Deep Dive
Shownotes Transcript
We underestimate China's ability to out-hustle everybody. It's a transcript that leaked from a war mobilization tabletop exercise in Guangdong province. It says, you know, basically, we've decided to invade Taiwan. How are we going to mobilize the population, mobilize the economy and industry?
America's science and technology ecosystem, what is happening to it? What is at risk? And what are the opportunities to make this better? Also, how is China interpreting what's happening in America's research ecosystem? How are they trying to build up their broader science and technology base? And what can we learn from Chinese war mobilization exercises to discuss? We have on Divyansh Kaushik.
and Alex Rubin. Both of them work at Bleak and Global Strategies. Divyansh has an AI PhD from Carnegie Mellon, and Alex spent the past decade at CIA looking at China and technology. Welcome to China Talk. Thanks for having us. Thanks for having us, Jordan. All right, Divyansh, let's do a one-on-one about the broader American research ecosystem and the interaction between universities, government funding, and corporations. What does it look like in the 2020s?
Yeah. So I think to better understand what does it look like today, we have to trace our steps back about 70 or so years and see how the American research ecosystem was conceptualized. It was that the universities would conduct curiosity-driven research funded by the federal government.
And American industry was more focused on taking that research and putting it into applications. Yes, there were certain industrial monopolies that were created by the government, and Alex can speak more intelligently to that, which also conducted basic research. But overwhelming majority of basic research happened in academia, universities that were created
by the uh as land-grant universities or other universities that existed before the war uh and that system served us really well uh throughout basic research uh developments that happened in 1950s 60s 70s bore fruit uh
you know, 10, 20, 30, 40 years later. That's the nature of basic research, where you do not necessarily have an application in mind right away, but one may emerge however many years down the line. So I'll give you an example. In 1970s and 80s, during the AI winter, when nobody was funding neural networks research because it was seen as a dead end and there was no application at all,
National Science Foundation that was created to fund basic research through the federal government was funding Jeffrey Hinton's work on neural networks, the only one to fund it at the time. And that work, fast forward 50 years or, well, 40, 50 years, let's say, has really shaped how we view AI today. That's the foundational technology behind all these large language models that you see. NSF also funded, for instance, the
entire PhD of Andrew Bartow, who's the father of reinforcement learning along with Richard Sutton. Andrew created the entire field of reinforcement learning with Rich Sutton.
At a time when there was little application of reinforcement learning, little practical application. And reinforcement learning is the other component that goes behind LLMs, that goes behind AlphaFold and the likes. So that's how America has pursued basic research. Now, looking to today, you see a lot of emphasis on why are we conducting research on reinforcement
uh, shrimps on a treadmill or, you know, fish being on cocaine or things like that. And there are very genuine reasons why those projects have been funded, but to the general public, it appears, oh, we are wasting a lot of federal research dollars. We are wasting money on things that have no actual application. Uh,
And whereas agencies like, say, the NIH, the National Institutes of Health, they fund more applied research on medicines and have tangible outcomes that they can point to. Look, NIH funded this drug, that drug. NSF, on the other hand, funding basic research cannot point to these tangible benefits for decades, as, for instance, happened with neural networks.
And so we are in this moment where there is a deeper conversation happening on what is waste, what is abuse of federal research dollars, and how to put that money to better use. Is industry funded research the way to go? Or does the federal government have any role in the R&D ecosystem? Should it have a role in the federal R&D ecosystem?
And what about public-private partnerships that were one of the anchors of the Chips and Science Act in creating the Technology Innovations and Partnerships Directorate at the National Science Foundation, for instance? That is a big conversation that we found ourselves in this year, especially as you see the National Science Foundation, the Department of Energy, Department of Defense, and the National Institutes of Health
being at the center of this question on where do federal dollars go? Should they go to research or not? Well,
Let's just do a little bit more history, because I think the sort of role of the broader defense community and funding and funding R&D over the past 75 years is also not something we can necessarily ignore. Alex, do you want to take that on or? Yeah, sure. I mean, I think it's a great example and I think it's best told through some some like an interesting case study. We're focused quite a lot right now on the semiconductor industry, rightly so, because of its big economic and strategic implications.
But even the foundation of the semiconductor industry, you can draw a line right back to funding from the Air Force and from NASA for the Apollo program and then from the ICBMs. I think one of the interesting examples here where we can kind of show a role for the federal government is in this kind of what's called the Valley of Death, which is bringing novel research from the lab and kind of giving it a pathway to market research.
And one of the things that the federal government has, and in particular, the kind of defense procurement establishment has been particularly good at is in jumping in at this kind of crucial stage in the R&D cycle and providing customers, providing a market for a lot of these technologies.
that then allow these companies to build up their production, build up their scale, reduce their costs, which then makes it feasible for them to go into the commercial market. You don't have the modern semiconductor industry in the way that you have it today without those initial purchases from the Air Force and from NASA and from some other government entities. It's just, you know, it's incredibly difficult to go from the lab to the market, especially when costs are super high. You know, it's hard to find, you know, a
consumer to buy those products when cost is high. And that's where I think that's one of the ways in which government is, I think, particularly important in pushing forward that cycle. We're talking about the role of federal government. We're talking about the role of federal government, specifically in early stage, right, basic research. There are examples where industry has taken this on. And I think, you know, it's particularly helpful to look at, you know, the AT&T system, the Bell system during the 20th century. It's a classic, classic case
of industry doing that, exactly the type of research that we're talking about. But there's a couple of unique parts or unique characteristics that I think made it particularly possible for the Bell system to do this. So you look at the Bell system throughout the 20th century, it's composed of a couple of key units. You've got its corporate headquarters up in New York. You've got its manufacturing arm, Western Electric, which is producing equipment that's going into telecom networks.
