A New Model for University Research?

Chapters

0:00 - Introduction

2:00 - Introducing Our Guest

3:09 - A ‘Cycle of Reactivity’ in Research

4:56 - Moving From Entitlement to Impact

7:12 - How Research Funding Needs to Change

11:45 - Can Private Enterprise Make Up For Lost Federal Funds?

13:41 - What About Global Competition?

16:36 - The Reverse-Pitch Model

18:22 - Should Colleges Benefit More When Research is Commercialized?

23:07 - Why Aren’t Companies Complaining About Cuts?

28:40 - How More Local Funding of Research Alters the Landscape

32:11 - How Will AI Impact University Research?

37:15 - Sponsor Break

37:48 - A Candid Assessment

39:13 - Making the Value of Research Clearer

41:57 - Defining What Research Serves ‘The National Interest’

45:12 - Cuts Could Have Long-Lasting Consequences

47:58 - Searching For a Long-Term Consensus

51:41 - AI Could Bring New Players In Research

53:44 - Wrap-Up

Relevant Links:

“The Endless Frontier,” by Vannevar Bush, 1945. 

“Science: A New Golden Age,” U.S. Office of Science and Technology Policy, 2026.

“CU Boulder joins CU spinout Inflection, other innovators in America's Quantum Space Initiative,” U.C. Boulder website.

Michael Crow says invention of iPhone was backed by the work of 3,000 researchers. ASU’s Instagram account.

Transcript

Justin Schwartz

They have to sort of look out for themselves. [They] almost have a fiduciary obligation to act as narcissists.

Michael Horn

Jeff, he's talking about corporate America, and he runs a university.

Jeff Selingo

That's Justin Schwartz, chancellor of the University of Colorado Boulder, engineer, inventor and founder.

Michael Horn

We asked him about the deal that's paid for American science since World War II. Washington funds the research, universities do it, and industry turns it into products.

Jeff Selingo

And that deal is being rewritten right now, and he thinks universities have to make a different case for it.

Justin Schwartz

How do we make sure that we're really engaging in a conversation that isn't about entitlement, but is about impact?

Michael Horn

Plus whether private money can fill the gap.

Justin Schwartz

I think it will be more of the leveraged value added than a replacement,

Jeff Selingo

And why he isn't panicking over the budget fights.

Justin Schwartz

I don't think any of us were alive the last time a president's budget was approved.

Jeff Selingo

And, yes, I asked him about AI and research.

Justin Schwartz

You had to go there, Jeff, didn't you?

Michael Horn

Some schools are buying all the computing power that they can get. He has a name for it.

Justin Schwartz

‘if you build it, they will compute’ kind of mindset

Jeff Selingo

Then Michael and I take on the harder question. What universities have to change to keep earning public support?

Michael Horn

Including an idea that higher ed's not gonna love, the federal government, it's a customer.

Jeff Selingo

That's all ahead on this episode of Future U.

Sponsor

This episode of Future U. is sponsored by Ascendium, a mission-driven nonprofit committed to improving learning and training systems to better serve learners from low income backgrounds.

For more information, visit ascendiumphilanthropy.org.

Subscribe to Future U. wherever you get your podcasts. And if you enjoy the show, share it with your friends so others can discover the conversations we're having about higher education.

Jeff Selingo

You know Michael, as we talk often, university research helps produce the medicines and the technologies we rely on and it trains the people who turn discoveries into something we could actually use.

But the system supporting that work is undergoing perhaps its biggest shift since it was developed in the aftermath of World War II.

You know, Washington is rethinking federal research policy, private investment is growing in university research, and of course, like everything, AI is changing how research gets done.

And so the question really is like what role will universities play in the next generation of discovery?

So our guest today is Justin Schwartz. He's the chancellor of the University of Colorado at Boulder.

He's an engineer. He's an inventor. He's an entrepreneur. And we're gonna get into all of that in the conversation today.

He previously served as Penn State's provost and the dean of its College of Engineering.

And we're gonna explore who should fund university research, what stronger industry partnerships could look like, and whether AI could expand access to discovery or just end up concentrating it among the haves in research.

So Justin, with that, welcome to Future U.

Justin Schwartz

Thank you. It's great to be here. Look forward to the conversation.

Jeff Selingo

So Justin, you know, the post-war deal around research was pretty simple. Right? Washington funded basic research, industry turned it into new products.

The White House just swapped that kind of post World War II blueprint called ‘The Endless Frontier’ for something called ‘A New Golden Age,’ which seems to be less institutional and less higher ed focused.

What are you most worried about when it comes to U.S. research policy right now?

Justin Schwartz

So I think what worries me the most is that we get into a cycle of reactivity.

So the White House puts something out, higher ed reacts, OMB or the White House reacts. And at no point do we does anybody call time out. Let's take a deep breath and really have a deeper conversation about what is the role of the federal government in advancing R&D? What is the role of higher ed in advancing higher R&D? Where does industry fit in?

We're in a different era, I think, than we've been in a long time, if ever, in terms of cash flow to big tech and the presence of, you know, the uber-successful entrepreneur, deep-tech guy, VC in engaging in the R&D space. Right?

I mean, you know, we're historically number one in the country for NASA science funding, but all of a sudden, you've got Blue Origin and you've got SpaceX and you've got, you know, multiple companies getting involved in space, and that's, you know, big, big money is behind some of them. And that's just one example.

So to me, the concern is that there's a rushed attention rather than a deeper conversation about what makes sense.

Jeff Selingo And

you really just want that certainty in some way, whatever the certainty is.

Justin Schwartz

Yeah. Right.

