Every year, thousands of students take courses that teach them how to deploy artificial intelligence models that can help doctors diagnose disease and determine appropriate treatments. However, many of these courses omit a key element: training students to detect flaws in the training data used to develop the models.

Leo Anthony Celi, a senior research scientist at MIT’s Institute for Medical Engineering and Science, a physician at Beth Israel Deaconess Medical Center, and an associate professor at Harvard Medical School, has documented these shortcomings in a new paper and hopes to persuade course developers to teach students to more thoroughly evaluate their data before incorporating it into their models. Many previous studies have found that models trained mostly on clinical data from white males don’t work well when applied to people from other groups. Here, Celi describes the impact of such bias and how educators might address it in their teachings about AI models.

Q: How does bias get into these datasets, and how can these shortcomings be addressed?

A: Any problems in the data will be baked into any modeling of the data. In the past we have described instruments and devices that don’t work well across individuals. As one example, we found that pulse oximeters overestimate oxygen levels for people of color, because there weren’t enough people of color enrolled in the clinical trials of the devices. We remind our students that medical devices and equipment are optimized on healthy young males. They were never optimized for an 80-year-old woman with heart failure, and yet we use them for those purposes. And the FDA does not require that a device work well on this diverse of a population that we will be using it on. All they need is proof that it works on healthy subjects.

Additionally, the electronic health record system is in no shape to be used as the building blocks of AI. Those records were not designed to be a learning system, and for that reason, you have to be really careful about using electronic health records. The electronic health record system is to be replaced, but that’s not going to happen anytime soon, so we need to be smarter. We need to be more creative about using the data that we have now, no matter how bad they are, in building algorithms.

One promising avenue that we are exploring is the development of a transformer model of numeric electronic health record data, including but not limited to laboratory test results. Modeling the underlying relationship between the laboratory tests, the vital signs and the treatments can mitigate the effect of missing data as a result of social determinants of health and provider implicit biases.

Q: Why is it important for courses in AI to cover the sources of potential bias? What did you find when you analyzed such courses’ content?

A: Our course at MIT started in 2016, and at some point we realized that we were encouraging people to race to build models that are overfitted to some statistical measure of model performance, when in fact the data that we’re using is rife with problems that people are not aware of. At that time, we were wondering: How common is this problem?

Our suspicion was that if you looked at the courses where the syllabus is available online, or the online courses, that none of them even bothers to tell the students that they should be paranoid about the data. And true enough, when we looked at the different online courses, it’s all about building the model. How do you build the model? How do you visualize the data? We found that of 11 courses we reviewed, only five included sections on bias in datasets, and only two contained any significant discussion of bias.

That said, we cannot discount the value of these courses. I’ve heard lots of stories where people self-study based on these online courses, but at the same time, given how influential they are, how impactful they are, we need to really double down on requiring them to teach the right skillsets, as more and more people are drawn to this AI multiverse. It’s important for people to really equip themselves with the agency to be able to work with AI. We’re hoping that this paper will shine a spotlight on this huge gap in the way we teach AI now to our students.

Q: What kind of content should course developers be incorporating?

A: One, giving them a checklist of questions in the beginning. Where did this data came from? Who were the observers? Who were the doctors and nurses who collected the data? And then learn a little bit about the landscape of those institutions. If it’s an ICU database, they need to ask who makes it to the ICU, and who doesn’t make it to the ICU, because that already introduces a sampling selection bias. If all the minority patients don’t even get admitted to the ICU because they cannot reach the ICU in time, then the models are not going to work for them. Truly, to me, 50 percent of the course content should really be understanding the data, if not more, because the modeling itself is easy once you understand the data.

Since 2014, the MIT Critical Data consortium has been organizing datathons (data “hackathons”) around the world. At these gatherings, doctors, nurses, other health care workers, and data scientists get together to comb through databases and try to examine health and disease in the local context. Textbooks and journal papers present diseases based on observations and trials involving a narrow demographic typically from countries with resources for research. 

Our main objective now, what we want to teach them, is critical thinking skills. And the main ingredient for critical thinking is bringing together people with different backgrounds.

You cannot teach critical thinking in a room full of CEOs or in a room full of doctors. The environment is just not there. When we have datathons, we don’t even have to teach them how do you do critical thinking. As soon as you bring the right mix of people — and it’s not just coming from different backgrounds but from different generations — you don’t even have to tell them how to think critically. It just happens. The environment is right for that kind of thinking. So, we now tell our participants and our students, please, please do not start building any model unless you truly understand how the data came about, which patients made it into the database, what devices were used to measure, and are those devices consistently accurate across individuals?

When we have events around the world, we encourage them to look for data sets that are local, so that they are relevant. There’s resistance because they know that they will discover how bad their data sets are. We say that that’s fine. This is how you fix that. If you don’t know how bad they are, you’re going to continue collecting them in a very bad manner and they’re useless. You have to acknowledge that you’re not going to get it right the first time, and that’s perfectly fine. MIMIC (the Medical Information Marked for Intensive Care database built at Beth Israel Deaconess Medical Center) took a decade before we had a decent schema, and we only have a decent schema because people were telling us how bad MIMIC was.

We may not have the answers to all of these questions, but we can evoke something in people that helps them realize that there are so many problems in the data. I’m always thrilled to look at the blog posts from people who attended a datathon, who say that their world has changed. Now they’re more excited about the field because they realize the immense potential, but also the immense risk of harm if they don’t do this correctly.

Below is the text of Melissa Nobles’ remarks, as prepared for delivery today.