You have the regional Bell operating companies, which eventually become companies like T-Mobile and Verizon. They're the ones that are providing local telecom service or local phone service. You have long distance service. And then the kind of key nexus throughout all of this
is Bell Labs, right? There's been a lot of really good stuff written about the history of Bell Labs and the role of Bell Labs, you know, things like the transistor discovery of cosmic microwave background radiation coming out of Bell Labs, inventing the cell networks. I mean, you can kind of keep going on down the list of amazing technologies and innovations that have come out of Bell Labs.
But the reason why Bell Labs was able to invest in both basic research and applied research is because of the consistent funding that they had, the consistent funding stream they had from this was a government-regulated monopoly. In the 1910s, government basically said to AT&T, we're going to give you the ability to be a monopoly. You're going to own the telecom industry. And in exchange, you're going to...
you know, essentially abide by certain conditions. One of which was if you create a new invention or a new innovation that is outside specifically within the scope of the telecom industry, you have to license it and it has to be very generous terms or no fee licenses.
So if you look at how Bell Labs is able to operate, they have a consistent revenue stream coming from, they're essentially their monopoly on cell phone service. They have a direct relationship with the manufacturer. So they're able to talk to the manufacturer and troubleshoot as a product goes to the manufacturer. They're able to troubleshoot, is this, are there ways that we can improve it to improve the manufacturing process?
and they have a direct connection to their customers through the operating companies so they can see, you know, what is the market demand.
A lot of these characteristics, most importantly, the consistent funding stream, that's exactly what we're talking about when we talk about why federal research dollars are so important for basic research. It's because they have that long-term consistent spending, which allows you to invest in things where you might not see a deliverable for 10, 15, 20 years down the line.
You know, that's different than if you're, you know, investing in, you know, what most companies are looking for, which is, you know, maybe a couple of years of a time horizon. But, you know, they want some return on the investment, right? You know, justifiably so, right? They have to justify that R&D spend. And that's, you know, incredibly difficult to do if you don't have this kind of consistent market. So I think, you know, there's a world, there's a way in which, you know, industry does this. But it's important to look at the characteristics that enable that.
which is similar to the characteristics that enable basic research spend from the public sector. Divyansh, do you want to hit the sort of like universities role in all of this angle? Today, for instance, universities are not, you know, they're not just doing basic research. They're also doing a lot of applied research, a lot of industry funded research as well.
But the other piece that gets missed about the university's role today is creating this talent pipeline. Who is going to industry to conduct this industry research? That's the researchers who the federal government is funding at these universities.
So these researchers come, they come to graduate programs, they develop intellectual curiosity by working on curiosity-driven research, by finding interesting problems to solve, because there's not a pressure to find a revenue stream for something right away. And they build that intellectual curiosity, they take that, and when they join industry, ultimately, let's say, they
They are the ones who are driving the industrial innovation that we see. And so universities play a big role here. They also play a big role in regional innovation and economic growth in their regions. A lot of spillover effects, the startups that are created, the jobs that are created as a result of this research funding. You see several economic studies out there showing
Multiple dollars returned as a result of a dollar spent on federal, you know, as part of federal R&D spend on universities. You see how these universities are furthering national security objectives, as Alex alluded to as well. For instance, just look at how Carnegie Mellon, as an example, my alma mater, has been working with the DoD on several projects.
Projects that directly impact the warfighter. You know, universities host FFRDCs. They host people from the Department of Defense as part of the research program. So the universities are playing a broader ecosystem driving role here, not just universities.
You know, a tunnel vision role that is focused on quarterly profits. And there's nothing wrong with that vision. It's just those are two complementary roles. And so you can't just replace one with the other. There are some companies, as Alex mentioned, like most companies do not do basic research, but some do.
You can take the example of Microsoft, as we saw with the Majorna chip on Quantum, that they spent nearly 10 years developing. And that was a result of consistent funding stream. But if you dry up that consistent funding stream for universities, what happens is, well, you have fewer PhD students coming into universities. You have fewer PhDs being
Granted, you have fewer people joining companies like Microsoft and you have fewer people working on new innovations like Majorna. So it's a whole broader ecosystem issue. Alex, can you talk about how jealous the rest of the world is of this ecosystem that America has built for itself? Well, so what I'll say is this. I mean, I think a couple of decades ago, China in particular looked at the US R&D ecosystem and kind of said, yes, we'd love that. Can we bring that over here?
You know, and they've been doing, you know, quite a lot of work, both in terms of dollars spent, time, you know, high level attention up to, you know, Xi Jinping is constantly talking about the importance of basic research, the importance of improving China's STEM education system, the importance of talent as kind of the key enabler of not just China's technology development, but as, you know, a key enabler for China's growth as an economic and global power.
And a big part of that is looking at the U.S. kind of R&D ecosystem and trying to, in some sense, replicate it. Like Devyansh said, it's an incredibly effective way of generating innovations, at leading them to market, at generating dominance, first mover advantages in critical new technologies.
And it's exactly what we're seeing going on right now. I mean, if you look at what China is doing in its own R&D ecosystem, its education system, I mean, they're kind of focusing in three, I think, three main broad categories. So you have and I think, you know, it really emphasizes the generational investment.
Just to throw some numbers on the board for people, it's like about $50 billion a year the US spends on basic R&D, and then you get another $50 billion from universities and businesses, so basically $100 billion a year.
American market cap for technology firms, 80% of the world's global technology market cap, 80% of Nobel prizes or science Nobel prizes over the past 50 years have had one of the winners have a US affiliation. And like,
that all those three things are connected. So like you get the biggest, baddest, most advanced companies, which then sell the biggest, baddest, most advanced technologies to your national security establishment. And it's a, it's a beautiful self-reinforcing thing because like the, the best scientists are,
are in the American universities. So then the best students all want to come to the American universities and enjoy the funding, which is only actually 25% of the world's basic funding, but it's just spent in such a higher quality way than the rest of the world, such that you do get the greatest global minds wanting to come here and do work on these topics.