I mean, I wouldn't use the word certainty because I do think there's a little bit of an air sometimes in higher ed of... I don't think privilege is the right word, but entitlement. Right?

So, you know, the Vannevar Bush model was a great one for the time. And let's be honest. A lot of it... While we talk about a lot of the basic science and the basic science was absolutely fantastic, of course. Right? I mean, you know, in our case, the decades of basic science collaboration between the federal government and the university created JILA, right, with this unique joint lab that we have with NIST. And that's really why we have multiple Nobel Prizes in the sciences that relate to quantum, and now there's this massive race for quantum technology. Right?

I mean, there's a lot of there, there, for sure.

Same thing in health related... You know, can't understate or can't overstate where NIH funding has come in terms of driving, you know, health health outcomes across the globe.

Nonetheless, a lot of the Bush model, which drove the creation of NSF, it kinda really pivoted to advancing Cold War initiatives. Right?

So much of the R&D spending outside of NSF and NIH was Cold War driven. And that was... I mean...

And I'm not saying that to criticize the funding, but that's just kind of fact.

And so to me, the question is, ‘How do we get away from this mindset of of, you know, sort of a built in expectation the federal government has to do this,’ to stepping up in terms of here is the real value that we are bringing to to our local communities, our states, the country, and the globe because we have this robust partnership with the federal government, which... You know?

And maybe I'm just an old school optimist democrat... Democracy guy. Right? But the federal government's role really should be representing the best interest of the country, right, and the population.

So how do we make sure that we're really engaging in a conversation that isn't about entitlement, but is about impact?

Jeff Selingo

Okay. So you talked about this different era. Right? This, you know, this expectation I think that higher ed for a long time has had that, well, the feds are always gonna do this. They always have done this.

You know, you've been on the... You've kind of been on every side of this, researcher, inventor, founder. You know, what part of the system do you think deserves to be shaken up, and what is the ideal system in your mind?

Justin Schwartz

So I'm an engineer, so there's no ideal system. I'm an old-school democracy guy. Right? It's the worst possible system except all others.

Jeff Selingo

That’s right.

Justin Schwartz

So I think where the conversation needs to be is, you know, where is it reasonable for the federal government when they fund research at a university to ask that there be other resources brought to the table?

I think that's a reasonable question from the taxpayer. Right?

So right now in in different states, you see, you know, significant investments from state governments in particular areas because it couples to economic development. Right?

And that's a very different enterprise than, you know, when the federal government invests through the NIH in in drug discovery, basic science. Right?

So things like things like that where it is clearly... You know, I think I'm... I don't think I'm being bold when I say health outcomes are a public good. Right?

So I think there's really a need to have some differentiation in terms of what type of research are we talking about. Right?

And then where is the role for private enterprise to to take part in the funding conversation, in the investment conversation? Right?

Do we think that our current IP model in terms of ownership when there's federal money involved, is that still the right one?

I'm not saying it's not the right one. Right? But if we're gonna have a... And what worries me, getting back to that initial question, is that we talk about a piece of the ecosystem and not the whole ecosystem. Right?

So for example, like the conversation that came in and maybe isn't still present around what is the right way for F&A to be to be funded to universities?

So my personal view is that there is significant room for improvement. I don't think going to an 18% flat rate is anything short of disaster. But I also think that higher ed needs to be open to a really constructive conversation because I think the current model really strongly influences how some universities become more of the haves versus the have nots. Right?

And so if your F&A rate is driven by the fact that you have a really large medical research enterprise and some of the other expensive areas, you are now getting a much higher F&A rate on some of the lower cost research activities also. Right?

And so that becomes... that's a very different narrative than if you're a smaller university, you know, an an emerging R1 or an R2 that's trying to grow research impact, and you don't have the internal expertise and capability to negotiate a higher F&A rate, and you don't have the basis for it because you're not running a teaching hospital or a research hospital. Right? You're perpetually trying to catch up.

And, you know, so a model that somehow balances out both the different costs based on geography. Right? I mean, it's more expensive to do research in Boulder and Berkeley and, you know, New York and DC, you know, Maryland than it is to do it in Lincoln, Nebraska. And I'm not knocking Lincoln, Nebraska. There's just a different cost of doing business. Right? So that piece clearly needs to be in the conversation.

But I also think there's a... There should be room for a conversation around the type of research that's being done. Right? And maybe that means going to a full cost-accounting model, which we haven't done. But maybe that is the the right way.

So, you know, and I'm not saying there's an ideal way, Michael. Myy point is we should have a time when we actually engage in conversation of what's the right way for the federal government to cover some of these F&A costs. Right?

There's inherent differences when, you know, some state universities get a lot of money for infrastructure. Others get very little for infrastructure. Privates are, of course, very different in that space.

So all these things are real costs and real factors, and we've just kind of peanut buttered them into this.

Michael Horn so we have

this sort of fundamental full conversation.

Justin Schwartz

A full conversation.

Michael Horn

Part of that, you mentioned private enterprise, and I wanna ask about that a little bit because one idea out there is that, ‘Hey, the private sector, you said match some of these types of research and development projects, probably not all.’

But this notion that it can step in, and you, of course, have forged a deal with Medtronic, in AI, healthcare, and robotics.

Is this possibly a bigger part of the future? Are there enough of these opportunities? And how much really are we talking about in terms of can private enterprise make up where there's lost funding in the current model?

Justin Schwartz

So I think it's going to be very much domain and technology specific for starters. Right?

So, you know, if you think about the way foundations have supported arts and humanities, right, there's been a foundation plus federal kind of mindset there.