Wow, thank you Emily and Andrew! Emily Jin on vocals and Andrew Li on saxophone, and their fellow musicians!

Class of 2025! Look at you, you’re looking really good in your regalia! It’s your graduation day! You did it! Congratulations!

And congratulations to all of your loved ones, all of the people who helped support you.

Your parents, your brothers and sisters, your aunties and your uncles, and your friends. This is a special day for them too. They are so proud of you!

A warm welcome to the loved ones who are here with us today on Killian Court — they’ve come here from all over to celebrate you!

And a special shout out to those who are watching from afar, wishing they could be here with you in person!

Class of 2025, you’ve made a lot of memories during your time here: from classes to crushes, from the East Campus REX build to the Simmons ball pit to Next Haunt, from UROPs to the Hobby Shop, and from the Outfinite to the Infinite!

So, I’d like to take you back to the fall of 2021, when you arrived here at MIT.

You traveled from all parts of this country and the world — from 62 countries, to be exact — and landed right here in Cambridge. Together, you became MIT’s Class of 2025.

And you arrived on campus — all bright-eyed and beaver-tailed — after missing a lot of in-person high school rituals, a lot of the high school experience. So, you were extra eager for college, and, more specifically, super excited to be MIT students!

Although the campus was officially fully open for the first time since the Covid shutdown — students, staff, and faculty were all here in person, with Zoom taking a back seat to meeting in real life — there were still a lot of protocols in place.

You had to get through all the Covid tests because we were still testing. Do you remember those Ziploc bags?

You swabbed and submitted attestations because you wanted the keys to unlock doors to labs, classrooms, and all the experiences that make MIT, MIT.

And once you gained access, you discovered a campus that was shiny and welcoming, yet dusty after being mostly empty for a long while. And there was no manual for how to reanimate this place.

You didn’t flinch.

You chose MIT because you like to solve problems, and your inner beaver came out to bring the campus back to life, to make it a home.

You were curious, you surveyed the landscape, and you started to dig into the past in order to build your future.

You sought out seniors, the Class of 2022, to read you in, to show you the ropes, and they really came through for you. They felt the urgency of their limited time left on campus, and they taught you “how to MIT.”

You also pored through archival records of clubs, soaking up history to guide you forward. You filled in the gaps by speaking with faculty and staff and alums. You evaluated the options, decided what you wanted to revive and what you wanted to scrap.

And true to your nature as MIT students, you launched new stuff. You innovated and invented.

And you built communities, from FPOPs and orientation through 8.01, 18.02, your HASS classes, and your p-set groups.

You built communities in your dorms and in your sororities and fraternities.

You built communities through your sports, through your hobbies and through the arts.

You built communities all across campus.

And you learned that building communities is not always easy and quick. It takes effort, patience, and a willingness to listen to and learn from others.

But, in the end, it is so worth it because you’ve met and made friends with really interesting people. Some with similar backgrounds and others from very different backgrounds. And from that interesting and diverse group, you’ve identified your crew — the people with whom you’ve shared not only interests — but your dreams, your fears, your concerns, laughs, and tears. You’ve made real connections — connections that lead to a lifetime of friendship.

And over the past four years, right before our eyes, you’ve demonstrated the enduring value and power of higher education to change lives.

Throughout your time at MIT, you ideated, prototyped, and tested. You created new knowledge, waded through ambiguity, worked collaboratively, and, of course, you optimized.

Now, on your graduation day, we send you on your way with enormous pride and hope.

But at the same time, we are sending you out into the world at a very difficult and challenging time. It’s a time when we all are being asked to focus on traditions that we should honor and defend. It’s also a time calling on us to create new traditions, better suited to human thriving in this century.

It’s a time when the issues are big, the answers are complex, the stakes are high, and the paths are uncharted.

But, Class of 2025, you are prepared to face these daunting conditions. In the words of one of your classmates: MIT taught the Class of 2025 to have “confidence in your competence.”

You are ready to assess your environment, diagnose what is stale and what is broken, learn from history, apply your talents and skills, and create new knowledge.

You are ready to tackle the toughest of problems! You are ready to shape the future.   

And while you are doing so, I ask that you keep MIT’s values and mission at the center of your efforts: to be bold and imaginative in tackling these big problems and to do so with compassion and generosity.

Now, more than ever, we — meaning the world’s people — need you to lean in.

Once again, Congratulations Class of 2025!

“Class of 2025, are you ready?”

This was the question Hashim Sarkis, dean of the MIT School of Architecture and Planning, posed to the graduating class at the school’s Advanced Degree Ceremony at Kresge Auditorium on May 29. The response was enthusiastic applause and cheers from the 224 graduates from the departments of Architecture and Urban Studies and Planning, the Program in Media Arts and Sciences, and the Center for Real Estate.

Following his welcome to an audience filled with family and friends of the graduates, Sarkis introduced the day’s guest speaker, whom he cited as the “perfect fit for this class.” Recognizing the “international rainbow of graduates,” Sarkis welcomed Mary Robinson, former president of Ireland and head of the Mary Robinson Foundation — Climate Justice to the podium. Robinson, a lawyer by training, has had a wide-ranging career that began with elected positions in Ireland followed by leadership roles in global causes for justice, human rights, and climate change.

Robinson laced her remarks with personal anecdotes from her career, from with earning a master’s in law at nearby Harvard University in 1968 — a year of political unrest in the United States — to founding The Elders in 2007 with world leaders: former South African President Nelson Mandela, anti-apartheid and human rights activist Desmond Tutu, and former U.S. President Jimmy Carter.