For sure. And I think it's, again, it's like, it's a generational investment, right? Like, it's not a coincidence that you have U.S. companies originally leading in the semiconductor industry and then U.S. companies originally leading in, you know, PCs and other competitive computer applications, you know, translate to, you know, the place we're at right now with AI, right? Like, they build upon each other.
So on the one hand, it's great because if you kind of get in early, if you're leading one industry, it kind of naturally builds, especially if you can continue to make these investments over the long term, it naturally builds into a great place for the next generation or next iteration of that. On the flip side of it, it also means if you make decisions now that underinvest in research or research
you know, in other ways kind of hamper the development of these kind of talent ecosystems, you know, the real impact you might not really see for 10, 15 years from now, but it's also, you know, once you start to see those impacts, it's kind of too late, right? You're talking about a long time coming before you can kind of really regain it.
So, you know, research investments now, research decisions now, whether it's, you know, grants or just, you know, graduate programs, all of this type of stuff, you know, it's the type of thing that's going to be, you're going to start to see the impact, you know, 10, 15 years from now.
All right. So that's our cheerleading session for the American R&D ecosystem. Divyansh, what has been happening over the past hundred days, which I think all three of us are worried about, may be risking the sort of, you know, world historical R&D golden goose that America has built since, you know, Vannevar Bush wrote a letter to FDR kind of imagining the
the glorious future which we could all live in thanks to the joint efforts of corporations, research universities and the U.S. government. Yeah, actually, I would reframe the question from just the last 100 days to like over the last year or so, because last year is when we saw NSF budget cut about by about 8 percent since the previous year.
The NIST budget cut about 13%. DOD basic research budget cut about 4%. What has happened this year is you started, for instance, early on in the administration, we saw NIH change its policy on F&A benefit and unilaterally changing it to 15%, which was challenged in court, then came
decisions on layoffs at several federal funding agencies. And more recently, you saw DOD and DOE follow NIH on capping FNA at 15%, which we can get into in a separate conversation. But you also saw NSF cutting about 400 grants that it had previously awarded, ending them
You saw the NSF director resign amidst rumors of potential additional layoffs coming at NSF. Former Trump appointee, mind you. Former Trump appointee, correct. And unanimously confirmed by the Senate. So the other thing we saw in the news was leaks from what the president's budget request might be.
where NSF could face about 55% budget cuts and IH could face similar cuts as well. So that is something that has been on top of mind for the research community for years.
both academia as well as industry i think uh i was reading this blog post that brad smith wrote on quantum recently where he talks about the importance of basic science funding to develop the workforce right so i think it is on mine for everybody uh
There was a letter that several former national security leaders, including former Trump appointees, including his former Homeland Security advisor and several others, signed to Congress laying out the impetus for funding basic science research in this moment. Because if you look back, as Alex was pointing out, when you compare with China, last seven years, China has increased its basic R&D spending by 10% year over year. That was the vision we set.
in the Chips and Science Act for us. We wanted to double our R&D, federal basic R&D spending over the next 10 years. And instead, we have not met the moment. There's about $50 billion or something like that of Chips and Science Act authorized money that has not been appropriated for the science part. So that is the moment that we are living in. Separately, on the talent side, I think there were some
There was some movement there where certain universities got letters or SEVIS terminations for some PhD students, which the administration then reversed.
And so I think that is also playing in the minds of several universities and researchers, like how will this broadly play out? At the global stage, you saw France and Australia try to capitalize on this, China too, by establishing specific programs for U.S. researchers who want to move to those countries and build there, conduct research there, provide a
long-term stability and funding and residency and whatnot. So I think this is kind of where we are. Well, you missed one, Divyansh. We've also got the war on specific universities with big fights with Columbia, with Harvard, where you just have, it's not just the 400 NSF things that said DEI and them getting canceled. It's also just like
You get your research canceled because you are, you know, you happen to be a Ph.D. student or a professor at Harvard. And that certainly happened. But, you know, it's it's kind of interesting. And in fact, as we are recording this, like President Trump just announced that he's going to name Secretary of State Marco Rubio as interim national security adviser.
and nominate NSA Mike Waltz for UN ambassador. So a lot is happening. What? Oh my God. Live reactions. Um...
Best possible outcome. That was the best possible outcome. No comment. I was ready for the Loomer National Security Advisor era, but I guess that might have to wait a year or two. The reason I point that out is because there's an important aspect that the National Security Council also plays in this conversation by emphasizing the national security aspect.
importance of federal R&D. And Alex knows this very well from his previous job on the critical benefits that brings. So... Wait, wait, this is why... We got to stay on this for a second. This is like... We haven't had this since Kissinger. I mean, this is like... Right, Alex? I mean, has there ever been another dual hat? No, I think Kissinger was the one and only. Yeah. I mean, I just... I wonder... There's been...
All this talk about like, like NSC reform, right? And like, it's not a fixed organization. And I wonder if this is a moment for some of that. But then again, like the, the Rubio state reforms were actually kind of underbaked relative to like what people were talking about and what even I expected him to do in terms of like structural organization. We have no idea what we're talking about. Um,
But I know this is we got to lean into this a little bit. I've never had like a breaking news moment. I mean, like, but it is very like there is a lot of Nixon energy, I think. And it kind of makes like I sort of see it. I think this is like I would be buying America on this news. Like, I'm I really think it could have got got a lot worse than Marco Rubio. So yeah.
I mean, as it relates to like the tech stuff, I mean, he's been like center stage on the importance of investments in technology, on the importance of, on the centrality rather of technology within the competition with China.
I mean, you go back to his time where I think they put out there's like a report when he was time on the Senate where he was like the maiden since he was looking at made in China 2025. Right. And I think, you know, again, it's it's it's all center stage there. Right. It's it's technology is is is the key the key battleground, if you will. And, you know, if we're going to use that metaphor, we can send it a step further and say, well, you know, researchers, scientists and engineers, they're the frontline soldiers, right?
