AI and the Future of the University

I'm Rufus Griscom, and this is The Next Big Idea. Today, the future of the American university in the age of AI. Is going to college still a good idea? Elon Musk says college is basically for fun and to prove you can do your chores, but they're not for learning. Entrepreneur and investor Peter Thiel, meanwhile, has called the college diploma.

a dunce cap in disguise. And he's putting his money where his mouth is, writing six-figure checks to more than 300 young people who agreed to skip school. Ben Horowitz, the venture capitalist, has called college a scam. Jeff Bezos, meanwhile, has made the point that the famous college dropouts we think of, Bill Gates, Steve Jobs, Mark Zuckerberg, are exceptions to the rule. Bezos said, I started Amazon when I was 30, not when I was 20.

And I think that extra 10 years of experience actually improved the odds that Amazon would succeed. So that would always be my advice, he said. I finished college, and I think it's been helpful to me. This is not just some kind of theoretical debate. It's personal. About 15 minutes ago, I processed a payment for my 20-year-old son's spring semester of college. Always painful.

My 17-year-old son, meanwhile, is in the middle of his college application process. And I'm not alone in this. The question of the future of the American university, not to mention more broadly, how our education system needs to evolve. touches most of us. And of course, it's a precursor to a larger question. What world are we preparing our children for? And how best can we prepare them?

What jobs will we have in a few years, not to mention a decade from now? And is the purpose of college to prepare us for work or for life more broadly?

I recently had the pleasure of discussing these questions with four people unusually well-qualified to answer them. Nabiha Syed is the executive director of the Mozilla Foundation. She leads the global charge for trustworthy AI, deploying their multi-hundred million dollar budget to ensure that the AI systems powering future education serve humanity, not just corporate profit.

For all the people who are like, college is dead. I'm going to pay these young people money to leave and come in this other building and build a thing. Sounds a hell of a lot like a college to me. In fact, they're paying them, and that sounds like a scholarship to me. Nicholas Dirks, president and CEO of the New York Academy of Sciences, is creating programs to produce the ethical, interdisciplinary leaders with expertise in both tech and humanities that the future university demands.

The real challenge then is going to be how do we simultaneously adjust to this new technological future about which we'll be talking, but also preserve access to quality higher education. Julie Samuels, president and CEO of TechNYC, is mobilizing New York's entire tech ecosystem to arm students from every major, not just engineering.

with the AI skills the modern economy requires. These inflection points, when the technology changes rapidly, it is so important that we do the work to ensure that we try our hardest not to leave people behind. And Matthew Johnson Robertson, is the dean of the brand new College of Connected Computing at Vanderbilt. This is the university's first new college in 40 years, with a mission to dismantle silos and inject AI into every single discipline from law to the liberal arts.

And I'm hoping that, you know, higher ed thinks about how it can use AI as an accelerator, as a way of enhancing the student experience, of doing more. We had this conversation live in Vanderbilt's new campus in the Chelsea neighborhood in New York City. in an expansive room with vaulted ceilings, which was the refectory of the General Theological Seminary. I could hear as we talked the echoes of God's new and old.

Coca-Cola. For the big. For the small. The short. And the tall. Peacemakers. Risk takers. For the optimists. Pessimists. For long distance love. For introverts and extroverts. The thinkers and the doers. For old friends and new. Coca-Cola. For everyone. Pick up some Coca-Cola at a store near you. This episode is brought to you by White Claw Surge. Nice choice hitting up this podcast. No surprises. You're all about diving into tastes everyone in the room can enjoy.

Just like White Claw Surge. It's for celebrating those moments when connections have been made and the night's just begun. With bold flavors and 8% alcohol by volume, unleash the night. Unleash White Claw Surge. Please drink responsibly. Hard seltzer with flavors. 8% alcohol by volume. White Claw Seltzer Works. Chicago, Illinois.