She described an “early lesson” in recounting her efforts to reform the laws of contraception in Ireland at the beginning of her career in the Irish legislature. Previously, women were not prescribed birth control unless they were married and had irregular menstrual cycles certified by their physicians. Robinson received thousands of letters of condemnation and threats that she would destroy the country of Ireland if she would allow contraception to be more broadly available. The legislation introduced was successful despite the “hate mail” she received, which was so abhorrent that her fiancé at the time, now her husband, burned it. That experience taught her to stand firm to her values.

“If you really believe in something, you must be prepared to pay a price,” she told the graduates.

In closing, Robinson urged the class to put their “skills and talent to work to address the climate crisis,” a problem she said she came late to in her career.

“You have had the privilege of being here at the School of Architecture and Planning at MIT,” said Robinson. “When you leave here, find ways to lead.”

Below is the text of President Sally Kornbluth’s 2025 MIT Commencement remarks, as prepared for delivery today.

Good afternoon, everyone! Governor Healey. The members of the Class of 1975, in their incredibly fashionable red jackets! And of course, all the members of the Class of 2025 — and your devoted family and friends!

At MIT, it’s customary for the president to deliver a “charge” to the graduating class. And I’ll start by reflecting briefly on the world we make together here at MIT.

At MIT, we allow a lot of room for disagreement, whether the subject is scientific, personal, or political. The friction of disagreement is a very effective way to sharpen each other’s thinking. (If you don’t believe me, I’d urge you to attend a faculty meeting!)

But in this disconcerting time, as we prepare to send the Class of 2025 out into the world, I want to 
celebrate three fundamental things we do agree on — the rock-solid foundation of our shared work and understanding.

First, we believe in the beauty and power of the scientific method. Winston Churchill once observed that, “No one pretends that democracy is perfect or all-wise,” In fact, as he famously acknowledged, “Democracy is the worst form of government — except for all the other forms that have been tried.”

And you could say the same — with reverence! — for the scientific method. None of us would argue that it’s “perfect or all-wise.”

But the scientific method remains the single most reliable tool humans have ever devised to arrive at the truth about the physical world. It’s designed to root out error, protect us against our own biases and assumptions, and provide a systematic way to turn facts we cannot see at first into knowledge we can act on. 

It’s hard to imagine anything more useful than that.

Second, we believe in the beauty and power of fundamental scientific discovery – that incredibly intricate, maddening, heroic, intoxicating process of exploration and testing that somehow got stuck with the bland label “basic research.” 
 

We believe scientific discovery is deeply valuable and inspiring, in itself — and we know that it’s absolutely essential for driving innovation and delivering new tools, technologies, treatments, and cures.

And finally — from direct personal experience here at MIT — we all know that we’re sharper, more rigorous, more curious, more inventive and more likely to achieve breakthrough results when we work together with brilliant people, across a broad spectrum of backgrounds, perspectives, and viewpoints, from across the country and all around the world.

You don’t find the big ideas in an echo chamber! 

And I want to say something I’ve said repeatedly: MIT would not be MIT without our international students!

*              *

The beauty and power of the scientific method. The beauty and power of scientific discovery. And frankly, the beauty and power of the Institute’s incredible global community. 

For those of us associated with MIT, these three concepts may seem almost too obvious to require explanation, let alone celebration.

But we find ourselves in a bewildering time, a time when these concepts have never been more important — and have rarely been in such peril.

So now, I offer my charge to the members of the Class of 2025.

To today’s graduates:

I hope and believe that, in your time here, you’ve prepared yourselves very effectively for the next steps in your life and career. I wish you every success in that next step, and all that come after it.

But I must ask that each of you take on another job. A lifelong job. An urgent job.

I need you all to become ambassadors for the way we think and work and thrive at MIT.

Ambassadors for scientific thinking and scientific discovery! For thoughtful research of every kind, here — and at universities across the country! For the importance of research to the advancement of our nation — and our species! And ambassadors for the limitless possibilities when we understand, appreciate and magnify each other’s talent and potential, in a thriving global community.

This ambassadorship has no salary besides your sense of its crucial importance. But I hope you will accept the responsibility — because no one else can make the case more effectively. And these concepts are the indispensable foundation of everything else we aim to achieve.

*              *

There’s only one way to get through MIT.

The hard way.

Each of you has done that — and in the context of historic challenges. May all the strengths and insights that you’ve gained here serve you brilliantly on the road ahead. Thank you — and congratulations!

An energetic OneMIT Commencement ceremony today featured calls for MIT’s newest graduates to have a positive impact on society while upholding the Institute’s core values of open inquiry and productive innovation.

“Orient yourself not just toward the construction and acquisition of new tools, but to the needs of people,” said science communicator Hank Green, in the event’s keynote remarks. He urged MIT’s newest graduates to focus their work on the “everyday solvable problems of normal people,” even if it is not always the easiest or most obvious course of action.

“Because people are so complex and messy, some of you may be tempted to build around them and not for them,” Green continued. “But remember to ask yourself where value and meaning originate, where they come from.” He then provided one answer: “Value and meaning come from people.”

Green is a hugely popular content creator and YouTuber whose work often focuses on science and STEM issues, and who has built, with his brother, John, the educational media company Complexly. Their content, including the channels SciShow and CrashCourse, is widely used in schools and has tallied over 2 billion views. Green, a cancer survivor, is also writing a book explaining the biology of cancer.

The ceremony also featured remarks from MIT President Sally A. Kornbluth, who delivered the traditional “charge” to new graduates while reflecting on the values of MIT and the value it brings society.