Right. They're the frontline soldiers that are, you know, contributing to to, you know, U.S. power or what have you in this space. This is wild. I mean, there were there was like New York Times articles earlier this morning where some people were saying Waltz got fired because he was too much of a hawk. And I mean, I guess that's just someone leaking that they want that to be the narrative. But.
Rubio, it's weird because we do have this very odd disjuncture, right? Alex, between him as the senator for a decade plus and then him in his MAGA era, which is like, I don't know. I think JD feels it in his soul a little more than Marco does. And we'll just have to see. I don't know. Wow. Wow.
All right, back to science and technology. So before we get into the on the plus side stuff, Divyansh, like I think it is important to grapple with like what we are.
you know, what the university ecosystem is facing from both from a talent and from a funding perspective, because international students are an extremely important part of the funding mix because almost all of them pay
full ride um and then you have kind of the government uh government funding which is a really important um funding stream and uh the fact that like there are only really like 10 or 15 universities that have these multi-billion dollar endowments that would be allow them to kind of ride out a massive external shock like um your international students halving or
um you know all of your funding going away such that like you don't have to just like pack up um you know pack up shop and you know close or sell to a private equity firm or something um and uh
the sort of you you glossed over it a little bit um the um the immigration stuff but it was very scary there for a wide range of uh students who are worried to leave the country and i think there's like like even though the courts have reversed it for now my my fear is that this is something that is going to be lingering in like every parents around the world and a lot of um
you know, PhD students when they're thinking about where they want to set up shop. We'll see how this plays out. We'll start seeing the yield data come out from universities soon enough, and we'll see how many students accepted the offers. So because, you know, we're at May 1 right now, April 15th is the deadline for students to accept or decline their offers. So we'll start seeing that. But just looking at this, we...
We've had an 11% drop in international student enrollment since last year, between March of 2024 and March 2025. This is a similar trend we saw in 2016 as well, between March 2016, March 2017. So we'll have to keep an eye out for the data. But just generally speaking on the importance of this population, I think one thing that people gloss over is
60% of computer science, AI, PhDs are international students. About 50% of all STEM PhDs and masters are international students. So if you want to replace that by just domestic population, it's going to take a lot of time and it's going to take a lot of concerted effort.
However, if you look at the data, the number of domestic computer science undergraduates who are going for a graduate degree has not changed since 1990 or 1995, something like that. China has a population that's four times the U.S. population. They produce twice as many STEM PhDs, twice as many STEM masters, four times as many STEM bachelors.
Gone are the days when you could credibly say that their quality of STEM education is not as good or their quality of STEM research is not as good. They have both the quality and the quantity. The only way we can really match is by recruiting a rest of the world team here. Just based on the numbers. They have way more bodies to throw at hard problems than we do.
And so that's an important factor that we should be considering as part of this conversation. All right, Divyansh, let's come back to your on the plus side. What was that thought? You know, just look at what Michael Kratzios said, you know, at the Endless Frontier retreat just about 16, 17 days ago, April 14th, where he was talking about
how you know this is going to be a golden age for america golden age of america how we are in the early light of this new golden age and uh the american hope and the possibility of progress power of science and technology you know
He was talking about that this golden age is only possible if we choose it. And then he laid out exactly like how he wants to pursue this, right? Like how ours was the atomic age and how we have to fight to restore that inheritance. And he laid out really like, OK, we have to rethink how we do federal R&D spending, smarter ways to do it. And that's not to say that the Biden administration didn't do it. They did.
They did a lot of those pilots, but making those the primary way of doing R&D would be a great way that a great thing that this administration would do. And, you know, new experimentation, price competitions, I think those things will be really, really good. We have to really think about
how do we do this better? How do we get the most juice out of every dollar that we spend on R&D? I think he really talked about how we cannot just have a protect agenda
to maintain American dominance. We also have to have a promote agenda. We have to shape a funding environment that makes clear what our national priorities are, that we enable the scientists to create new theories and empower engineers to put them into practice, use advanced market commitments, and just multiply the impact of government-funded research here. So I think there was a lot of positive in that address that, oh,
gives way to a lot of new opportunity for the administration to really scale this up. I think now's the time if people have bold ideas to go forward with them. Yeah, I mean, it's sort of...
Like we saw the Doge playbook, right. Of the first hundred days. And there was none of the, Oh, let's actually do the reforms we need to unlock like the better version of government. It was just like, we're going to cut stuff, but the Doge era is ending. And, um, you know, there's like, it is, you can't like smash your way into creative meta science, uh,
reforms. But you do need to shake up these organizations. And, you know, the fact that there is this energy and insight and understanding that, you know, not everything is hunky dory. Like it, it is, it is harder and more expensive. And the breakthroughs have been fewer and farther between relative to the dollar spent than what you were looking at in the fifties, sixties and seventies, such that, you know, it's,
Now is as good a time as any to run experiments and try new stuff, but that is a lot harder with half as much money and without foreign talent, which is what we're risking by playing, you know, by playing the budgetary and visa and sort of like war on university type games that we've seen over the past few months.
Yeah, and I would say to that, my good friend Caleb Watney has a good saying here that if he were to see for a federal R&D from a VC lens, given the big presence of VCs within the government, you have to measure the rate of return on the investment. The return on your investment as a metric of your performance rather than
how little you invested. And that's an important factor to consider here. How do we get the most out of what we are investing? The other thing on just to your point about reforms and shaking up the agencies, I think the agencies are very much open to that. The agencies have the National Science Foundation created the tip directorate before Congress even passed Chips and Science Act. I
even when several other directorates were quietly not happy with that, right? The agencies are very open to this change. I think you look at how the Department of Energy National Labs are experimenting with OpenAI's models as like scientific peers for brainstorming. I think those are really interesting examples
things that research agencies are trying to do to shake themselves up, to rethink how they do research or how they do funding. So I think if the administration wants to pursue that path, I think they will find a lot of support from within the agencies themselves, from the universities and from industry.