Thank you. Wow, what a panel, what a room, what a topic. It's great to be here with you all. Vinod Khosla. The Sun Microsystems co-founder and venture capitalist said recently, college degrees are dead. Sam Altman, the CEO of OpenAI, was asked recently if his child, who was just born this year, would go to college. He said, probably not. Are the rumors of the death of the American university...

exaggerated. Let's start with Nabiha. People gathering together to learn from experts is not going anywhere. That shape of it is not going anywhere. Who will the experts be? Where do they gather? Who the people are? That's malleable. And that has changed in universities throughout the world. And I say this, Mozilla Foundation has been funding responsible technology education at...

57 different institutions on three continents for the last decade. And so that act, it's going to look a little bit different. It'll evolve. But that core of learning, not changing. The thing is that like a college isn't just about learning. There's other things that the college does. So if you think about the other functions of a university, class mobility between getting jobs and building networks, the future of creativity and confidence in your ability to learn.

and then a site of knowledge production, like deep research and knowledge production. The shifting economics... of getting a college degree affects those three things really significantly not just like will people get together to learn because it's worth noting for all the people who are like college is dead i'm going to pay these young people money to leave

and come in this other building and build a thing, sounds a hell of a lot like a college to me. In fact, they're paying them, and that sounds like a scholarship to me. Well, Nick, you have a lot of experience with the evolution of the American University. You've been a leader at Columbia, at Berkeley, now the New York Academy of Sciences.

You've seen the universities face any number of challenges and opportunities, and AI is the latest of those. What's your perspective? So first of all, you started with... Sam Altman among others making predictions about the future and of course if you want to follow Sam Altman you can put 10 billion dollars into open AI which still doesn't have a revenue model and see how that works out but

The truth is that very, very wise people have been predicting the end of college and the end of the university for decades. Peter Drucker did 35 years ago. Famously, Clayton Christensen, I think in 2005, said that in 15 years, half of the current degree-granting colleges and universities in the United States would have folded. And, of course...

We hear this all the time from the Silicon Valley that college is over. It's really a form that cannot survive, especially now with LLMs and AI. The truth is also, though, that this is a very, very challenging time for colleges and universities. And I think AI added on to all the other kinds of things, the growth of indebtedness after 2011, the rise of distrust in universities.

Clearly, the recent political assaults by the administration on research and even the effects of the pandemic leave us in a position, I think, where the top tier of universities, the Vanderbilts, the Berkleys and so on. will survive, but a lot of them won't. And I think the real challenge then is going to be how do we simultaneously adjust to this new technological future about which we'll be talking.

but also preserve access to quality higher education. Julie, as CEO of TechNYC, you've been actively helping schools adapt to the changing tech landscape. through initiatives like Decoded Futures and Future Ready. In your opinion, what do you see happening right now with schools and AI and where are we headed? So it's all over the place. To answer your question, it is all over the place. You've got school districts, you've got...

private schools, public schools, and, you know, there is no single playbook right now. It's really, really messy. We in New York City partner a lot with the public schools, you know, helping think through curriculum. We have much more success partnering with the schools so that students... have access to meet people who work in tech, to learn about tech companies, to see how the products work in the real world because they will graduate into a world where...

AI tools are prevalent. Whether you want them to or not, it doesn't matter. The other end of the spectrum A project that we've worked on that I'm almost the most proud of of any project is something called Empire AI, which is something we've done in partnership with the governor and her team and a number of New York state institutions, SUNY, CUNY, Columbia, Cornell. NYU, RPI, et cetera, et cetera, we're building a state-of-the-art computer in Buffalo, and it will rival.

the private sector companies. There is no other academic computer like it in the world and the academic partners, the researchers are the only people who have access to that. machine and to that compute. And it is in N of 1, and New York State is the only place doing it, and it's called Empire AI. is already up and running, it is already transformative, and we are just a fraction of the way in. So we also need to think about not just our curriculum, but how we are ensuring that the academy...

doesn't get left behind when the private sector is just like chasing the profit, whether or not that profit exists, as you said. And Matt.

You are, in some respects, the answer to all this, the new both challenges and opportunities that AI and the convergence of AI and education present. You are the dean of the new College of Connected Computing at Vanderbilt.