“We believe scientific discovery is deeply valuable and inspiring in itself — and we know that it’s absolutely essential for driving innovation and delivering new tools, technologies, treatments, and cures,” she said.

Kornbluth challenged graduates to be “ambassadors” for the open-minded inquiry and collaborative work that marks everyday life at MIT.

“I need you all to become ambassadors for the way we think and work and thrive at MIT,” she said. “Ambassadors for scientific thinking and scientific discovery. For thoughtful research of every kind — here, and at universities across the country. For the importance of research to the advancement of our nation — and our species. And ambassadors for the limitless possibilities when we understand, appreciate and magnify each other’s talent and potential, in a thriving global community.”

Kornbluth also elaborated on the core elements of the work MIT has always pursued.

“At MIT, we allow a lot of room for disagreement, whether the subject is scientific, personal or political,” Kornbluth said. Still, she noted, “in this disconcerting time, as we prepare to send the Class of 2025 out into the world, I want to celebrate three fundamental things we do agree on — the rock-solid foundation of our shared work and understanding.”

The first of these, Kornbluth said, is that “we believe in the beauty and power of the scientific method. … It’s designed to root out error, protect us against our own biases and assumptions, and provide a systematic way to turn facts we cannot see at first into knowledge we can act on. It’s hard to imagine anything more useful than that.” Secondly, she said, in a similar vein, “we believe in the beauty and power of fundamental scientific discovery.”

A third element, Kornbluth observed, is that “we all know that we’re sharper, more rigorous, more curious, more inventive and more likely to achieve breakthrough results when we work together with brilliant people, across a broad spectrum of backgrounds, perspectives and viewpoints, from across the country and all around the world. You don’t find the big ideas in an echo chamber.”

Kornbluth added: “I want to say something I’ve said repeatedly: MIT would not be MIT without our international students.”

MIT’s Commencement celebrations are taking place this week, from May 28 through May 30. The OneMIT Commencement Ceremony is an Institute-wide event, held in MIT’s Killian Court and streamed online. MIT’s undergraduates, as well as advanced degree students in its five schools and the MIT Schwarzman College of Computing, also have additional, separate ceremonies in which graduates receive their degrees individually.The OneMIT event also featured remarks from Massachusetts Governor Maura Healey, who said she was “incredibly proud” of the graduates and the Institute itself.

“You stand for the qualities that make Massachusetts special: a passion for learning and discovery that is so powerful it changes the world,” Healey said. “Curing disease. Inventing technologies. Solving tough problems for communities, organizations, and people all around the globe. Making lives better and powering our economies. Thanks to you, Massachusetts is No. 1 for innovation and education.” She added: “MIT’s contributions to our knowledge economy — and our culture of discovery — are a pillar of Massachusetts’ national and global leadership.”

Speaking of the economic impact of MIT-linked businesses, Healey had an additional suggestion for the graduates: “Put your talents to work in Massachusetts, a place where you are valued, respected, and surrounded by incredibly talented, engaged innovators and investors. Make your discoveries here. Found your startups here. Scale your companies here.”

She even quipped, “We put forward some pretty good incentives through our economic development legislation and we’ll help you find a way to spend that. Just reach out to my economic development team.”

Green imparted general life advice as well.

“One of the problems you will solve is how to find joy in an imperfect world,” Green said in his Commencement address. “And you might struggle with not feeling productive, unless and until you accept that your own joy can be one of the things you produce.”

On another note, Green added, “Ideas do not belong in your head. They can’t help anyone in there. I sometimes see people become addicted to their good idea. … They can’t bring themselves to expose it to the imperfection of reality. Stop waiting. Get the ideas out. … You may fail, but while you fail, you will build new tools.”

Throughout his speech, Green emphasized the humanitarian qualities of MIT’s students. This past semester, after being named Commencement speaker, he sent the graduating class a survey that about half of the class responded to.

The survey included the question, “What gives you hope?” In his speech, Green said the many of the responses involved other people. Or, as he characterized it, “People who care. People who focus on improving life in their communities. People who are standing up for what they believe in. People who see big problems and have the determination to fix them.”

The OneMIT ceremony began with the annual alumni parade, this time featuring the undergraduate class of 1975, while the Killian Court Brass Ensemble, conducted by Kenneth Amis, played the processional entry music.

The Chaplain to the Institite, Thea Keith-Lucas, delivered the invocation, while the campus a capella group, the Chorallaries of MIT, sang “The Star Spangled Banner,” and later, the school song, “In praise of MIT,” as well as another Institute anthem, “Take Me Back to Tech.”

Despite many uncertainties facing higher education, the MIT students, families, friends, and community members present reveled in a festive moment, celebrating the achievements of the graduates. A total of 1,158 undergraduate and 2,593 graduate students received MIT diplomas this academic year.

“There’s only one way to get through MIT,” Kornbluth quipped. “The hard way.” 

Below is the text of Hank Green’s Commencement remarks as prepared for delivery on May 29.

I don’t really do imposter syndrome, that’s where you feel like you don’t belong. I have a superior syndrome called “Hahaha I fooled them again” syndrome where I know that I don’t belong, but I also am very pleased that I have once again cleverly convinced you that I do.

I, a man you might very well know as a tiktoker, a man who recently blind-ranked AI company logos by how much they look like buttholes, have snuck into giving MIT’s Commencement speech. And I can admit this because you can’t kick me off now, I’ve already started speaking! It would be weird if you stopped … but still, I’m going to try to do a good job.