To conclude our America section, my take here is the fan of our Bush endless frontier ecosystem over the past 75 years has delivered like
some of the greatest benefits to humanity that have been created in like the history of our species and um you know setting the national power stuff aside which i hope is kind of self-evident um from the fact that this is the thing that beat out the soviet union and created you know the richest country in the
Yes, it is a thing that could use reform, but it is like the golden goose of all golden gooses that we have been able to sort of, you know, through work and consistency and some luck, been able to cobble together. And it's such a, you know, national gift that...
I don't know. It's like one of the things that's upset me the most is the sort of risk that, you know, we could be walking through doors. We can't walk back because, you
ecosystems like this, if they're supported, have an enormous amount of staying power. But there are also fragilities that are baked into them when you think about the institutions and funding streams and talent that you need to continually be replenishing in order to be able to deliver the sort of results that you got in the past. So I don't know. We've covered this a lot over the past eight years, kind of following the whole
excitement to and passage of the Chips and Science Act and this moment where you did feel like there was a real bipartisan consensus to do more and push harder on basic research and then to kind of watch the sort of immigration stuff and the culture war issues and the Doge energy kind of coalesce around the
to the biggest risk that I think this ecosystem had faced in decades at least. It's just a really concerning thing and something that we're going to be watching closely in the coming months and years. Yeah, my take is American R&D is the envy of the world and we should honestly double down on it. The universities are a powerhouse here. At the same time,
they're not blameless in a lot of manners, right? And they need to do some self-reflection as well as to why our commitment to academia and to our universities is in question today. And I hope many of them will do that self-reflection and will come out better from here.
If I can, I would also just like to add, I mean, I think we've talked a lot about labs and academia and kind of higher education levels, but I think it's also just important to highlight, right, like the R&D ecosystem is even broader than that.
It includes community colleges, it includes vocational schools, technical training programs, right? These are, you know, pumping out, you know, the technicians, the ones that are actually running the equipment in these labs, which again, so, so, so crucial. I think if you look at some of these like big tech companies, like especially the ones that have, you know, significant manufacturing roles, most of their employees aren't PhDs. They're community college graduates or, you know, they've gone to some sort of technical training program.
There's actually a pretty significant veteran population in the semiconductor industry, like hiring a bunch of former mechanics because they've got the requisite skills. Again, not PhDs, but skills necessary to kind of maintain this equipment. And I think the last part of it, which gets at just the generational investment is K through 12, right? If you really want to boost the quantity and quality of graduates coming out of your PhD programs and your master's programs, it starts at those early stages.
We'll talk about what China's doing in that space later, but it starts at K-12 because it's about slowly over time building technical literacy to the point where when you get to an undergraduate program or a graduate program, your math and your science skills are already at par with where the rest of the world is at. I think the administration does recognize that, right? You saw the AI in K-12 executive order come out recently, which very much
tries to steer the education system to better incorporate AI education and AI in education through K-12 just to make sure that we have an AI-ready workforce over the next couple of years. So I think Alex is just spot on here. And this, again, goes to my optimism with
you know, the opportunities that lie ahead for us. All right, Alex, China. What are the long-term strategic intentions of the Chinese government when it comes to science and technology? I kind of like landed a while ago on what I think is the best way to capture it. And I'm going to paraphrase Matt Damon in The Martian. So I think it's, you know, Xi Jinping's planning to science the shit out of China. Like that's the tagline. That's his big approach. He's going all in on this bet.
I mean, I think if you look at whether it's the economy, the military, internal stability, I mean, everything there, there is tech through it all. You know, if you start looking at the economy, I mean, you look at the economy, it's, you know, this kind of his catch, his newest catchphrase is new quality, productive forces, which is, you know, a.
A new interpretation of a classic Marxist, Marxist-Leninist, you know, productive forces. But basically, it's how can we use tech to make the economy better? You know, if you look at the key parts of it, it's upgrading traditional manufacturing. Well, how do you do that? You install a bunch of robots. You automate it through AI. So again, tech.
What else is there? There's eliminating choke points, right? So it's trying to make China more self-sufficient by, again, investing in innovation, investing in R&D that Chinese companies come up with their own versions of technologies that they're currently reliant on foreign companies for. Classic case study here would be photolithography, where you have significant investment going into China's lithography companies, in particular SMEE, to replace their reliance on lithography systems coming out of the Netherlands.
And then I think the last part of it would be future industries. So I think it was last year, China identified in their future industries development action plan, they laid out like six categories, six broad categories and a bunch of specific technologies they were going to target for support. And I think...
I think you might laugh at some of the ones that they listed out because they sound a bit far-fetched. But on this list for them were things like humanoid robots, quantum technologies, artificial general intelligence, brain-computer interfaces, right? And they're serious about it, right? They are serious about leveraging all these technologies for the economic benefits because they view it as – I think they're very much viewing it as the U.S. –
to a point that Divyansh mentioned earlier, the US by leveraging the golden goose of its R&D ecosystem positioned itself to kind of dominate the big revenue generating sectors of the modern economy. And they want to be the ones to dominate those sectors moving forward. And so they're making those investments now. I've mentioned socially even, you look at some of the things that
that Chinese leaders like Xi are particularly concerned about, top of mind are things like social stability, things that potentially run the risk of undermining the party's hold on power. And how are they trying to boost their control? Again, it's technology. It's leveraging AI-based tools to improve surveillance,
Whether it's like facial recognition, gait recognition, voice recognition, predictive analysis. Again, it's technology-based solutions to monitor and control the population. Even things, you know, I think one of the things that often gets overlooked is if you look throughout Chinese history, the leading cause of revolutions, rebellions, you know, during the dynasty periods was famine and drought, right? Leading cause.