What is it that you're building and what do you see as the risks and opportunities that you face? Yeah, we're trying to build something that I think is going to be really, really... exciting both for students but for researchers and hopefully for the larger community the idea is that you know there's so much research and thought and work going into ai but you know

There hasn't been a ton of that invested in thinking about how AI is going to transform other fields and thinking about it with a much broader lens. And so this idea of computing for all. So how do we think about what this technology means to everyone? Because I think for so long it's been siloed and we think about computing.

scientists and in many cases the for-profit sector really dominating the way that this technology affects all of our lives and I think the study of that the research of that and thinking about how many fields are not gonna be challenged by this but hopefully transformed, improved, and to some degree advanced. I think that to me is what excites me about the College of Connected Computing.

And I'm hoping that higher ed thinks about how it can use AI as an accelerator, as a way of enhancing the student experience, of doing more. And I think this dichotomy of we should get rid of college because we can do it better some other way, I think fails to see the iterative nature of what we've been doing in colleges and universities for hundreds of years, which is to adapt, to reflect.

what we see in society and to figure out how to better improve and deliver what I think is still something really valuable, which is an entity dedicated to the creation of knowledge, the dissemination of knowledge, and not for simply profit. And I mean, I'm sure we could turn...

everything in our lives over to for-profit companies but that seems like it'd be a real bummer and like I kind of like the idea that not-for-profits people that are dedicated to some other mission are out here teaching our kids but you know that's just I guess my two cents.

I should say, by the way, that I was intentionally being provocative by opening with statements from Silicon Valley heavyweights saying that the universities is over. Their children, I've noticed, are all attending universities. Can I also say? Yeah.

Only in Silicon Valley do they say those things. Exactly. They do not say those things on the East Coast. Right. True that. True that. But in our conversation, Matt, you've said that you see this as a time that's both challenging and exciting for students today. Why do you think it's difficult to be a student today and why is it exciting?

Two things. This isn't a time of immense change. So I don't want to underplay that. And I think students are in a pretty vulnerable position because they are embarking on, in many cases, a kind of multi-year journey, taking themselves out of the workforce to some degree out of society.

kind of learn and then hopefully go and do something new with that knowledge. And I think it's really hard if you're tracking a moving target to figure out what's going to happen to you on the other side of this. I think the combination of that and, you know, the sort of economic impacts of AI, it fully realized. could result in less jobs for people really low on the ladder of progression. And so I think those are real concerns. And so when I meet with students, one of the things that they...

continually say to me is their concern that they're going to graduate out into a job market for which there's no job for them. And I think if we want to fulfill not the only promise, but one of the promises of what we're trying to do with the college education, we have to think about how do we prepare students.

for graduating out into an economy that may look somewhat different or very different depending on how true these predictions end up being. Shifting to the student's perspective, the experience that students are having today.

Julie, you told me something about what your kids did recently. before a trip to the Metropolitan Museum of Art. Can you share that story and maybe what its implications might be for us? Yeah, so I've got three kids. We live on the west side, but we live by the museum. They've been to the Met a million times. It's not their favorite place, usually. But my... Ten-year-old has been really into the Percy Jackson book, so he's getting very interested in Greek and Roman stories. And the other day...

they went to the Met. I stayed home with one of the kids. My husband took the two older kids to the Met and they had a great time and they came home. And my 10 year old's like, you're never going to believe what I did. And he said, I went on GPT and I asked.

GPT to create a scavenger hunt. They went to see the Greek and Roman section of the Met and he created a scavenger hunt and they printed it out and he had checked everything and it was like their best experience in the museum they've ever had and again they've been going to this museum their entire lives and it was awesome and it made the museum come alive it made the books he's reading come alive and by the way he did it

all on his own. I did not know until he got home that it happened. And it just blew my mind. It was totally amazing. And it was like, oh, right, this is the potential. This is the opportunity. And my kids got resources, but... The truth is you don't need a lot of resources to do what he did. So to put those tools in the hands of kids all over New York City, all over the world really is an amazing thing to think about.

And when you're working with schools, do you think that the teachers have as much to learn from the kids when it comes to AI and how to integrate it with the school systems?