And that's, of course, you know, less zeros than there is when you're talking about space research and quantum and and and health research. But there may be some lessons learned. Right?

So, you know, the Medtronic and the health tech is a great example.

Thinking about, you know, the... This massive growth in quantum R&D in the Chicago area here in the front range in Boulder and across Colorado, in Maryland, right, in other spaces, right, there is a real, you know, anticipation that there's a massive economic impact of where there is success. Right?

And so, you know, the federal government has consistently, over the last few years, invested heavily in quantum. We've just got a big award. You know, I think North... I think University of Chicago just did. Maybe it was Illinois just did. And so you actually...

And so where is the industry role in that? Right? That's the conversation we haven't really had in-depth.

Michael Horn

Well... And that's where I'm curious. Right? Because, you know, today, I think you get seven out of 10 of those research dollars from federal sources, if I understand it correctly.

If we start to move to more of this matched model, how much realistically, say, five years from now, if we have a bigger conversation and reset, what's the actual appetite or possibility of private enterprise stepping into some of these specific domains, and and how much of that can they realistically do at a place, you know, like your institution?

Justin Schwartz

Right. So, I mean, I think it will be more of the leveraged value added than a replacement, to be honest. But I think at the same time, we recognize there have been, you know, topics of fancy that have gotten significant federal input and federal resources for a while and then fade down.

And the question then became, you know, is this an area where now universities are just gonna step out and it's up to industry to continue the R&D or not? Or is there a more of a smooth transition as the technology matures, right, from the federal investment to the industry partnership mode as things go from, you know, different TRL levels from, you know, one to four into the six, seven up to the eight, nine space. Right?

Because this gets into... You know, we're talking about it from a U.S. university industry transition context, but there's another piece of this. Right? And the other piece of this that we've seen over and over again is what about the global competition? Right?

And so, you know, one of the risks that we talk about often, which is legitimate, is when we get to that point where, you know, with the very traditional and historically effective capitalist model of, you know, federal investment partnership with universities, state dollars, university monies in the early stage transitioning to to full on industry and then, you know, much of the university science role is done.

Well, then we've now got industry in a mode where it's competing, right, with governments that have a very different mindset around IP and tech development.

So, I mean, this goes back... I mean, I'm thinking back to 30-something years ago, right, when there was widespread alarms that the entire steel industry was gonna be in China. And the federal government stepped in with NIST.

Michael Horn

Because there's a national security manufacturing base, etcetera, interest in that.

Justin Schwartz

Right. So we see this where then all of a sudden, the federal government has to come back into an industry and say, we need to invest in this because it's a competitiveness issue.

So how do we think about these things more holistically? Right?

So right now in quantum, sure, we're competing with Chicago. Everybody knows that we talk about it all the time. But in Chicago and Maryland and Texas, we're also competing with China.

Michael Horn

Right. Right.

And so there's that question of unity.

I wanna get, like, sort of further that as well because part of this is there's obviously commercial interests. There's commercial interests that spin off of some of these things that we can't anticipate.

But as I understand it, you've developed this model for industry research partnerships, that involves sort of a reverse-pitch session. walk us through what sounds great. It sounds sort of like something Silicon Valley might get behind, but I don't know what it means.

So tell us what it... What that actually means.

Justin Schwartz

Right. So, I mean, the conversations around how do we get industry in the earlier stage. Right?

How do we really create a deeper partnership which serves our mission? Right? Because our mission is creation and dissemination of knowledge. Right? And dissemination is through papers and partnerships with industry, and dissemination is really what we do in the classroom. Right? A student learns and then they leave. Right? When they leave, that's what the transition from education to dissemination.

So what does a what does a deeper partnership look like that is, you know, research translation, but also research partnership that's also student engaged.

So if you're Medtronic, you're not just tapping into and benefiting from the intellectual capacity of our faculty. Right? But our students are involved. So it's almost the, you know, Medtronic or whomever you're partnering with having a hand in their employee development in an earlier stage in the educational stage. Right?

So more of a, you know, less of a four years here transition to industry, but really an integrated flow through.

Does that make sense?

Jeff Selingo

Yeah, it does. And I’m kind of interested in talking about another company, this quantum technology company that spun out from Boulder, Infleqtion, it it kind of has unicorn status now.

You know, obviously unicorns in the market are obviously few and far between, which is why they’re called unicorns.

Justin Schwartz

We have 11.

Michael Horn

That's pretty good.

Jeff Selingo

Pretty good.

So... But why does so little... It seems like so little research really pays off for higher ed, but yet we keep hearing these stories about how much higher ed research is used, in the, you know, in the world outside of university gates.

So what is it about that transfer that doesn't... Maybe is it working like it should do you think?

Or do you think that university should be profiting a lot more off of so much of the research that now goes into these, not only unicorn companies, but all these other companies as well?

Justin Schwartz Right. I mean, I think from... You know, our mission, of course, is what I said in terms of creation dissemination of research.

And I will counter my own self in a moment. I'm usually one who will say internally to my faculty, my staff, my leadership, but externally also that when higher ed starts the conversation talking about the dollars and not the sense, that's a mistake.

At the same time, though, to your question, Jeff, I think, yes, we should be... We talk about being stewards of our resources. One of those resources is our intellectual property. Right?

And so, you know, we... We're very defensive if anybody uses something we publish without crediting back. Right? That's like one of the golden rules in in, you know, academia.

And then historically, we've probably been much less effective stewards of our IP in terms of that translation.