Hello and thank you very much to everybody for welcoming me out, all the lovely people up here, the president, the governor, the alumni, Class of 75, and also of course, thank you especially to a class of extremely impressive charismatic and attractive students of the Massachusetts Institute of Technology graduating Class of 2025.

To express my thanks: The average human skeleton has more than 25,000 calories. More than half of your bones are in your hands and feet, and all together your skeleton contains enough oxygen atoms that, if you freed them, you could produce around 24 hours of breathable air.

Those were some of my best bone facts, and I assume that good bone facts are a totally normal way for humans to show gratitude.

I gave you my very best bone facts because I owe an extra debt of gratitude to you, the Class of 2025, because more than half of you filled out a survey I sent you! I assume you did it late at night while you should have been p-setting, whatever that is, but instead you did this.

And I have loved looking through your responses and learning a little bit about you, and a little bit from you.

One of the things asked you what the most MIT thing you did at MIT was, and this was my favorite section to read.

Some of it was definitely not meant for me to understand, like several of you counted up all the smoots on the Harvard Bridge.

Whatever that means … good work.

One of you was Tim the Beaver. Another tried to impress a date with train facts.

I see you. Same … but with bones.

A lot, and I mean a lot of you simply said the word “hack,” and the lack of specificity there, I have to say, does make me feel like whatever you did, the statute of limitations has not yet kicked in.

But by far the most common beginning of a sentence in this section was “I built…” You built robots and bridges and incubators and startups and Geiger counters and a remote-controlled shopping cart and a ukelele and an eight-foot-wide periodic table. Y’all built … a lot.

And that is something I found reassuring. We are going to need to do a lot of building.

I took a look at your shoes as you were coming, but it turns out I didn’t need to see them to know I wouldn’t want to be in them.

I think the only people jealous of you right now is the Class of 2026 because I’m sure things will be even more screwed up by the time they’re sitting where you are. But what a terribly messy time to be graduating from college. The attacks on speech, on science, on higher education, on trans rights, on the federal workforce, on the rule of law … they’re coming from inside the house.

Meanwhile, the world is getting hotter faster. And the sudden acceleration in the abilities of artificial intelligence, communication, and biotechnology promise huge opportunities, and massive disruption.

So, if I were you, I would want some advice! But as previously mentioned, I am a TikTok-er who will now forever be known as the first person to ever say the word “butthole” during an MIT commencement speech. So the advice — some of it — is going to come from you. I asked you, in my survey, what you would say to your classmates from a stage like the one I am now on. And here’s a selection.

One of your classmates wrote:

I always forget which Green brother is Hank and which is John!

There is no one definition of success. The idea you have in your head of what success is, it’s going to change, and you should let it.

Is one of your classmates 45 years old?

And here’s another 45-year-old hiding among you:

Open a Roth IRA.

Jeez! Did your dad fill out my survey for you? Seriously though, you should.

Here’s one of my real favorites:

Collaborate and help each other, be brave in reaching out, and be forgiving in your interactions.

Even if it probably won't work, try anyway.

Don’t start with the solution, start with the problem.

Now a lot of you might be thinking right now: Did he just make us write his Commencement speech for him? And the answer to that is, well, at least you know that Claude didn’t write it.

I’ve had a good time here focusing on the ludicrous aspects of my career, and I do want to emphasize its ludicriousness.

I’ve done TikTok dances to Elmo remixes, and I’ve also published two best-selling science fiction novels. I’ve written fart listicles, and I’ve interviewed presidents. I’ve made multiple videos about giraffe sex, and I’ve sold multiple companies. I helped build an educational media company that provides videos for free to everyone with an internet connection, and our content is used in most American schools.

And yes, that was the section I put in so your parents could feel better about me being here. I left it as long as I could.

I am good at having an idea I believe in and then just doing it, consequences be damned, and that has served me well, though it has not always been relaxing.

And I did that all on the uncertain and rapidly changing ground of online video and social media over the last 20 years. So perhaps I do have something to say to a class of graduates heading out into an uncertain and unstable world.

If I could attribute my success, whatever it is, to anything besides luck, it’s that I literally can’t stop believing that there is any better use of time than learning something new.

And curiosity doesn’t just expand the number of tools you have and how well you’re able to use them, it expands your understanding of the problem space.

And so maybe the advice is very simple. Just be curious about the world and you’ll have everything you need for the future and, look, it is almost that simple.

There’s a really important question I asked y’all in my survey that I haven’t mentioned yet. I asked, “What’s giving you hope?”

And though one of you wrote “Macallan 12,” most of you, in your response, talked entirely about people: my friends, my family, my peers, over and over.

People who care. People who focus on improving life in their communities. People who are standing up for what they believe in. People who see big problems and have the determination to fix them.

At a school like MIT, I imagine that the focus can definitely be on the building and less on the people. This is an institute of technology, not of humanities. But I read the humanity in your answers.

And this brings me back to the simplicity of curiosity leading you both toward understanding problems and acquiring new tools. Because your curiosity is not out of your control. You decide how you orient it, and that orientation is going to affect the entire rest of your life. It may be the single most important factor in your career.

And my guess is that it’s going to be really easy to be focused on the problem of just building ever more powerful tools. That’s exciting stuff and also it can be surprisingly uncomplicated. But even though the problem space is much bigger than just “build bigger tools,” it is surprisingly easy to simply never notice that.

The most powerful mechanisms that steer our focus are … I’m just going to say this … not always designed for our best interests, or the best interests of our world. Social content platforms are great at steering our curiosities and they are, often, designed to make us afraid, to keep us oriented toward impossible problems, or toward the hottest rifts in society.