They, you know, Chinese leaders are very concerned about food security. They're very concerned about China's ability to feed itself. For the last, I think it's 22 years, the first document, the number one document that comes out every January from the state council and from the CCP Central Committee, it's a policy for the year on rural policy for agriculture. Again, it's because they're super concerned about food security. Partly it's because of just the lack of arable land, you know, pollution, all that type of stuff.
But again, their answer for improving grain output, farm output, it's technology. It's technology-based solutions to address all of these major concerns. And so, you know, if you look at, you know, what is China's plan, both moving forward, but also what their plan has been
From the reform and opening period up till now, it's been, let's invest as much as we can in technology and in science, and let's find a way of leveraging that for the whole host of things that we're trying to do. So yeah, science is shit out of China. Yeah. And just to add to that, just to add to what Alex was saying, they've openly said in many documents how they want to copy the US system.
I remember I was testifying last year to the Senate Energy and Natural Resources on some of this where the Chinese 13th five-year plan, for instance, explicitly calls out like Argonne, Los Alamos, Lawrence Berkeley National Labs as like crown jewels of U.S. innovation and how China wants to mimic the U.S. national laboratory system more.
to basically focus on national goals, strategic needs, target international technological frontiers, all the things that Alex was pointing out. Yeah, Alex, do you want to sort of tie that to the basic research ecosystem?
Yeah. So, I mean, I think all the things that I'm talking about, all the things that China is trying to do, it's all fundamentally based in the basic research ecosystem and in the talent going into the basic research ecosystem. Like I said, I think we said this earlier, but you even have Xi Jinping talking about how basic research is the foundation of China's technological progress and how talent is the key enabling factor, enabling their ability to develop. I think there's a
There's a recognition within senior leadership circles in China that for them to succeed in particular in dominating these kind of future industries, these technologies of the future, well, it starts with investments in basic research. I know they're facing challenges and actually following through on those investments and shifting their investments away from applied research and toward basic research, just given kind of the longstanding investments that they've put in that space.
But there is a broad recognition that if you want to be competitive, if you want to be a global technology leader, you have to invest in early stage research, you have to invest in basic research, and you have to, importantly, train the people to be able to go and staff those facilities. I think when we talk about basic research, often we end up talking a lot about building out the infrastructure, whether it's buying up GPUs or building up data centers to train models.
But fundamentally, you can build out the best infrastructure in the world. You can fill it with the best equipment in the world. You can give it as much money as you want. But if you don't have the people that know how to use the equipment and they know what to do with the equipment, then it does nothing for you. It all inevitably comes down to, do you have the people and do you have them collected enough, working together enough with the right training, with the right experience, with the right connections to really be able to push science and technology forward? All right, Alex, what's your take on the line that
China is looking for good enough technology as opposed to like, you know, Nobel Prizes and really frontier stuff. So, I mean, I think it's it's essentially China's doing both. China has a a different interpretation of what it means to be a to be the technology leader than than the U.S. does.
I think the US defines technology leadership as who has the most advanced technology, who is the one leading cutting edge research. I think China defines technology leadership as that, plus who dominates the markets, who owns most of the world's markets for these key technology products. And for that key second part of their definition, you don't necessarily need the most advanced technology. What you need is technology that achieves, let's say 80% of the capability at let's say 80% of the cost.
Because, again, when you're looking at, you know, dominating markets for a lot of that, you're looking at sub-Saharan Africa, you're looking at Southeast Asia, you're looking at Latin America, you're looking at places that aren't necessarily capable of affording the most advanced technology, but still want the benefits that advanced technology can provide. And I think, you know, if you want to look at a classic example of this, let's look at Huawei circa like 2019. Yeah.
Huawei, for a very long time from its early stages, it wasn't seeking to be the world leader in telecommunication equipment. It was seeking to develop technology that was just good enough, and it was offering it to countries that couldn't afford the best that the U.S. had to offer. And then eventually, once the U.S. got out of the telecom equipment game, the Europeans had to offer. But it was offering it at discount, right? It was offering it with generous financial incentives and just generally cheaper costs, which...
They could do that because of all the government money and subsidies and state support and all this other kind of stuff. So if you kind of track the rollout of telecom networks through the 2000s and 2010s, where you start going from 2G networks, 3G networks up to 4G networks, Huawei's kind of secret sauce, its niche area is going into essentially emerging markets in the global south.
offering technology that's maybe just not as good, but it's cheaper. And then, you know, getting their foot in the door in all of these network build-outs. What that gives them is significant revenue and market share, which they then turn around and invest in research and development.
That then positions them so that, you know, circa 2019, they not only own the world, most of the world 4G network infrastructure, but they've also been able to leverage all of that money they've been making by selling to those networks in places like sub-Saharan Africa and investing in 5G technology, which at the time was both better and cheaper than what, you know, competitors were offering. And, you know, that put...
the U.S. in a very difficult spot because it found itself in, for the first, I would argue, for the first time in its modern history, where it was at the, you know, the precipice of a major infrastructure buildout, critical technology, critical infrastructure buildout. It didn't have a U.S. company and it was facing a Chinese company that was just as good and was cheaper. And so that's when we talk about good enough, you know, I think it's about
Just broadening our definition of what we say it means to be a global technology leader and emphasizing it's not just about the most cutting edge, but it's also about scale and presence in markets throughout the world.
Yeah, I mean, Ava Doe, who wrote The House of Huawei, is on maternity leave, but we're going to get her on at some point. And I think like, you know, one of the fascinating lessons of that Huawei story is like, even though the government was pushing Huawei,
to get firms to do more R and D like that was a Ren Zhengfei decision. That was not a government decision to spend. What was it like 50% of, of, of profit on R and D or something like that. And it was revenue on R and D, like some absolutely absurd number. And this was an outlier very much like compared to his rivals in China. ZTE is like, you know, 5%, 7% or something such that, um, you know, it's this, it's this interaction between yes, the government support, um,
Yes, the sort of domestic scale, but also just like really incredible sort of visionary founders.
For sure.
