (MITHIC), one of the Institute’s strategic initiatives. Since last year, he has served as MITHIC’s faculty lead. Under his leadership, MITHIC has awarded $4.7 million in funding to 56 projects across 28 units at MIT, supporting interdisciplinary, human-centered research and teaching.

Trained initially as a violinist, Makan earned undergraduate degrees in music composition and religion from Oberlin and a PhD in music composition from the University of California at Berkeley.

A critically-acclaimed composer, Makan is the recipient of a Guggenheim Fellowship and the Luciano Berio Rome Prize from the American Academy in Rome. His music has been recorded by the Kronos Quartet, the Boston Modern Orchestra Project, and the International Contemporary Ensemble, and performed at Carnegie Hall, the Lincoln Center for the Performing Arts, and Tanglewood. His opera, “Persona,” premiered at National Sawdust and was performed at the Isabella Stewart Gardner Museum in Boston and by the Los Angeles Opera. The Los Angeles Times described the music from “Persona” as “brilliant.”

Makan succeeds Philip Khoury, the Ford International Professor of History, who served as vice provost for the arts from 2006 before stepping down in 2025. Khoury will return to the MIT faculty following a sabbatical.

In its first moments, the infant universe was a trillion-degree-hot soup of quarks and gluons. These elementary particles zinged around at light speed, creating a “quark-gluon plasma” that lasted for only a few millionths of a second. The primordial goo then quickly cooled, and its individual quarks and gluons fused to form the protons, neutrons, and other fundamental particles that exist today.

Physicists at CERN’s Large Hadron Collider in Switzerland are recreating quark-gluon plasma (QGP) to better understand the universe’s starting ingredients. By smashing together heavy ions at close to light speeds, scientists can briefly dislodge quarks and gluons to create and study the same material that existed during the first microseconds of the early universe.

Now, a team at CERN led by MIT physicists has observed clear signs that quarks create wakes as they speed through the plasma, similar to a duck trailing ripples through water. The findings are the first direct evidence that quark-gluon plasma reacts to speeding particles as a single fluid, sloshing and splashing in response, rather than scattering randomly like individual particles.

“It has been a long debate in our field, on whether the plasma should respond to a quark,” says Yen-Jie Lee, professor of physics at MIT. “Now we see the plasma is incredibly dense, such that it is able to slow down a quark, and produces splashes and swirls like a liquid. So quark-gluon plasma really is a primordial soup.”

To see a quark’s wake effects, Lee and his colleagues developed a new technique that they report in the study. They plan to apply the approach to more particle-collision data to zero in on other quark wakes. Measuring the size, speed, and extent of these wakes, and how long it takes for them to ebb and dissipate, can give scientists an idea of the properties of the plasma itself, and how quark-gluon plasma might have behaved in the universe’s first microseconds.

“Studying how quark wakes bounce back and forth will give us new insights on the quark-gluon plasma’s properties,” Lee says. “With this experiment, we are taking a snapshot of this primordial quark soup.”

The study’s co-authors are members of the CMS Collaboration — a team of particle physicists from around the world who work together to carry out and analyze data from the Compact Muon Solenoid (CMS) experiment, which is one of the general-purpose particle detectors at CERN’s Large Hadron Collider. The CMS experiment was used to detect signs of quark wake effects for this study. The open-access study appears in the journal Physics Letters B.

Quark shadows

Quark-gluon plasma is the first liquid to have ever existed in the universe. It is also the hottest liquid ever, as scientists estimate that during its brief existence, the QGP was around a few trillion degrees Celsius. This boiling stew is also thought to have been a near-“perfect” liquid, meaning that the individual quarks and gluons in the plasma flowed together as a smooth, frictionless fluid.

This picture of the QGP is based on many independent experiments and theoretical models. One such model, derived by Krishna Rajagopal, the William A. M. Burden Professor of Physics at MIT, and his collaborators, predicts that the quark-gluon plasma should respond like a fluid to any particles speeding through it. His theory, known as the hybrid model, suggests that when a jet of quarks is zinging through the QGP, it should produce a wake behind it, inducing the plasma to ripple and splash in response.

Physicists have looked for such wake effects in experiments at the Large Hadron Collider and other high-energy particle accelerators. These experiments whip up heavy ions such as lead, to close to the speed of light, at which point they can collide and produce a short-lived droplet of primordial soup, typically lasting for less than a quadrillionth of a second. Scientists essentially take a snapshot of the moment to try and identify characteristics of the QGP.