Now there's all sorts of counterexamples. Right? I mean, we are... you know, as I wrote recently, we are a mosaic of types of institutions, and there are institutions that have done a really great job of managing their IP for a long time. There are others that are are newer to the game. You know?

So there's a lot of variance.

Like, I mean, I spent time at FSU, and Taxol was their big win. And then I was at NC State. Right? And and Cree came out of the department I was department head of. So there's examples in one direction. And, you know, NC State built Centennial Campus as a as a... Almost became a feeder to RTP. Right?

And certainly, one's not going to accuse Stanford of missing the boat on Silicon Valley.

But there are a lot of us who are later to the game.

You know, we've risen a lot at CU Boulder in the last few years.

So we've had multiple really strong years of creating startups. And, you know, as we look at it and we talk about should we build our own, you know, venture studio and what would it look like and how do we do it, a lot of the challenges, of course, you know, as in any venture case, the start-up costs. Right?

And, you know, where patenting is not cheap, patent protection is even more expensive, especially when you're looking at global patents.

So there's a lot of things that go into it.

And I think where we've started to have the success, you know, you asked about Infleqtion. Infleqtion is, obviously, a true point of pride for us.

But the founder of Infleqtion is still part-time faculty with us. Right?

So this wasn't a we created IP and then gave it to somebody else who created a startup, and we didn't make a good deal. Right? This is someone who's much more connected to us continuously.

And we're still in regular conversation with Dana Anderson. He's, you know, he's someone I consult with often on all sorts of things.

I think what we need to think about is, you know, how do we as an individual institution in our particular environment in terms of what are Colorado laws, what is the state expectations, what is the ecosystem in Boulder and in the front range for creating startups and having an entrepreneurial culture, right? All of those things factor into how you approach it.

But I do think more and more institutions of higher ed are getting into this game because we're realizing, you know, we need to have diversity of our funding streams, and we should certainly benefit from the things our faculty do.

There are faculty at my institution who will tell me, and they do tell me, that they've got, you know, multiples, very successful startups that they created years ago despite the university.

And they didn't have the support they should have. And now that we're bringing it, you know, we're jumping to the top of the number of startup companies created every year.

Jeff Selingo

Yeah. So, Justin... I'm just kinda curious because I was with your colleagues recently... Your former colleagues recently at Penn State. And we were just talking about the pipeline from university lab to, you know, consumer product, for example, or, you know you know, government product or whatever it might be. Right?

You know, you were talking about SpaceX earlier. And, you know, those pipelines, I think the public kind of misunderstands how long some of those pipelines are. And, you know, and and there's this flywheel. Right? You constantly have to have stuff in the pipeline if you're a pharmaceutical company, right, if you're a consumer product company, if you're Apple, whatever it might be, you always need something in the pipeline because you know five or ten years from now or two years from now, you're gonna need a new product, you know, coming out into the marketplace.

And so the concern now, I think, among many research universities is that given what's been happening to science over the last couple of years, that is that pipeline going to be a little smaller, a little tighter in a couple of years?

And I just... I guess I'm wondering, you know, you've been raising... Not you specifically, but higher ed in general has been raising the alarm about this. So why has an industry been objecting to some of this also? Because they really do rely on university research.

Justin Schwartz

Right. So I can't speak for industry directly, but I do think there's been appropriate alarm in higher ed because of things that were proposed by the White House. Right?

So, you know, I very much in this role have become a Greek stoic. Right? And there is... You know, we suffer more in our minds than in reality.

There was reason to raise the alarm and to, you know, put into hyperactivate all of our government relations teams from across higher ed when Trump's first budget proposal came out.

But I don't think any of us have been... were alive the last time a president's budget was approved. I'm not sure that the president's budget is... I mean, maybe Washington's budgets were approved. I'm not sure.

Michael Horn

I'm not even sure about that. Yeah.

Justin Schwartz

And so if we look at what the changes were in terms of what was actually passed by the congress and signed by the president in last year's budget, this year's budget, it wasn't as dramatically different as what we had the year before.

And so that doesn't mean there haven't been significant changes. And I, you know, someone said to me a long time ago, there's no such thing as a recession. Right? If you have your job, you're fine. If you don't, it's a depression.

It's kind of the same thing in terms of our faculty. I don't wanna discount those whose funding really was hit significantly. But a lot of it was also, if you remember, like, canceling grants and then pushback and then getting reimplemented.

So there's been a lot more churn, I think, than than change would be, I think, a fair statement. And if we... I haven't done the analysis, and so my gut instinct could be wrong. But if we look at step function changes in agency funding levels as we've gone from one... A Republican to a Democratic administration or vice versa, I'm not sure the step function changes here have been significantly bigger than in past major transitions. Right?

If we think about Carter to Reagan or Bush to Clinton or Bush to Obama. Like, it would be interesting to actually do that analysis and say, how big was the change now in reality?

Certainly, there's been a doubling down in certain research spaces like quantum in the current administration, like all the things in the Genesis envelope as well.

But there was a doubling down on cancer research under Obama. Right? And that was great. Right? That that should be.

So I think if I were to guess, because I don't like to speak for other people, my guess is industry hasn't, like, stepped in that sort of context or verbiage because they don't wanna be seen as being counter to an administration either.

But that doesn't mean they haven't stepped up when it's in their best interest. Right?

One thing we could expect from industry is to... You know, they have to sort of look out for themselves. They're almost... they have a fiduciary obligation to act as narcissists. Right? If you have... If we're gonna treat them like people, then we ought to put them in psychological categories. Right? They have to operate in their best interest.

Now some companies respond and are thinking only in the short term, and other companies respond and think about long-term also.