Even my own children, they've got some teachers who think it's interesting and cool and some who don't, and it's like any new technology in that way. But in these moments, these inflection points, when the technology changes rapidly, it is so important that we do the work to ensure that we... try our hardest not to leave people behind and you know my son has two parents who work in tech and so that was like really easy and accessible for him but he shouldn't

You know, like as a society, we need to make sure that it's not just because like Dylan's got two parents in tech that he's going to be so much better at this than everyone else. So we spend a lot of time thinking about the really important equity questions. And these inflection points are when they really become just.

so much more crucial. Yep. Nabiha, you mentioned to me that many of your friends in the city are moving their kids to alpha schools. Could you share what alpha schools are and what they might say about where education's heading? Yeah, alpha schools are very particular philosophy that you could condense the act of learning into a two-hour AI-mediated interface, sort of gamified, and the rest of the time is used for project-based learning.

If you think that learning is basically like shoving facts into a duck's mouth, preparing it to be foie gras, sure. Like, fine. That seems fine. But that is not how I see the act of learning. I think it's like a very narrow use case of AI. And to like Julie's point, I think. Articulating the really interesting use cases of how AI can unlock learning beyond just like I can shove more facts into this little kid faster is like one part of that universe is like grim.

and sad. And there's a great wired article from a couple of weeks ago on alpha schools in Texas and how many parents felt like truly like their kids were learning facts, but not learning to love learning, not learning how to

be lifelong learners and how to teach themselves. And so it feels like it creates a very strange dependency on the technology. Then take this entirely different universe of like, how are people using whether it's a... off the shelf, like open AI LLM or something they build themselves to.

create flashcards or gamify learning, or if they have an educator who doesn't speak the same language as them or it's their second language, how you can bridge those divides. There's an entire universe of deeply pro-social ways to use it.

technology to enable learning. And then there's a sort of like stupid version, which is like, eat some facts, have a nice day. But what about sort of push back or take the alpha school side of the argument? I had a conversation with Sal Khan, founder of Khan Academy. He's of the view that if you look at the history of learning, that one-on-one tutoring has been shown to be incredibly effective. And the problem we have in a typical classroom is there's some kids who are...

and others who are behind in your teaching to the middle. There certainly is something compelling, I would think, about the idea that we will all, that all students at some point in the coming years will have their own AI tutor. Maybe we will all have our own AI tutor.

that knows exactly what we know and what we don't know, knows how to fill in the gaps, knows what our interests are and how to use metaphors and engage us in a way that's compelling and can report back to the teacher or at the alpha schools, they call them guides. where we are, where we need help. Do you not see this as where we're headed?

do, but I think the human component is not one that we can exclude. Like I was fortunate to have like one law school experience in the United States. It was like sort of a normal classroom. And then my second law degree was a tutorial model. And so I deeply understand the difference between. being taught something where you have a person who's invested in learning and you learning in a particular way.

The potential for AI to understand what you know, to test you from different angles, to say, well, have you thought about it this way or this way or this way is useful. But I think my critique is in isolation, that's not sufficient, right? So, and part of the critique of alpha schools is that the guides are often not deeply pedagogically trained. These are not people who are like skilled with saying there's a gap here.

Or are you connecting these two different concepts in a way that's meaningful? Are you seeing the interdisciplinary nature of this? That's not what's happening in those interactions. I think we leave something very critical on the table by just saying like, memorize as many facts as possible. Matt, I'm curious about your take on this. Do you foresee AI agents that would have a direct relationship with each student and perhaps be talking with either teachers and or the teacher's AIs?

I do think that seems really exciting. When I think about the challenge of alpha schools or maybe the opportunity of them, the thing that jumps to mind is, man, somebody should think about a way of studying whether that works or not, and they should...

maybe design some experiments and do some research and try to look at the prior literature. And then it sounds like a thing we do, but I can't think of the word of it. Maybe it's universities, I don't know. But to my mind, I think that this is the early days of what I think a technology that has a lot of promise.

offers. And if you look at the early days of the internet, similar things in thinking about the way we communicate knowledge with each other. I think one of the things that I'm excited about is that there's clearly immense opportunity.

ability to address some of the resource constraints that education has traditionally had because it's so driven by human labor that really you are constrained in the number of kids you can teach, the degree to which you can do that. And so I think I'm excited by the idea.

that If we figure out the right way of using these tools and we think about understanding what our goals are and how we want to deploy this, that there are huge opportunities for customized learning agents to really accelerate and enhance what we do in traditional classroom learning.