To identify quark wakes, physicists have looked for pairs of quarks and “antiquarks” — particles that are identical to their quark counterparts, except that certain properties are equal in magnitude but opposite in sign. For instance, when a quark is speeding through plasma, there is likely an antiquark that is traveling at exactly the same speed, but in the opposite direction.

For this reason, physicists have looked for quark/antiquark pairs in the QGP produced in heavy-ion collisions, assuming that the particles might produce identical, detectable wakes through the plasma.

“When you have two quarks produced, the problem is that, when the two quarks go in opposite directions, the one quark overshadows the wake of the second quark,” Lee says.

He and his colleagues realized that looking for the wake of the first quark would be easier if there were no second quark obscuring its effects.

“We have figured out a new technique that allows us to see the effects of a single quark in the QGP, through a different pair of particles,” Lee says.

A wake tag

Rather than search for pairs of quarks and antiquarks in the aftermath of lead ion collisions, Lee’s team instead looked for events with only one quark moving through the plasma, essentially back-to-back with a “Z boson.” A Z boson is a neutral, electrically weak elementary particle that has virtually no effect on the surrounding environment. However, because they exist at a very specific energy, Z bosons are relatively straightforward to detect.

“In this soup of quark-gluon plasma, there are numerous quarks and gluons passing by and colliding with each other,” Lee explains. “Sometimes when we are lucky, one of these collisions creates a Z boson and a quark, with high momentum.”

In such a collision, the two particles should hit each other and fly off in exact opposite directions. While the quark could leave a wake, the Z boson should have no effect on the surrounding plasma. Whatever ripples are observed in the droplet of primordial soup would have been made entirely by the single quark zipping through it.

The team, in collaboration with Professor Yi Chen’s group at Vanderbilt University, reasoned that they could use Z bosons as a “tag” to locate and trace the wake effects of single quarks. For their new study, the researchers looked through data from the Large Hadron Collider’s heavy-ion collision experiments. From 13 billion collisions, they identified about 2,000 events that produced a Z boson. For each of these events, they mapped the energies throughout the short-lived quark-gluon plasma, and consistently observed a fluid-like pattern of splashes in swirls — a wake effect — in the opposite direction of the Z bosons, which the team could directly attribute to the effect of single quarks zooming through the plasma.

What’s more, the physicists found that the wake effects they observed in the data were consistent with what Rajagopal’s hybrid model predicts. In other words, quark-gluon plasma does in fact flow and ripple like a fluid when particles speed through it.

“This is something that many of us have argued must be there for a good many years, and that many experiments have looked for,” says Rajagopal, who was not directly involved with the new study.

“What Yen-Jie and CMS have done is to devise and execute a measurement that has brought them and us the first clean, clear, unambiguous, evidence for this foundational phenomenon,” says Daniel Pablos, professor of physics at Oviedo University in Spain and a collaborator of Rajagopal’s who was not involved in the current study.

“We’ve gained the first direct evidence that the quark indeed drags more plasma with it as it travels,” Lee adds. “This will enable us to study the properties and behavior of this exotic fluid in unprecedented detail.”

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

The Pappalardo Apprentice program pushes the boundaries of the traditional lab experience, inviting a selected group of juniors and seniors to advance their fabrication skills while also providing mentor training and peer-to-peer mentoring opportunities in an environment fueled by creativity, safety, and fun.

“This apprenticeship was largely born of my need for additional lab help during our larger sophomore-level design course, and the desire of third- and fourth-year students to advance their fabrication knowledge and skills,” says Daniel Braunstein, senior lecturer in mechanical engineering (MechE) and director of the Pappalardo Undergraduate Teaching Laboratories. “Though these needs and wants were nothing particularly new, it had not occurred to me that we could combine these interests into a manageable and meaningful program.”

Apprentices serve as undergraduate lab assistants for class 2.007 (Design and Manufacturing I), joining lab sessions and assisting 2.007 students with various aspects of the learning experience including machining, hand-tool use, brainstorming, and peer support. Apprentices also participate in a series of seminars and clinics designed to further their fabrication knowledge and hands-on skills, including advancing understanding of mill and lathe use, computer-aided design and manufacturing (CAD/CAM) and pattern-making.