Meanwhile, the capitalist impulse is very good at keeping us oriented toward the problems that can be most easily monetized, and that means an over-weighting toward the problems that the most powerful and wealthy people are interested in solving.

If we let ourselves be oriented only by those forces, guess what problems we will not pay any attention to. All of the everyday solvable problems of normal people.

I desperately hope that you remain curious about our world’s intensely diverse and massive problem space. Solveable problems! That are not being addressed because our world does not orient us toward them. If you can control your obsessions, you will not just be unstoppable, you will leave this world a much better place than you found it.

This is not about choosing between financial stability and your ideals. No. There is money to be made in these spaces. This is simply about who you include in your problem space, about what you choose to be curious about.

So with that in mind, here’s my advice, from my heart and from my experience.

First, don’t eat grass.

Second, more importantly, one of the problems you will solve is how to find joy in an imperfect world. And you might struggle with not feeling productive unless and until you accept that your own joy can be one of the things you produce.

Third, ideas do not belong in your head. They can’t help anyone in there. I sometimes see people become addicted to their good idea. They love it so much, they can’t bring themselves to expose it to the imperfection of reality. Stop waiting. Get the ideas out. You may fail, but while you fail, you will build new tools.

And fourth, because people are so complex and messy, some of you may be tempted to build around them and not for them. But remember to ask yourself where value and meaning come from, because they don’t come from banks or tech or cap tables. They come from people.

People things are the hardest work, but also often the most important work. Orient yourself not just toward the construction and acquisition of new tools, but to the needs of people, and that include you, it includes your friends and your family. I think we can sometimes feel so powerful and like the world is so big that throwing a birthday party or making a playlist for a friend can seem too insignificant when placed against the enormity of AI and climate change and the erosion of democracy. But those thoughts alienate you from the reality of human existence, from your place as a builder not just of tools, but of meaning. And that’s not just about impact and productivity and problem solving, it is about living a life.

Do. Not. Forget. how special and bizarre it is to get to live a human life. It took 3 billion years for the Earth to go from single-celled life forms to you. That’s more than a quarter of the life of the entire universe. Something very special and strange is happening on this planet and it is you.

The greatest thing you build in your life will be yourself, and trust me on this you are not done yet, I know I’m not. But what you will be building is not just a toolkit. You will be building a person, and you will be doing it for people.

When I asked you what you did at MIT, you said you built, but when I asked you what was giving you hope, you did not say “buildings” you said “people.” So, to the graduating Class of 2025, go forth, for yourself, for others, and for this beautiful, bizarre world.

Thank you.

The MIT Corporation — the Institute’s board of trustees — has elected 10 full-term members, who will serve three- or five-year terms, and three life members. Corporation Chair Mark P. Gorenberg ’76 announced the election results today.

The full-term members are: Wes Bush, Ruby R. Chandy, Hala Fadel, Jacques Frederic Kerrest, Michelle K. Lee, Bianca Lepe, Natalie M. Lorenz Anderson, Sebastian S. Man, Hyun-A C. Park, and Thomas Tull. The three life members are: Orit Gadiesh, Jeff Halis, and Alan Leventhal. Gorenberg was also re-elected as Corporation chair.

Stephen P. DeFalco ’83, SM ’88, the 2025-2026 president of the Association of Alumni and Alumnae of MIT, will also join the Corporation as an ex officio member. He succeeds Natalie Lorenz-Anderson ’84.

As of July 1, the Corporation will consist of 80 distinguished leaders in education, science, engineering, and industry. Of those, 24 are life members and eight are ex officio. An additional 31 individuals are life members emeritus.

The 10 term members are:

Wes Bush, former chair and chief executive officer, Northrop Grumman Corporation

Bush has worked in the aerospace and defense industry since starting at COMSAT Labs under MIT’s co-op program. After graduation, he first worked at The Aerospace Corporation, then became a systems engineer at TRW’s Space Park facility in 1987. Prior to Northrop Grumman’s acquisition of TRW in 2002, he led numerous space program activities, served as vice president of TRW Ventures, and was the president and chief executive officer of TRW’s U.K.-based Aeronautical Systems business. At Northrop Grumman, Bush served as the president of the company’s space technology sector, then as its chief financial officer. He became president of the company in 2006, served as chief executive officer from 2010 through 2018, and became chairman in 2011.

Ruby R. Chandy ’82, SM ’89, CEO, Luminas Advisory Services

With 20 years of public company board experience, Chandy currently serves on the boards of Dupont, Thermo Fisher Scientific, and Flowserve. She is on the advisory board of Pritzker Private Capital and serves on boards of its portfolio companies. She was formerly president of the industrial division and a corporate officer at Pall Corporation, which was acquired by Danaher Corporation. Prior to Pall, she served as chief marketing officer at Dow Chemical, Rohm and Haas, and Thermo Fisher Scientific. She has extensive general management experience at Dow Chemical, Thermo Fisher Scientific, Boston Scientific, and Millipore. Chandy also currently serves on the Board of the NACD Philadelphia Chapter, the Board of Trustees for Cristo Rey Philadelphia High School, and the MIT Sloan Executive Board.