And maybe even think about new models that we haven't really dealt with yet. But I will say from the EdTech lens, over and over again, we get told that everything is going to change. And the traditional model of how we do this is all going to go out the window.

And none of those things have come to pass. I think a lot of the technologies we saw from MOOCs have been really useful and actually been very effective and helpful, but it hasn't radically changed the way that I practice in my classroom. And so... And that's not because I don't want to use those tools. I think there still are gaps. And so I think I'm excited by this technology, but I think I'm more excited by understanding where the best way and how the best deployment of it might be.

Running a business online? Instantly build trust with your customers with a domain and matching professional email from GoDaddy so people know you mean business. And there's never been a better time. Visit GoDaddy.com slash GD now and you can get a domain and professional email plan for just 99 cents a month for one year. That's a little price for a lot of credibility. Choose the domain that's right for you and a professional email to match.

Go to godaddy.com slash gdnow and look legit. That's godaddy.com slash gdnow. New customer purchases only. Products auto-renew separately. See terms on site. Have you ever called a loved one, gotten their voicemail, and thought, yes, that worked out perfectly? You're getting credit for a genuine act of thoughtfulness. You were thinking of them trying to connect, and yet the whole thing only took 30 seconds, which was all you had.

Well, that's the feeling I'm getting this season, giving aura frames to both of my siblings, my parents, and a few friends. How can one gift be perfect for all these different people I love? Because you can personalize each aura frame after a chips. You can deliver photos from anywhere, anytime, right from your phone. The opportunities for humor and personalization are endless.

For a limited time, save on the perfect gift by visiting AuraFrames.com to get $35 off Aura's best-selling Carver Matte Frames, named number one by Wirecutter, by using promo code IDEA, I-D-E-A. at checkout. That's A-U-R-A frames.com promo code idea. This deal is exclusive to listeners and frames sell out fast. So order yours now to get it in time for the holidays. Support the show by mentioning us at checkout.

Terms and conditions apply. When we talk about AI in education, the topic that tends to come up is cheating. And Nick, I gather that little blue books into which children or students, I should say, write essays is coming back into fashion. Do you have some experience with this?

Well, I have indirect experience. These days I'm not teaching undergraduates. I still work with graduate students, but I do have a wife who is a professor at Berkeley, and she has revived, and I don't think she's the only one who has done so. the blue book. And some of you may remember, those are the kind of stitched blue books with white ruled paper that you used to get for an exam in class. And all of a sudden, many professors are beginning to say, come into class.

write an essay on the following subject so that we know it's not going to be produced by ChatGPT and the student has to write on the spot, figure out how to make an argument, how to... progress paragraphs, and of course, show that they've not only read, but thought about the homework. Now, you know, the question of plagiarism has been a perdurant one. We have programs that can detect it.

Every paper that is written by ChatGPT is new, and some people say they have programs that can deal with these things, and we yet haven't really seen that work. But I think the challenge here and the larger question that goes to the question of the... of what and how AI is going to start changing the student experience is, are we going to find that our capacity to teach students how to write essays?

on their own, that is to say, think through a problem and really develop an argument, or for that matter, to read the books on which and about which they would write these essays. without having ChatGPT do the summaries or do the work for them or somehow or another intermediate the process of long exposure to text. that education used to consist of. And if you recall, in the 1970s, between 1972 and 1976, when the calculator was introduced, it took four years to get rid of the slide roll.

at that time were deeply worried, was this going to lead to the end of being able to teach mathematics? Well, it didn't. It didn't change the way in which certain kinds of tables were memorized. But the point... is of course that we are in early days. We don't know yet what the effects are going to be, and we have to try to not only test these new models, but figure out what is fundamental to education and how to preserve those kinds of things.