Totally. Right. And I mean, I think another key point, which we mentioned this earlier on kind of that that that interconnection between research, customers and manufacturing. Right. That's again, that's that's what Huawei has built for itself. It's looked at the U.S. at success stories in the U.S. and elsewhere. And it's built that model for itself. Right. It's it's a manufacturer. It's a designer. It has a company.
a secure domestic market, which is where government support is particularly crucial, especially in Huawei's early days. But it has this interplay that makes it a very successful, very effective model for pushing forward R&D. At the other point, that's obviously a look back. But right now, we're seeing a lot of the same dynamics play out on a whole host of sectors.
Where if we're not careful, we can find ourselves in the same exact position later this year, two years from now, where again, we're seeing critical infrastructure sectors being reliant or being either reliant on Chinese technology or having to choose between choosing a Chinese company as your supplier or having to pay more money and be slightly less competitive by going elsewhere.
Yeah. I mean, I think that's something that like the broader American media political ecosystem is only starting to process is like China will be ahead in major commercial technologies over the next five years. You've already seen it in drones. You already saw it in telecom.
You already see it in electric vehicles, but like this is like regardless of where kind of like the Chinese macroeconomic environment or the American science funding environment is going in the coming years, like we're entering a new dimension, a new dynamic where like this sort of.
Like the trade-offs of keeping these technologies out of the US, which is like broadly what we've decided to do for drones and for sort of for drones, but for telecom and for cars at least is like, you know, we're creating this like weird, you know, captured like...
you know, I don't want to say like it's Japan in the 1980s with a 1990s with cell phones, but like there, I think another part of the story in China was the export discipline. And the fact that a lot of these firms, at least in their early days did really have to compete to get at scale domestically, both internally with other firms from different provinces and with, with them, you know, fighting against Apple and with Tesla. And, um,
You know, one of the sort of things that I think U.S. from a policy perspective needs to keep its eye out is like just because we don't see the cars here doesn't mean they they don't exist and they aren't getting better and they aren't winning over third markets and they aren't forcing GM and Ford and Tesla to to get their shit together. So yeah.
I don't know, that was a lot of domestic policy, Alex. No, no, and even if what a Chinese company is offering right now isn't as good or is more expensive or whatever as what a U.S. company is offering right now, they've consistently shown that they're able to
leverage legacy technology to then eventually move up the stack so that they find themselves, they put themselves in a position where they can get to world leading technology. But I think just so much focusing so much on where are they right now and how good is what they offer right now, I think misses the risks in years to come of their ability to
To, again, leverage good enough technology, leverage legacy technology to eventually generate for themselves, either through theft or innovation, world-leading technology. Divyat, do you want to say anything about this? Just look at where Huawei is now from, you know, where it was in 2019. As Alex was pointing out, right, it's...
going everywhere. It's building data centers, it's building cell phones, laptops, it's operating undersea cables, it's invested in EVs, it's not just a telecom company anymore. Same is true in many other scenarios for many other companies. And so to Alex's point that we have to be careful and we have to try and project it out rather than just look at a static moment in time is extremely critical.
especially as we try to take more protect and promote actions. Where do they want to be? And you don't even have to assume. They've all laid it out. You know, Made in China 2025 was not a hypothetical document. They met every single objective there. The AI 2030 plan was not a hypothetical document where they said they would be in 2025. In that document, they are there. So they're very open about where they want to be.
And we're just sometimes overconfident about how much our lead is. And, you know, we underestimate, we have an uncanny ability of underestimating China's ability to out-hustle everybody. And so I think that is something that people should be careful about.
you know, when I was fundraising for the China talk Institute, which now exists and is still taking donations, we're doing great work around, uh, uh, Chinese AI biotech and strategic competition. Um, uh,
A number of funders asked me, doesn't the IC just kind of like have all this like China technology stuff covered? You know, what would you say to them when they're thinking about what the IC can and can't do and the utility of folks kind of writing about this stuff independent from the government and publicly? Fundamentally, we're talking about commercial technologies. We are talking about actors that are commercial entities, right? Companies, academia, right?
We're talking about, I think the other part of it is we're talking about industries that for a lot of them are relatively, everybody knows each other, right? Like key players know each other. And so, you know, there's an inherent need because of that for things like China Talk or just generally speaking, right, for a key role for industry, for all of these players, right?
for a couple of reasons. I think one of them is so much of the key insights that, you know, might be technological in nature, but can have significant strategic implications are rumors circulating within industry or, you know, insights that a company is gaining from talking to a customer or talking to a partner, whether it's in China or elsewhere. And so there's significant value to be gained in that regard. And I think the other part of it is
Unlike looking at something like the Chinese military or Chinese leadership, which is I think a very different question, the targets that you're looking at and the developments that you're looking at fundamentally is just a different target set. And so I think there has to be a role for entities or just generally pulsing the open source ecosystem
Otherwise, you run the risk of kind of really missing significant trends and developments. And the last point I'll mention there is, you know, I think we spent a lot of time talking about why China is different and how China is different and how Chinese companies operate differently. But they also have that kind of profit motivation, even if it's somewhat reduced or if they can get around it with when they need to.