So here's an example. I hope they don't get mad at me. BP. Right? Formerly British Petroleum, then they became Beyond Petroleum. I haven't seen them talking and advertising about being Beyond Petroleum and looking at the energy transition. But I think if you look at what they're doing, they haven't really changed much. Right?

So they've changed what they talk about, what they market, but I think they are actually still thinking long term, right, in terms of being an energy company rather than a petroleum company.

Now that doesn't mean everyone's done that. Right? There are some who sold their wind leases back to the federal government like that,

and others

who have said, no. We'll stay where we are. Right? And just kept a little more quiet on it.

So I don't think it's fair to lump... So industry is not homogenous in that respect.

Michael Horn

Yeah.

So Justin, another part of this sort of if you think about the various folks that can fund research and you mentioned it that there's the national interest. There also might be local interest in sort of home, closer to home interest, if you will. And so this goes to states. Right?

Do you see Colorado potentially putting a greater share of its funding into your research?

What would be the pitch to states more generally to do this?

And sort of could that then force universities to actually become more differentiated? You mentioned all the competition around quantum. That's a national interest, so maybe that makes sense. But maybe there's other parts where, you know, certainly, we know in Arizona, ASU has leaned greatly into certain environmental concerns that are of note for that particular region.

Do you think there might be more of that if states take on more funding?

Justin Schwartz

Yeah. It's... You know, every state is very different, I've learned.

So I've been, you know, I've been in Florida, I've been in North Carolina, I've been in Pennsylvania, I grew up in Illinois and was short on Illinois faculty for a bit and now Colorado.

Arguably the biggest outlier and this isn't a criticism is Arizona State, Right?

So Michael Crow came in a time 20 years ago, and at the time I had family living in Phoenix, I kinda watched it real time, when that urban area was really poised for growth, right, for connectivity between Phoenix and Tempe and Scottsdale and Glendale and all the communities.

And, you know, I lovingly joked that ASU is an urban development organization that runs a higher ed institution. Right? And Michael's fantastic. I consider him a friend, and I... you know, he's a role model of sorts, but I don't think what he did maps to many other places. Right?

But what he did was amazing.

And hen I say every state is different, that's my euphemistic way of saying that Colorado lives under a thing called TABOR, the taxpayer bill of rights, which is essentially a voter approved measure 20 or so years ago, that really limits to how much the legislature can just raise taxes.

And so recently, and recently, mean, like, in the last year, there has been uprising is unfair. There's been an outpouring of conversation driven by some of our most successful Colorado entrepreneurs basically saying, ‘Hey, Colorado has gone from one of the very best places for high-tech entrepreneurship to dropping down a bit, and we... If we need to do something now before it's too late.’ Right?

And so governor Polis created a Colorado Competitiveness Council to really talk about these issues. I... I'm on the council, but I'm not a driver. Like, it's truly a multi enterprise group.

And, you know, the timing is such that governor Polis... His former governor Polis come January, he's term limited. You know, we know who our candidates are, the Democratic candidates. My former law dean. He started Flatirons... The entrepreneurship... The first entrepreneurship wing out of Boulder 20 something years ago.

So we're really thinking through what can we do, what are the factors that drive an ecosystem's ability to thrive. Right?

And so in Colorado, like, when... In our case, Colorado tends to lean more into the tax credit space than it does to the cash on hand space because of the limitations of TABOR.

Jeff Selingo

So, Justin, we were just talking about the cost side or the funding side of this. Let's now talk about the cost side of this, and let's wrap up here with... Because we can't leave without a question about AI these days.

Justin Schwartz You had to go there, Jeff.

Jeff Selingo

We had to

go there.

Well, and there's so much discussion around AI helping researchers do research faster, better, at bigger scale.

But I'm also wondering if it also becomes the next arms race. Right? You were talking earlier about the haves and have nots among research and higher ed. And, you know, is it now going to be the cost of compute? Is now going to be the next lab space, you know, deal around the competition in higher ed.

How are you thinking about AI and research? Do you see it as something that's going to help research or just make it more expensive or both?

Justin Schwartz

So probably both, but it depends on the field.

So, you know, we look at what's happening in a few places. Right?

So OpenClaw has a partnership with MIT specifically on research investigation. And one of the areas they're focusing on is, of course, materials development.

So as a former head of material science department, right, let's be honest, experimental work in science is expensive too. Right? And so, you know, is AI going to be a more cost-effective way of discovering new materials than the experimental way, right?

Because, I mean, in material science over time, last 20 years has gone from a, you know, truly a hands on experimental kind of discipline to more and more computational presence also. Right?

And that's helped drive the relationship between material science and biological impact. Right?

So in some cases, one might argue that AI could make it more cost effective because doing those computational experiments is cheaper than doing a lab experiment. And, also, you know, there's a cost benefit when you can accelerate research. Right?

And so I think the AI people, if you sit in the AI chairs, we're gonna accelerate research. We're gonna accelerate outcomes. You're gonna do, you know, experiments on the computer so much faster that you're gonna narrow down what you need to do in the lab.

I suspect the drug discovery people for AI are thinking the same thing.

And then there's other areas of research where, you know, the cost of compute, the cost of AI is gonna be a new cost that they hadn't thought of before that could become, like you said, the computational arms race.

What I see in... like, I'm seeing a number of different reactions in higher ed to the AI enterprise. Right? And they vary from we're going to concentrate our AI on this discipline, right, and go deep.

And then there's the ‘if you build it, they will compute’ kind of mindset where you're looking holistically. Like, if we just, you know, build these massive computing capabilities, then great things will happen. Right?