It seems to me, though, that what we've considered to be math skills has changed because of the calculator. It used to be that we were very impressed when people were able to conduct large calculations in their heads. We're no longer impressed by that because it's not useful. And I wonder if we won't find that.

The whole idea that a piece of writing is an authentic indication of how someone else thinks, it strikes me I'm going to make what might be a contrarian argument that I think that's obsolescent. I think we're reaching a point right now where... If you really want to know how somebody thinks, you need to talk to them. We're arguably at a point at which...

All writing is mediated by AI or might be mediated by AI. And so maybe what teachers should be doing in schools is return to focus on oratory. We're moving towards a more oral culture, arguably.

And this would please Socrates, because Socrates, of course, was very concerned that when writing emerged, it was going to cause students to no longer have memories and have command of facts. So is there an argument here that we're moving towards a more... rural culture and that our criteria for success for students should be more based on their ability to do what we're doing right now, which is to talk to each other.

I think this gets to a great debate in education that exists at the K-12 level and at the post-secondary, which is the tension between are we educating for education in some intellectual sake versus are we educating for jobs? And there is no one... answer and there's no one single answer for any single student probably but it is a debate I live I spend a lot of time working in kind of the political and civic landscape of New York City and this is a raging

debate that we have. Because, you know, we have the biggest school system in the world, I believe, New York City public schools. And how do you gauge success? Is it that kids who graduate have jobs? And if so, what are the skills? Are they oratory? What are the skills? You know, 10 years ago, we all talked about computer science education. Now we talk about computational thinking. And it keeps kind of evolving. And I don't know how you weigh those two things. I hope my children...

are able to go to school because they love learning. But I don't know. Those answers are really complicated. Let's talk about how AI can empower teachers. We have teachers among us here. How is AI changing?

teach, how we assess students, how we design classes, how we engage classrooms. Let's open this up. What do you think, Matt? Yeah, I'll jump in and say, you know, one of the things that's so interesting is that I think computer science education is a real bellwether for what may happen in other fields. So we have to radically reconsider the way we teach computer science. We usually teach people to program, and it's becoming...

rapidly apparent that these coding models are sufficient to generate not only simple programs, but really all of what we would teach in an undergraduate curriculum, probably graduate curriculum as well. And I think what's interesting about that is not that... that scares me, but it really requires us to rethink how we're going to go about teaching

someone how to program and what that even means. And so if the ultimate goal is to teach someone to be able to create computer programs that can do interesting things that we care about, teaching them to do really toy programs, which is what we would traditionally do, probably doesn't make sense. So what excites me from an educational standpoint... point is I can now ask people to create things of

almost sort of enterprise scale in the course of a class where before I had to get them to create a tic-tac-toe program. And so I think what that enables is ultimately us to get much closer to what the dream of computer science was, which is to allow people... to really

generate things that they're excited about or passionate about or whatever. And I think that to me suggests that we could see this in other disciplines as well. And there's so much focus on coding because of the economic benefits for massive companies. But I think ultimately... those same tools in terms of essay writing, in terms of even oral communication, I can produce programs that will...

orally communicate with you better than I can and be more snappy and more responsive and do all of those things. So I'm not even sure that the saving grace will be our ability to talk. I think it really does come down to what the purpose of education is. And I think we have to sort of reckon with that in the face of these sort of new technologies.

I think there's two frontiers here that are actually legitimately very exciting. One is sort of the creativity of pedagogy, like what experimentation is possible. And the second is to give really robust feedback, which is otherwise very difficult to do with time. So on the first one. I did a little experiment. So I teach First Amendment adjunct at Columbia and I had the.

pleasure of teaching that during COVID lockdown. And it was not fun to teach First Amendment during Zoom school. And so I had decided that I was going to turn each case like. First Amendment case into a choose your own adventure. Like, would you cross this line and have trespass? Would you do this? Would you do that? It took me weeks to design this game. And I was very proud of myself in 2020 for doing that.

As an experiment, I asked Claude if it could just make a game for me on like six different cases. And it did it in four minutes. And, you know, would I use each of those? Maybe, maybe not. But it gave me something to start with that would allow me to experiment with a totally different format of teaching something as, you know, potentially dry as.