Putting this learning into practice, junior apprentices fabricate Stirling engines (a closed-cycle heat engine that converts thermal energy into mechanical work), while returning senior apprentices take on more ambitious group projects involving casting. Previous years’ projects included an early 20th-century single-cylinder marine engine and a 19th-century torpedo boat steam engine, on permanent exhibit at the MIT Museum. This spring will focus on copper alloys and fabrication of a replica of an 1899 anchor windlass from the Herreshoff Manufacturing Co., used on the famous New York 70 class sloops.

The sloops, designed by MIT Class of 1870 alumnus Nathanael Greene Herreshoff for wealthy New York Yacht Club members, were a short-lived, single-design racing vessels meant for exclusive competition. The historic racing yachts used robust manual windlasses — mechanical devices used to haul large loads — to manage their substantial anchors.

“The more we got into casting, I was modestly surprised that [the students’] exposure to metals was very limited. So that really launched not just a project, but also a more specific curriculum around metallurgy,” says Braunstein.

Metallurgy is not a traditional part of the curriculum. “I think [the project] really opened up my eyes to how much material choice is an important thing for engineering in general,” says apprentice Jade Durham.

In casting the windlasses, students are working from century-old drawings. “[Looking at these old drawings,] we don't know how they made [the parts],” says Braunstein. “So, there is an element of the discovery of what they may or may not have done. It’s like technical archaeology.”

“You’re really just relying on your knowledge of the windlass system, how it’s meant to work, which surfaces are really critical, and kind of just applying your intuition,” says apprentice Saechow Yap. “I learned a lot about applying my art skills and my ability to judge and shape aesthetic.”

Learning by doing is an important hallmark of an MIT MechE education. The Pappalardo Apprentice Program, which celebrated its 10th year last spring, is housed in the Pappalardo Lab. The lab, established through a gift from Neil Pappalardo ’64, is the self-proclaimed “most wicked labs on campus” — “wicked,” for readers outside of Greater Boston, is slang used in a variety of ways, but generally meaning something is pretty awesome.

“Pappalardo is my favorite place on campus, I had never set foot in any sort of like makerspace or lab before I came to MIT,” says apprentice Wilhem Hector. “I did not just learn how to make things. I got empowered ... [to] make anything.”

Braunstein developed the Pappalardo Apprentice program to reinforce the learning of the older students while building community. In a 2023 interview, he said he called the seminar an apprenticeship to emphasize MIT’s relationship with the art — and industrial character — of engineering.

“I did want to borrow from the language of the trades,” Braunstein said. “MIT has a strong heritage in industrial work; that’s why we were founded. It was not a science institution; it was about the mechanical arts. And I think the blend of the industrial, plus the academic, is what makes this lab particularly meaningful.”

Today, he says the most enjoyable part of the program, for him, is watching relationships develop. “They come in, bright-eyed, bushy-tailed, and then to see them go to people who are capable of pouring iron, tramming mills, teaching other people how to do it and having this tight group of friends … that's fun to watch.”

Collaborators from across the Commonwealth of Massachusetts came together in December for a daylong summit of the Massachusetts Prison Education Consortium (MPEC), hosted by the Educational Justice Institute (TEJI) at MIT. Held at MIT’s Walker Memorial, the summit aimed to expand access to high-quality education for incarcerated learners and featured presentations by leaders alongside strategy sessions designed to turn ideas into concrete plans to improve equitable access to higher education and reduce recidivism in local communities.

In addition to a keynote address by author and resilience expert Shaka Senghor, speakers such as Molly Lasagna, senior strategy officer in the Ascendium Education Group, and Stefan LoBuglio, former director of the National Institute of Corrections, discussed the roles of learning, healing, and community support in building a more just system for justice-impacted individuals.

The MPEC summit, “Building Integrated Systems Together: Massachusetts Community Colleges and County Corrections 2.0,” addressed three key issues surrounding equitable education: the integration of Massachusetts community college education with county corrections to provide incarcerated individuals with access to higher education; the integration of carceral education with industry to expand work and credentialing opportunities; and the goal of better serving women who experience unique challenges within the criminal legal system.

Created by TEJI, MPEC is a statewide network of Massachusetts colleges, organizations and correctional partners working together to expand access to high-quality, credit-bearing education in Massachusetts prisons and jails. The consortium works on all levels of the pipeline, from academic programming, faculty support, research, reentry pathways, and more, drawing from the research and success of the MIT Prison Education Initiative and the recent restoration of Pell Grant eligibility for incarcerated learners.