And that's not new. It was 1985, because it was right before I graduated with my bachelor's degree, right, that the federal government started putting supercomputers at different universities. Right?

So I was graduating from Illinois right when Illinois got named the first... one of the first supercomputing centers. Right? So it's not that this is the first time we've seen massive infrastructure in computing. It sort of comes around every time there's a generational change in the type of compute.

The question today goes back to our first conversation. Right?

So how much is the federal government gonna go and say, okay. We're gonna pick five places, and we're gonna buy each of those five places the absolute best that just came out of, say, NVIDIA and create a national computing hub?

And how much of it is going to be individual institutions sitting down with NVIDIA like UF did in NVIDIA's early days and Michigan is doing as a OpenClaw investor or partner in the OpenClaw Foundation.

And, like, where is it gonna be... where is the federal government gonna... have a role in this, and where is it gonna be industry institution doing it together?

So it's gonna actually be, I think, a great test case for this earlier conversation we had about what's the federal government's role and how much is simply industry/university partnership.

Michael Horn

It's just gonna be fascinating to sort of watch this tug and pull of deliberate versus emergent strategy, if you will, almost.

And Justin, this has been a fantastic conversation. Jeff and I both really appreciated the nuance you've brought to this and also the humility, to think about this whole pie, in a rapidly moving for research, guess you could say, situation.

So huge thanks for joining us.

Justin Schwartz

Thanks for having me.

I've enjoyed it. The time went by fast.

Michael Horn

Yeah. Well, there you go. And I think our listeners are gonna love it.

And for them, we'll be right back on Future U.

Sponsor

This episode of Future U. is sponsored by Ascendium, a mission-driven nonprofit committed to improving learning and training systems to better serve learners from low income backgrounds. Ascendium envisions a world where low-income learners succeed in postsecondary education and workforce training as paths to upward mobility. Ascendium's grantees are removing systemic barriers and helping to build evidence about what works so learners can achieve their career goals. For more information, visit ascendiumphilanthropy.org.

Michael Horn

Welcome back to Future U.

Off an important discussion with the chancellor University of Colorado Boulder about a topic that I think many in higher ed have long just sort of assumed. Right?

The compact between the feds, universities, private industry around research, the way we've done it sort of just keep doing it. It's worked for many many decades.

And Jeff, in that discussion with Justin, I, you know, I found it really interesting. Like, was willing to both question how the research system works even as he defended the importance of having such a compact between these different entities.

And what I appreciated, I'll say upfront as we get into this, Jeff, is that as he was talking about it, his basic thing was like, let's not just piecemeal our way here. Let's think strategically about what we're still trying to accomplish from the system. And then we can... You know, once we have sort of those clear, clear values, priorities laid out, then we can think about from a whole-system perspective, how should we rethink this? Because he was very candid. Not everything is working.

Now I'm curious, what do you think, you know, universities would need to change in themselves to make their argument for continued public support credible?

Because it can just sound like... And he made this point as well, just sort of I'm entitled to it, an entitlement. And he was very emphatic, that's not what this should be.

Jeff Selingo

Yeah, Michael, know, what stayed with me was his point about impact.

You know, he asked universities to explain the value of their partnership with the government and what it brings to communities and the country.

And that's a much more useful starting point for the conversation I think about the Future U. know, really universities need to move beyond kind of chasing prestige and chasing grant volume, you know, they love to say, ‘Oh, we did more university research this year than last year.’ Right? And they need to develop an approach around, I think, the fields where they can actually make a contribution.

You know, if you look at some universities, do you really wanna always be the fourth or the fifth in a field or maybe the ninth or tenth in a field? Or do you really want to be the first or second? And that's really where we should be spending our money.

And for me, that means a university president or a VP for research should be able to explain why they are investing in particular research strengths, right? How those connect to their region and how students ultimately benefit, whether that's undergraduate or graduate students.

You know, I think a larger research budget by itself really doesn't answer those questions.

And I also don't think for the public at large, Michael, this idea of dividing research into basic and applied really tells us what we're funding.

I think if you ask any person on the street, do you know what basic research is? Do you know what applied research is? I don't think anyone could answer that question.

I think they wanna know what purpose is, they wanna talk about purpose. Is it solving a community problem? Is it developing a new product? Or is it expanding knowledge? And I really do think the best example of making an argument for continued public support are just giving us concrete examples of what university research has done.

And you know, Michael, there's this video that went viral of Michael Crow who, you know, the audience knows I work for ASU as a special adviser. Michael Crow is the president of ASU. And he was on this panel at Rice University earlier this year. And there was a point during the conversation, he held up his phone and said, ‘You know, everyone thinks the phone, the mobile phone, the iPhone, he held up an iPhone in this case, right, was the product of the genius of Steve Jobs. And he said, by his estimation, it's 5,000 researchers that went into that.’ It's a pretty powerful statement. So let's listen to it for a minute.

Michael Crow

I left my iPhone out because everybody thinks that this iPhone 16 is the product of the genius, and he was a genius, of Steve Jobs. Hardly. There were probably, by my estimation, 5,000 academic research groups through the decades that had something to do with the technology I'm holding in my hand. Nobody even knows they exist.

Michael Horn

Jeff, what I take from you and all that is one, something we talk a lot about is, like, own, you know, the entity or the person that owns the story or the metaphor owns the debate a lot of times.

And often it can sort of fall back on platitudes is what I'm taking from, what you just said. And really forget about those specifics, how I really show what this is.