Hala Fadel MBA ’01, managing partner, Eurazeo

Fadel is a member of the management committee of Eurazeo and leads the investment committee of the growth equity team. Prior to joining Eurazeo in 2022, she built Comgest’s inaugural private growth equity program. She also spent nearly 15 years at Comgest as a portfolio manager within the European growth equities team, leading investments in technology, as well as in health care and consumer goods. From 2014 to 2022, she served as co-founder and managing partner of Leap Ventures, an early-stage technology venture capital firm that invests in Europe and the Middle East. While there, Fadel led several early-stage tech investments in France, Sweden, and the U.K. She started her career as an investment banker in mergers and acquisitions at Merrill Lynch in London.

Jacques Frederic Kerrest MBA ’09, vice chair and co-founder, Okta; managing partner and founder, Windproof Partners; senior advisor to Blackstone

As Okta’s chief operating officer from 2009 to 2023, Kerrest was responsible for Okta’s day-to-day operations, drove Okta’s corporate priorities, accelerated innovation across the company, worked closely with customers, partners and prospects, and served as a key liaison with the investor community. He oversaw corporate strategy, corporate development, strategic partnerships, and Okta’s social impact arm, Okta for Good. Previously, he worked in sales and business development at Salesforce.com, and in venture capital at Hummer Winblad Venture Partners. Kerrest also served as the chair and co-founder of Herophilus, a neurotherapeutics drug development company acquired by Genentech Roche. He is the author of “Zero to IPO,” a guidebook to building startups.

Michelle K. Lee ’88, SM ’89, CEO and founder, Obsidian Strategies, Inc.

Prior to founding Obsidian Strategies, Lee was vice president of the Machine Learning Solutions Lab at Amazon Web Services. She also served as the U.S. under secretary of commerce for intellectual property and director of the U.S. Patent and Trademark Office from 2015 to 2017 and was the first woman to serve in this role in the country’s history. Before entering public service, she served as an executive for eight years at Google. Lee also held the appointment of the Herman Phleger Visiting Professor of Law at Stanford University from 2017 to 2018. She began her career as a computer scientist at the MIT Artificial Intelligence Laboratory and Hewlett-Packard Research Laboratories.

Bianca Lepe PhD ’24, data scientist, City of Boston

Lepe’s PhD research focused on computationally guided vaccine design for tuberculosis and the development of a surface functionalization platform for M. tuberculosis to study host-pathogen interactions. As a Graduate Student Union member and REFS conflict coach, she supported fellow researchers, helped resolve conflicts, and represented student concerns. She also served as a student leader on the Graduate Student Council and the Corporation Joint Advisory Committee, where she facilitated dialogue on critical issues and advocated for people-centered solutions. Lepe has professional experience in technology policy consulting, venture capital, and biotech strategy. She recently joined the City of Boston’s analytics team as a data scientist, where she collaborates on projects to improve the city’s decision-making and operations.

Natalie M. Lorenz Anderson ’84, chief operations officer and board director, 247Solar, Inc.

Before joining 247Solar, an MIT startup commercializing a modular, scalable thermal energy solution, Lorenz Anderson was a partner at Booz Allen Hamilton, where she was a senior vice president and subject matter expert in cybersecurity, privacy, risk management, IT, and advanced technologies in the defense, national security, and civilian agency domains. She has served on several advisory and corporate boards with MIT roots, including Gigavation, Embr Labs, and Lutron, and is a former board member and current advisory board member for Ocean Power Technologies. Lorenz Anderson has also been a limited partner of Safar Partners LLC and is a former board director and vice president of the Girl Scouts Nations Capital Board.

Sebastian S. Man ’79, SM ’80, chair and chief executive officer, Chung Mei International Holdings Limited

Since 1990, Sebastian S. Man has helmed Chung Mei International Holdings Limited, which was co-founded in 1963 by his family and is a leading manufacturer of domestic kitchen electrics and air treatment products for major international brands. He is affiliated with several trade organizations, including as honorary vice president of the Hong Kong Electrical Appliance Manufacturers Association and a board director of the Pacific Basin Economic Council. He is also a council member with the Better Hong Kong Foundation and a member of the Vision 2047 Foundation. Man has been an executive committee member of the International Chamber of Commerce and the Hong Kong China Business Council. He is also an executive committee member of the Young Presidents’ Organization Gold HK and the North Asia Chair of the Chief Executive Organization.

Hyun-A C. Park ’83, MCP ’85, president, Spy Pond Partners, LLC

Park started her career working for MIT professor Tunney Lee at the Massachusetts Division of Capital Planning and Operations, and then worked on the Central Artery (“Big Dig”) project. From there, she went to Cambridge Systematics, where she was in charge of a business line focused on transportation asset management. Park recently chaired the Technical Activities Council of the Transportation Research Board, where she led a group of chairs that oversaw more than 200 committees and 6,000 volunteers on research activities related to all modes of transportation and a wide range of transportation topics. She also served as co-chair of the Women’s Transportation Seminar’s Public Art Project that resulted in the installation of a new public art piece at Boston’s South Station.  

Thomas Tull, co-chair, TWG Global

In addition to his role at TWG Global, Tull founded and chairs the United States Innovative Technology Fund, and is the founder, chair, and CEO of the private holding company Tulco, LLC. Previously, he founded and served as CEO of the media company Legendary Entertainment, which produced films like “The Dark Knight” trilogy, “Inception,” and “Jurassic World.” Tull is part of the ownership groups of the Pittsburgh Steelers and the New York Yankees, and he is deeply committed to philanthropy and advancing innovative solutions to global challenges through the Tull Family Foundation. He serves as an advisor to the chief innovation and strategy officer at MIT, is a member of the MIT School of Engineering Dean’s Advisory Council, and recently served as a Visiting Innovation Scholar at MIT.

The three new life members are:

Orit Gadiesh, partner and chair emeritus, Bain and Company Inc.