First Amendment jurisprudence. And so that ability to play with different modalities, to experiment, to learn, which you could do the old-fashioned way, but would take a really long time, which especially adjuncts don't often have a lot of.

feels very... incredible the other is feedback right like you want to give people like five pages of feedback on the essay they've given you but like time is real and that's hard but you could be using a model to help say well okay i'm going to give you feedback on this

essay and it has this, this, and this structural weakness. Here's the gap analysis. Here's something you didn't cite. There's a lot of potential there to equip teachers who know what good education looks like and is using a tool to further their own skill set. in their own control versus replacing them.

You know, I'm really interested in this question of how Matt was referring to changes in how we think about teaching computer science. I have a 17-year-old son, and I've been, of course, encouraging him to take coding classes. He's taken a couple. But he says, Dad, you don't seem to understand.

We're not going to need to know how to code. We need to know how to, you know, I'm interested in economics and psychology and data science and computational thinking more so than writing code. That's obsolescent. You're not with the program here. And I will say... I find myself feeling kind of delighted that the language with which we speak to computers appears to be English.

now, which for those of us who are humanities majors, feels kind of like a long time coming. Is there reason to believe we've seen in the last couple decades a swing towards more and more focus on the STEM subjects? I'm interested to know if you all think that that's going to shift, that we're going to see a swing back towards humanities.

You know, I'm at the New York Academy of Sciences, and I'm supposed to say that STEM is the most important thing to study. But the truth is that there is an interesting swing back to the humanities. It's a very challenging moment for the humanities in many ways, not least because of... the job market, and because of perceptions of what an English major can actually do after they graduate, can they only deconstruct.

The refrain I kept hearing when I was at Berkeley from people who were hiring at Google and elsewhere was they wanted people who could think, who could think critically, who could... understand what it was to do research about new areas and indeed to have a kind of broad capacious sense of how you think differently in a world in which obviously

Google among others is always trying to find a new way forward and Humanities had begun to surface again as a critical resource in the heart of the Silicon Valley leave alone

in all the other kinds of fields, I think, in which, you know, we're going to rethink what it means to know how to code or to have a certain level of coding capacity. We're going to rethink how science works when you can start doing, as Demis Hassabis has said, decades worth of scientific research in the course of just a month or so and indeed when you can access an entire archive whether you're a historian or a

art historian or whatever in the matter of a few minutes where you upload a file. So we're still at the cusp of really understanding this, but I think the... the opportunity is that we can both reconceive the way in which interdisciplinary work can be done, both teaching and research, and come up with some new ideas.

about how to make the university a much more interconnected place, which is feeding it to you, Matt. That's the thing that got me so excited about the idea of what we're trying to build with the college. Like if you ask me where I see the most interesting next 30 years. It is to think about how these disciplines for which, you know, there was really a lack of focused effort on trying to deploy technology in and thinking about how technology could advance. I think...

A focus on that now with access to this technology that is much more general purpose than I think we've had before, I think does two things. It takes it out of the hands of experts. So you don't need to be a computer programmer to make a computer program that solves your problem and your discipline.

And I think that's exciting. That speaks to the creativity we were talking about earlier, you know, people bringing their own lens, their own creativity to problems they want to solve. And ultimately, I think this idea that we want to have everyone be a programmer frustrated me because I want everyone to do what their passion is. And then I think programming is an amazing tool to accomplish that well, efficiently, quickly, whatever.

But it would be such a boring world if everybody only cared about programming for programming's sake. We need the humanities. We need all these other areas. We need science. We need all these things because the human experience is so much more diverse than programming. And ultimately...

if we build artificial intelligence that only cares about computers, it really doesn't leave much space for us as people. And so I worry deeply that if we don't figure out a way of connecting this with other fields, we'll be left in a space where we're increasingly squeezed out.

And of course, perhaps from the perspective of parents in the room, the latent question is, what world are we preparing students for? What skills will actually be necessary in five to 10 years? Bill Gates has called this a confounding paradox. It's becoming easier to learn, but it's not clear what we need to learn if AI will be able to do everything. What skills do you all think we'll need?