You know, Harvard obviously has been defending itself a lot in the court of public opinion right now, and it sort of gets vague, oh, cancer research, things like that. We obviously, you know, mom and apple pie agree with, but like give us the real tangible stories, make this really feel very practical. I think that's a really good point you just made.

The other point that you made that I just wanna comment on as well, and it gets to both the clip we just saw and something you said is the purpose. What's the purpose of this? And something Justin, made the point about that I think sort of stems downstream of something he said was in many ways, and higher ed's not gonna like hearing this, but federal government is the customer of this research. They're... by definition, it's their dollars and your payer is your customer.

And I guess what I found compelling in Justin's point about the importance of R&D and and the federal government in this compact was the importance of being clear about how that... Those research dollars tie to the national interests with clarity around that. Not amorphous, ‘Oh, it's good for, you know, the U.S. to somehow be first here,’ but real clarity.

And his point about how the initial focus in this compact that came out of the 1940s was largely around the Cold War and competing with the Soviet Union really did land for me.

And so I take from it, I guess, a couple things here. One, I think there are large swathes of higher ed that aren't going to love it, but I think that we have to have focus. What is the national interest? What is your part of the compact, higher ed, in exchange for those dollars from the customer of the taxpayer?

But two, I think it requires something from the federal government, Jeff, which is I think clarity around just what is the national interest and why is it the national interest? What are they trying to advance and why? And then therefore, where would basic research, applied research, forget about the distinctions, research in general, how does it actually advance us toward that ideal?

And I don't think the federal government has done a very good job as they've thought about the transition to something different in explaining that either, Jeff.

And so, look, the answer might not always come from higher ed for research. I'm very happy to acknowledge that there are other sources of research in the future.

But I think we need to be super clear about what that outcome and purpose is, and then we can think about where's the best place to fund it.

Now along those lines, one possibility that Justin raised was deeper partnerships with industry. How do you build those partnerships though while preserving room for research that... I mean, as you and I both know, there's a lot of research that should be done that doesn't have an obvious commercial application.

Can private funding really fill that space, replace federal dollars there? I admit I'm skeptical on that.

Jeff Selingo

Yeah, Michael, I thought his answer about replacement was pretty important, right? He sees industry adding value to the public investment.

And I wanna pick up on another thing he said.

When a student learns something and leaves the university, that's also the dissemination of knowledge, right? It's not just the actual research piece, right? The paper, the finding, the actual product.

And I think his Medtronic example makes that tangible, right? The company is involved with faculty research and with students who really become its future workforce, so another important piece of the research ecosystem.

But I think there are limits to the idea that another source of money will fill every gap.

You know, I remember our mutual acquaintance, Josh Jarrett, when he was at the Gates Foundation and everybody thought, well, the Gates Foundation, they have a ton of money. Right? They could just replace the feds on, you know, education stuff.

And as he put it on this panel that I moderated years ago at Georgetown University, he said, you know, even the Gates Foundation with its endowment, I don't even know what that endowment is, but in the billions of billions of dollars, right, he said we're pennies to the federal government's dollars. You know, the scale simply isn't there.

You know, Michael, the other worry that I have is capacity. And I think of this in a very personal term, you know, in a very personal way.

You know, as you know, since he's moving up to your neck of the woods, my nephew just finished his PhD in chemistry from Colorado State and he now has a job in Cambridge working for a pharmaceutical company. And that is the direct result of our investment in graduate students, in the infrastructure of graduate education and research.

You know, funding a project is only part of that infrastructure.

You know, my nephew worked for years in the classroom and in the lab and his lab work was like any science, right, some led to findings and research, but a lot of it was about his education and what is to come.

And that's why I keep coming back to the pipeline. You know, Justin's quantum example started with decades of research before any commercial opportunities that we discussed.

And my concern is that weakening that capacity now could mean fewer opportunities years from now.

And I think, Michael, just like corporate America doesn't like uncertainty in geopolitics, right, we're seeing that now, you know, with oil and the war in Iran. Neither does higher ed. They don't like uncertainty either.

And say, you know, when there is a change in administrations and a grant is eventually restored, a researcher may have already postponed hiring somebody or a student may have already chosen another career.

And my worry is that those decisions can really have consequences well beyond this administration.

Michael Horn

I'll just say quickly on that last point. This is why I just wish the... There would be that larger conversation that Justin effectively asked for on our podcast, to create a more enduring consensus around this direction as opposed to this...

This is a theme on our show. Right? The ping-ponging back and forth between radical extremes between, as different administrations come in over the last several years.

And, this is another place, Jeff, that I think a real public conversation, that was thoughtful and acknowledged is gonna take a few years to hammer out with an enduring consensus would be worthwhile, I think.

The other piece of this I think also gets back to clarity around purpose. Right?

And I'll pull it out a different way here because of your anecdote and good job telling a story to own the, own the conversation. But frankly, like what the purpose of graduate education is sort of connected to that story you just told.

And we had a discussion on this podcast a few years ago with Len Cassuto, of course, about how few graduate students actually take tenure track jobs in the academy or even get tenure.

So I would say, like, let's build those pipelines into industry, into productivity for the country with a lot more clarity so that it's easier to show the ROI, I think, and make that case, Jeff.

And that will require, just to be clear, some significant redesign on the part of higher ed. That's something we talked about certainly there. Clarity to the students when they come in about what the outcome could be, how the different pathways, maybe different number of years. Right? Because maybe the purpose isn't publication and things of that nature.

It also will mean... And again, this will, I think, be controversial in some quarters as well, but I'm just gonna say it. I think it'll... It also means some clarity around international students as well as our K-12 pipeline, in this country.