Gadiesh joined Bain and Company in 1977 and served as chair from 1993 to 2025. She is currently based at the group’s London headquarters and remains active in client and advisory work in North America, Europe, and Asia. She has counseled top-level management in structuring and managing portfolios, developing and implementing global strategy, designing both cost reduction and growth programs, embedding technologies in organizations, and more. Gadiesh currently serves on the World Economic Forum board of trustees, the International Business Leaders Advisory Council to the Mayor of Shanghai, and the board of governors at Tel Aviv University, as well as on the advisory boards of the James Martin 21st Century School of Oxford University and the Peres Institute for Peace and Innovation.

Jeff Halis ’76, SM ’76, president and CEO, Tyndall Management, LLC

Halis founded Tyndall Management, an investment firm specializing in publicly traded securities, in 1991. Prior to that, he held positions in the finance and investment industry working for Citibank, Merrill Lynch, and Sabre Associates. He is a former director of several publicly traded companies, including Enstar USA, Inc., KinderCare Learning Centers, and PriceSmart. His civic involvement included his membership on the state of New York’s financial control board, the investment committee of the New York State Common Retirement Fund, and the Citizen’s Budget Commission. He has also been on the boards of WNET, CaringKind, and Bridge Over Troubled Waters.

Alan Leventhal, former U.S. ambassador to the Kingdom of Denmark

Prior to his appointment as a United States ambassador from 2022 to 2025, Leventhal was the chair and chief executive officer of Beacon Capital Partners. He previously served as president and chief executive officer of Beacon Properties Corporation, a publicly traded real estate investment company that merged with Equity Office Properties in 1997. He is the former chair of the board of the Damon Runyon Cancer Research Foundation and also served on the executive committee. Leventhal is a trustee emeritus of Boston University, where he served as chair from 2004 to 2008. He also served as a life trustee of Northwestern University and on the boards of the Friends of Post Office Square and the Norman B. Leventhal Map and Education Center at the Boston Public Library.

Hank Green, prolific content creator and YouTuber whose work has often focused on science and STEM-oriented topics, is delivering today’s OneMIT Commencement address. Green, along with his brother John, has built the educational media company Complexly, racking up over 2 billion views for the their content, including the channels SciShow and CrashCourse. MIT News talked with Green in advance of his commencement remarks.

Q: MIT’s president, Sally Kornbluth, often talks about the value of curiosity. How much of curiosity do you think is natural, or alternately, how do you keep cultivating your sense of curiosity?

A: There’s a line in my talk today, something like, if I could attribute my success to anything besides luck, it is always believing that there is no better use of a day than learning something new. And I don’t know where that came from. I feel like everybody is like that. I have an 8-year-old son and he’s like that. My wife texted me last night and said, “He wants to know what dark matter is.” Well, wouldn’t we all?

I don’t know exactly know how to cultivate that, but I do have strategies for orienting [toward] that. … The reality is that it’s very easy to orient my curiosity toward what would make me the most money or what makes me feel better than other people. I’m very aware of this as founder and host of SciShow, that people might watch because they want to feel superior to people who don’t know stuff. And that’s a motivation, and at least it’s oriented toward knowing more stuff, but it’s not the best motivation. I think one of the great powers people can have is being able to orient your curiosity around what your values are, and how you’d like to see the world change. And that’s something that I have worked a lot on.

Q: It seems like you’re not just learning about new things, but also, in the process, aren’t there a lot of new challenges in figuring out how to communicate things best?

A: Tons! I mean the thing about it is that the communications landscape changes very fast. Five years ago, TikTok wasn’t really a thing. When I heard about it, I thought, “You can’t do science communication in a minute. That’s impossible. All you can do is dance videos.” And then I saw people doing it and said, “Well, you can.”

I’m also working on a book-length science communication project right now. When I say book-length, it’s a book about the biology of cancer. And that process, it doesn’t end there, but for me that’s the largest, longest communication you can do.

[But alternately] my friend Charlie made one of the first science TikToks I saw. It’s a skit about how vaccines work, where one character was a vaccine and one was an immune cell. That was probably 30 seconds long and it’s probably better than any way I would have communicated about vaccines in the midst of the Covid epidemic on the new platform, pre-bunking fear about vaccines from the very beginning, very simply explaining what they are in a way that’s very accessible and not going to turn anybody off.

Q: What are you talking about in your remarks today?

A: Yeah, I mean we are in a super-weird moment with regard to the amount of power humanity has. We’ve been in moments like this before, where the amount of power at our fingertips increases exponentially very quickly. The nuclear age is the big one in terms of the speed of that change. But it feels like biotechnology and AI and communications are all adding up to being a really big deal.

The thing I kept coming back to was — I didn’t put this in the talk, but it inspired the talk: Okay, so we had a period of time where humans powered the world through muscle. And now human muscle is not the [most] important part of how we build. Intelligence and dexterity are important, but in terms of calories expended, [that’s done] by machines. If we end up in a world where that [also] becomes more the case for intelligence, what do we still have a monopoly on? A lot of people would still answer that question with “Nothing,” I guess.

I think that’s really wrong. I think we’ll still have a near-monopoly on meaning, and what we mean to each other. So, what I wanted to get at is, all the stuff that we do, all the things that we build, at the root, the base, we do it for people in some way. It might be a playlist for your friend, or the Human Genome Project, but all of that, we’re doing for people. And so keeping [ourselves] oriented toward people, and not building around them as an obstacle but building for them, is the thing I’ve wanted to be focused on.