The summit was hosted by TEJI co-directors Lee Perlman and Carole Cafferty. Perlman founded the MIT Prison Initiative after years of teaching in MIT’s Experimental Study Group (ESG) and in correctional classrooms. He has been recognized for his work in bringing humanities education to prison settings with three Irwin Sizer Awards and MIT’s Martin Luther King Jr. Leadership Award.

Cafferty jointly co-founded TEJI after more than 30 years’ experience with corrections, including working as superintendent of the Middlesex Jail and House of Correction. She now guides the institute with the knowledge she gained from building integrative and therapeutic educational programs that have since been replicated nationally.

“TEJI serves two populations, incarcerated learners and the MIT community. All of our classes involve MIT students, either learning alongside the incarcerated students or as TAs [teaching assistants],” emphasizes Perlman. In discussing the unification of TEJI with the roles and experiences MIT students take, Perlman further notes: “Our humanities classes, which we call our philosophical life skills curriculum, give MIT students the opportunity to discuss how we want to live our lives with incarcerated students with very different backgrounds.”

These courses, offered through ESG, are subjects with a unique focus that often differ from the traditional focus of a more academic course, often prioritizing hands-on learning and innovative teaching methods. Perlman’s courses are almost always taught in a carceral setting, and he notes that these courses can be highly impactful on the MIT community: “In courses like Philosophy of Love; Non-violence as a Way of Life; and Authenticity and Emotional Intelligence for Teams, the discussions are rich and personal. Many MIT students have described their experience in these classes as life-changing.”

Throughout morning addresses and afternoon strategy sessions, summit attendees developed concrete plans for scaling classroom capacity, aligning curricula with regional labor markets, and strengthening academic and reentry supports that help students remain on the right path after release. Panels explored practical issues, such as how to coordinate registration and credit transfer when a student moves between facilities and how to staff hybrid classrooms that combine in-person and remote instruction, as well as how to measure program outcomes beyond enrollment.

Co-directors Perlman and Cafferty highlighted that the average length of stay within these programs in county facilities is only six months, and that inspired a particular focus on making sure these programs are high-impact even when community members are only able to participate for a short period of time.

Speakers repeatedly emphasized that these logistical challenges often sit atop deeper, more human challenges. In his keynote, Shaka Senghor traced his own journey from trauma to transformation, stressing the power of reading, mentorship, and completing something of one’s own. “What else can you do with your mind?” he asked, describing the moment he realized that the act of reading and writing could change the trajectory of his life.

The line became a refrain throughout the day, a question that caused all to reflect on how prison education could not only function as a workforce pathway, but as a catalyst for dignity and hope after reentry. Senghor also directly confronted the stigma that returning citizens face. “They said I’d be back in prison in six months,” he recalled, using the remark from a corrections officer from the day he was released on parole as a reminder of the structural and social barriers encountered after release.

The summit also brought together funders and implementers who are shaping the field’s future. Molly Lasagna of Ascendium Education Group described the organization’s strategy of “Expand, Support, Connect,” which funds the creation of new educational programs, strengthens basic needs and advising infrastructure, and ensures that individuals leaving prison can transition into high-quality employment opportunities. “How is this education program putting somebody on a pathway to opportunity?” she asked, noting that true change requires aligning education, reentry, and workforce systems.

Participants also heard from Stefan LoBuglio, former director of the National Institute of Corrections and a national thought leader in corrections and reentry, who lauded Massachusetts as a leader while cautioning that staffing shortages, limited program space, and uneven access to technology continue to constrain progress. “We have a crisis in staffing in corrections that does affect our educational programs,” he noted, calling for attention to staff wellness and institutional support as essential components of sustainability.

Throughout the day, TEJI and MPEC leaders highlighted emerging pilots and partnerships, including a new “Prisons to Pathways” initiative aimed at building stackable, transferable credentials aligned with regional industry needs. Additional collaborations with the American Institutes for Research will support new implementation guides and technical assistance resources designed by practitioners in the field.

The summit concluded with a commitment to sustain collaboration. As Senghor reminded participants, the work is both practical and moral. The question he posed, “What else can you do with your mind?,” serves as a reminder to Massachusetts educators, corrections partners, funders, and community organizations to ensure that learning inside prison becomes a foundation for opportunity outside it.