I spend a lot of time in this space also because literally every politician I know asks every day for these answers and the truth is no one knows. I just want to say that, Matt, what you just said really resonated with me. It reminds me of this time in the late 90s. I know some of us were playing around on the internet then, and it felt...

The vastness felt amazing, and it felt so creative. And there were these corners of the internet, and if you knew where to find them, there were early blogs, and there was this website I went to that had these short stories, and you went, and it was like, it was magic.

And I'm reminded sometimes when we talk about what do the jobs look like, it can feel kind of bleak, honestly. But then you talk about the creativity, which is something that universities uniquely... can foster and that makes me feel amazing because from that creativity There will be jobs and there will be progress. That is the human experience. That will not change with AI. I hope that AI will only make it more exponential. Let's just give our panelists a huge applause.

Before the evening ended, a member of the audience asked a question about labor markets and how they might change in the age of AI. I responded this way. Based on all the evidence, it seems to me that... we are seeing an exponential increase in technological capability and not a linear increase. And that as much as we talk about historical technologies that have existed before, this is unprecedented at the pace at which it's changing.

What's happening in other convergent fields like robotics, what we're missing for humanoid robots is a level of dexterity that can be solved by... in theory, next generations of AI. So I think people can disagree about how long it will take, whether it will take, some would say five years, some say 10 years, some would say 20 years, some might say 30 years. But there is a future that is visible from here where all human...

labor is replicated by robots. And that possibly, to try to land on an optimistic note, lands us in a place where the liberal arts education returns to what it was arguably originally created for because the liberal arts education was originally

for the arts befitting a free person or for men and women of leisure could think about all the great ideas. And so the hopeful note I would like to land us on is that we will... that we or our children or grandchildren will have the great benefit of being men and women of leisure and that the liberal arts education will be all the more...

attractive and desirable. I would like to go back to school. I would like to go back to a college campus so that we can learn for the sake of learning, opposed to for the servile arts, which were for the purpose of commerce. I, for one, think there's never been a more fascinating time to be alive, and there have never been more ways to learn. College is a great privilege, as it always has been.

And I think I'm inclined to agree with Mark Cuban, the entrepreneur and Shark Tank host who said, there's going to be a greater demand in 10 years for liberal arts majors than there were for programming majors and maybe even engineering. Because when the data is all being spit out for you, you need a different perspective in order to have a different view of the data. I spoke with Sal Khan of the Khan Academy about the future of AI in education.

Though that conversation is nearly two years old, his advice continues to be prescient. Give it a listen if you're interested in this topic. There's a link in the episode notes. One thing I've always thought to be funny is that we have such a compartmentalized relationship to education. We do nothing but learn for 20 years, sometimes more, and then we do nothing but work.

Why not let kids apprentice in jobs that fascinate them, interlard work earlier into the learning experience? And why not give adults more access to vibrant and scintillating educational experiences throughout their lives? As I...

tour colleges with my kids. I'm dying to go back myself. Well, that's what we have set out to do in our own small way at the next big idea club. It's really an extension of what we're doing on this podcast. We're inviting the most fascinating thinkers in the world. onto the show to engage on the topics that matter most right now. What we do when you join the club is we send you the six most important books of the year from the six thinkers who most excite us on the show.

so you can discuss them directly with the authors themselves. To learn more, go to nextbigideaclub.com. Use promo code podcast for 20% off. Big thanks to Vanderbilt University for organizing this event. To learn more about what they're doing to transform higher education, check out their podcast Quantum Potential. There's a link in the episode notes. Today's show was produced by Caleb Bissinger and mixed by Mike Tota. I'm Rufus Griscom. See you next week.

The Uniswap wallet makes it easier and safer to own and use crypto. Created by pioneers of the crypto economy, the Uniswap protocol has powered over $3 trillion in trading volume, and it's trusted by tens of millions worldwide. With the Uniswap wallet, you can discover, swap, and manage your crypto all from your phone. Buy your first crypto assets in just a few taps and start exploring the freedom of decentralized finance with Uniswap. Tap the banner to get started.