There's, I think, a lot of murkiness over why we take so many international students in this... In STEM fields and it's frankly because our K-12 schools are not producing the students we need to be like, your nephew in that case.

And I think there's a real question of what's higher ed's role in this. Higher ed historically has said, you know, test …

Jeff Selingo

Not our problem.

Michael Horn

Not our problem. Like we use them because we know you won't succeed here. I get it.

I don't want you accepting students that aren't gonna succeed either, but I guess the question is do you need to redesign so that more of them can succeed given that this is what the pipeline is producing, I think.

And then on the on the international front, again, I think this might be controversial, but I don't actually think it should be, which is I I think it's fair of the federal government to expect that in exchange for letting someone study here with a student visa, there should be some expectation of what's the benefit for the U.S. as a country.

And that's not always gonna be the same answer to be clear, Jeff. So there's some nuance here, I think. Right? It might be diffusion of values and knowledge and cooperation with other countries when they go back. That could be a valuable outcome.

It could mean that people are gonna stay here and build businesses or contribute in some significant ways. There could be an ROI, right, as a result for the country. Different answers.

But I do think some clarity and some expectation that the students coming here understand on the front end that they are buying in to the values of this country to some degree and the university's historical traditions at least of open inquiry and liberal liberal education.

And I think those are important values if they're gonna take advantage of these treasures, in our country.

But let's bring AI into this. If it... That's a little twist here, Jeff.

Because Justin described both the potential for cheaper experiments and the cost of computing that could actually just create a lot more cost because you're just doing a lot more work.

You know, there... I think there's a question of whether this will push universities toward more collaboration, more specialization, a whole bunch of questions.

We also got this comment, from one of our listeners, longtime followers of ours, Dr. Andrew Sears, I just thought I'd read it because I wasn't super familiar with the example, but I thought it was interesting. Discoveryloop.com and similar initiatives like it using AI to reinvent research around AI and robotics.

And his argument was this is a pretty massive potential disruption to traditional university research that's not coming from other universities, but from the private sector.

And he's sort of imagining a two tiered system in the future with AI to identify potential promising discoveries and then robot automated research gigafactories that are actually, you know, sort of the wet lab if you will for the experiment.

I think it remains to be seen where this goes, but as you know, Jeff, there's been this huge controversy over the last several weeks in the math area, with this not... I'm gonna mess mess up the pronunciation, but then, and show my limits, but the Navier-Stokes existence and smoothness problem with OpenAI claiming a swarm of their agents solved this problem. Other people who had been tackling this for many years sort of, threw in their shade at it saying they're not getting the contributions they put forward and and and things.

But I think the point it illustrates is there's this whole other dynamic of sort of a disruptive entrant, if you will, that we haven't even thought about.

It's not gonna be in Google Labs or or, you know, IBM Labs or you know, the sort of antecedents or or or the new forms of the Bell Labs of yore, that come in and do some of this basic research.

But there may be AI specific entities that are fit for certain kinds of research, Jeff.

So that's like another huge twist and how does that connect into all this question around collaboration or specialization?

Jeff Selingo

Yeah. And, you know, Michael, one of the worries that I have, and first of all, we're... I don't think we're gonna be solving any math problems on this, on show.

Michael Horn

Clearly.

you know, what we did in college.

I'll give you... I'll just leave it at that. Right?

But, you know, one worry that I have speeding headfirst into AI and science is, you know, the haves and have nots is just going to... That divide is just gonna increase.

As we heard this summer from the chancellor at UC San Diego, unequal access to tools and infrastructure could really concentrate research power further.

And, you know, we also know that sometimes research can move too fast without the proper rigorous controls.

And I think that makes Justin's distinction between going deep in a particular discipline and building computing capacity in the hope that useful research follows is kind of really important.

You know, my takeaway is that universities should start with the research they want to advance and then ask AI, you know, and ask then what AI would actually improve. Right?

Buying capacity doesn't really settle that question.

And that leads us back to collaboration. And this is where I think, you know, going back to UC San Diego's chancellor, I think computing resources that institutions can share. And then we're... If you think of it as a hub and spoke, the hub being the computing resources, then what does each university bring expertise to make those resources useful?

And what I like to see is a university that can explain where it will lead, where it will work with others, and I think that's going to give government and industry a clearer understanding of what their investment makes possible.

Say it upfront, right, and it would help universities, I think, protect the long term research capacity that we've had because upfront they're going to say, here's what we're gonna do and they're gonna be held accountable for that.

Michael Horn

Jeff, I think it's a good place to leave it.

And it actually relates to some themes that we've had all year on this show.

One, focus on where you're gonna be excellent. Don't try to be all things to all people. Really nail that down.

Do you think there's this disruption angle that there could be some low cost entrants coming in, and so maybe some universities will make their mark as part of that. It won't just be the private sector.

TBD around all of that and more, but I think it really does play.

And then one more thing, geography still does matter in this because different geographies are gonna have different research interests, different specializations, different priorities, and different opportunities unique to that region where they can really solve these problems.

So all things that I think really do connect back to a lot of the themes we've been grappling with on the show.

And in a topic that is so so important for the future of humanity, not just our universities, but sometimes gets short shrift in the conversations around the future of higher education.

So I'm glad we went on this conversation. Huge thank you again to Justin Schwartz, for joining us and for all of you for tuning in as always.

And keep sending your comments and feedback. Check us out at Future U podcast on all social media around Future U podcast. Or of course Jeff Selingo, @JSelingo, on the various social media channels, myself at @michaelbhorn.

And we'll see you next time on Future U.