And so, you know, a lot of these developments that companies are doing, they're messaging it, right? They're messaging it in industry news, in industry press. They're messaging it kind of within the ecosystem because they want to be seen, right? They want to push out that message of technological advancement and development. And so there's so much out there in the open source that you can find, right?
that, again, small technological developments, but massive strategic implications. Something as like China making progress toward more advanced node semiconductor manufacturing, super technical, but massive strategic implications for US export controls, US AI policy. Yeah. I mean, I think my answer to that is, if you're looking at the sort of military to
Like maybe it's 80, 20 or 90, 10 on the military side. The interesting stuff you got to like hunt and dig in a intelligence community way for. And I would assume that like the inverse, it's the inverse on the commercial side where
like 90, even 95% of what you would need, maybe not in a very specific tactical way, but at a larger strategic level of like, what does this mean for America? What does this mean for industrial policy? What does this mean for science and technology policy? You know, you, you can get by just reading stuff and like the, I think like the relative, um,
openness of the Chinese media ecosystem when talking about commercial technologies versus talking about, you know, operational plans of like, you know, how we're going to shoot down American satellites or whatever is just an entirely different ball game because, you know, they need to, um, you know, win domestic market share and hire people and get folks excited to work for different companies and yeah, raise money from investors and all that is, is, uh,
happening under like a journalism ecosystem, which is like, for the most part, pretty free, which is a cool thing to get to sort of surface if you have the language skills and context to be able to sort of process that and share with an English speaking audience. Yeah, I mean, you know, China's war plan for tech is, you know, that's their industrial policies that they're releasing constantly. Like Made in China 2025, it was a very specific policy
tech dominance plan and it you know it got down into the detail of you know we they want to they want to control x percent of this industry or produce x percent of these components right i mean you can't get more detailed than that so so um
Before this episode, Alex, I asked you if there was one document you wanted to talk about, and you pointed me to this mobilization war plan. Why don't you, we'll link it in the show notes, but why don't you give listeners some context on why you think it's interesting and important?
Sure. So it's from back in May 2022. And it's a transcript that leaked from a tabletop exercise, like a war mobilization tabletop exercise in Guangdong province. I mean, what's interesting about it is it's, you know, representatives from the party, from the military, from the government. And it kind of brings together, you know, all these folks in the room. And it says, you know, basically, we've decided to invade Taiwan. Like, what does the province do?
Not necessarily from a like, you know, moving ships in this, you know, naval vessels in this way or that, but more, how are we going to leverage? How are we going to mobilize the population? And I think relevant to this conversation, relevant or mobilize the economy and industry.
Um, and it was, you know, in very detailed fashion, it went through and it talked about mobilizing and converting civilian manufacturing industries, commercial manufacturing industries over to wartime production. So it called that in particular, the shipbuilding sector called out drones, called out other high tech industries. Um, and I think it's, it's, it's a, it's a fascinating example of, um,
China preparing for, you know, pairing and talking through the the kind of steps that they would need to take in the event that there's, you know, a major war with the U.S.?
not at the national level, but at the provincial level, and thinking through how do you leverage an economy. And it's kind of ringing bells for being similar. It's basically the US's approach to war mobilization during World War II. That's the scale. That's the framework. It's leveraging the benefits of China's decades-long investment in expanding manufacturing capacity
to, you know, essentially outproduce the U.S. in the event of a conflict. And so, yeah, it's a great transcript that's kind of, you know, surprisingly very detailed and kind of goes through pretty much all the kind of considerations that they're thinking about, whether it's, you know, mobilizing reserves, recruiting people, mobilizing maritime industries, you know, aerospace, repair yards, like militias, everything is in there. And...
when I read this, I was like, Oh, this is like kind of cute, but maybe it's sort of like cosplay like, and that like Americans, like, you know, we do like nuclear war games for fun, I guess. Um, and you know, there, there's something about, uh,
the sort of history of the party in the Chinese system where like national mobilization is like the coolest, most exciting time to be alive. But your sense is that, no, I shouldn't just take this with a complete grain of salt. Convince me otherwise, Alex. Yeah, I mean, I think if you look at, you can even find this in like local Chinese press, which I was able to just by like literally asking ChatGPT to find articles for me.
So you can actually see other examples in local press at the county and prefectural level, as well as at the provincial level of them essentially like running similar exercises or at the very least mentioning like prefectural and county level national defense mobilization committees. Yeah.
It's just, you know, it's part of their, there's this big national system called the National Defense Mobilization System, which essentially puts these kind of cross-party government military committees at the national, provincial, county, prefectural level, right? So there's an interesting example in November of 2020 in Chongqing municipality where they conducted a mobilization exercise and they actually had like civilian manufacturing companies temporarily switch over their production lines.
And the example here was ATVs, which, you know, not super interesting. But again, it's flexing the muscle. It's testing it. And, you know, to put it in the U.S. context, right, county prefectural level, like that's the equivalent of, you know, folks over in like Fairfax County to be local to D.C., like the city of New York, thinking through how it would mobilize in the event of a conflict to support a national level, you know, war in the Pacific. Right.
So just talking about kind of the scale and the depth of preparation and building out the system. All right. So for our next episode on China Talk, we'll Twitch stream America's national mobilization war plan. Divyat Alex, thank you so much for being a part of this rather loopy, but hopefully informative edition of China Talk. Thanks for having us. Thanks for having us.
♪
There's antimony, arsenic, aluminum, selenium, and hydrogen, and oxygen, and nitrogen, and rhenium, and nickel, neodymium, neptunium, germanium, and iron, americium, ruthenium, uranium, europium, zirconium, ruticium, vanadium, and lanthanum, and osmium, and astatine, and radium, and gold, and propactinium, and indium, and gallium, and iodine, and thorium, and thulium, and thallium. There's yttrium, ytterbium, actinium, rubidium, and boron, gadolinium, niobium, iridium, and strontium, and silicon, and silver, and samarium, and biseth, bromine, lithium, beryllium, and barium.
There's homium and helium and hafnium and erbium and phosphorus and francium and fluorine and terbium and manganese and merconium, olivdenum, magnesium, dysprosium and scandium and cerium and cesium and lead, praseodymium and platinum, plutonium, palladium, promethium, potassium, polonium and tantalum, technetium, titanium, tellurium. Ah!
and cadmium, and calcium, and chromium, and curium. There is sulfur, californium, and fermium, berkelium, and also mendelevium, einsteinium, nobelium, and argon, kryptonium, radon, xenon, zinc, and rhodium, and chlorine, carbon, cobalt, copper, tungsten, tin, and sodium. These are the only ones of which the news has come to Harvard. And there may be many others, but they haven't been discovered.