00:00:06
5 4 3 2 1 zero. All engine running.
00:00:14
Liftoff. We have a liftoff. That's one small step for man, one giant
00:00:20
lift for mankind. The world's largest airliner. Each wing is big enough to hold five tennis
00:00:26
courts. This new technology made it possible to meet the user's crucial needs. Enter the computer and a new age.
00:00:33
What a computer is to me is it's the most remarkable tool that we've ever come up with. And it's the equivalent of
00:00:40
a bicycle for our minds. Here I am playing a game of chess with a
00:00:45
computer which is analyzing board positions and and applying a certain kind of intelligence to figure out what its next move should be. That's the
00:00:52
subject of our program today, artificial intelligence.
00:00:58
[Music] The good future of AI is one of immense
00:01:03
prosperity where there is an age of abundance. Everyone can have whatever
00:01:09
they want. We're still in the very early innings of AI. I would say the rate of progress is
00:01:14
exponential right now. Every time I think that we are overstating the impact of artificial intelligence, something comes along that
00:01:20
tells me we aren't making enough of it on the show. You know, there's no 60-minute clock on this thing. This is
00:01:26
an infinite game. Think about solving a problem that would take humans thousands of years to solve. Those who can harness and govern the
00:01:33
things that are technologically superior will win and it will drive economic
00:01:38
vibrancy and military supremacy. The Trump administration believes that
00:01:44
AI will have countless revolutionary applications. We believe that America's destiny is to dominate every industry
00:01:51
and be the first in every technology and that includes being the world's number one superpower in artificial
00:01:58
intelligence. It feels like every tech revolution of our lifetime has been leading to this
00:02:03
moment. [Music] [Applause]
00:02:09
[Music] See, you're going there.
00:02:16
All right, everybody. Welcome to winning the AI race. Uh, this is our first event
00:02:22
in DC. Can I can I get permission from our leader to sit down? Yes, you're here. You may sit. You may
00:02:28
sit. Thanks for coming out, everybody. And, uh, we put this event together in just a
00:02:34
couple of weeks uh, in order to have a really important discussion, winning the AI race. This is something America has
00:02:41
to do and it's something we will do and we're going to do it through the way we've won every other technological race
00:02:48
through grit, entrepreneurship, uh, and dogged competition. The difference with this administration is
00:02:54
they're actually engaging with the technology industry. And, um, today we're bringing together many members uh,
00:03:01
or all members of the administration here to talk about it. And none of this would have been possible uh without our
00:03:08
bestie David Saxs deciding that he would take um some time and become our zar of
00:03:14
crypto and AI. And I would like to just start with a huge round of applause for
00:03:19
David Saxs. David, you've been here for
00:03:26
six months. I'm sorry, but did you actually prepare? This is excellent so far. No, I'm just speaking for
00:03:31
No, it's excellent. Keep going. Keep going. He wants to be invited back to DC.
00:03:37
They told me I've got 12 hours left on the ground. I think that White House tour is going to happen after all.
00:03:43
It might just happen. Might just happen in the air. But in all seriousness, u you've been here for 6 months and we all know how
00:03:50
capable you are. Uh but my lord, uh this administration is on a heater when it
00:03:56
comes to crypto and AI. I am absolutely and I think I speak for everybody in our
00:04:01
industry thankful and wildly impressed but not surprised at the pace at which you've led crypto and AI. What's the
00:04:08
first six months been like for you? It's really been incredible. I mean I never expected to go into government at
00:04:14
all and uh really as a result of President Trump coming on our podcast uh a year or so ago. Uh that began uh a
00:04:21
relationship that you know eventually led to uh me being offered this this job
00:04:26
and I took it because I just thought it was a once in a lifetime opportunity to work for a president who really wants to get things done for the American people
00:04:32
and you can see that he that just every day he works so hard uh to to push forward his agenda for the American
00:04:38
people and I think uh AI and crypto have just been two of those issues. Uh but it's been a lot of fun to work on these
00:04:44
things because we are getting a lot done. Yeah. last week and this date, this date today, you put we put together in just
00:04:50
like the last 10 days as an opportunity to talk about your action plan that was getting the
00:04:55
president's action plan getting published today. But we we should invite Jacob out because Jacob partnered with us from Hillen Valley from Jacob
00:05:00
Hellberg. Uh come on out and join us. Yep. Our our new fifth bestie. Yeah, Jacob Hellberg.
00:05:07
There you go. Nice to see you, brother. Good to see you. and um
00:05:12
good to see you. Good to see you. Freeberg, your team and Jacobs uh put a ton of work into this and
00:05:20
we have a lineup that is just absolutely outstanding. So, thank you Jacob for the
00:05:26
for the hard work from your team and Freeberg the hard work from your team to put this all together. Maybe you could
00:05:32
give everybody an idea of the questions we want to address today and what the format's going to be.
00:05:38
Absolutely. So um the Hill and Valley Forum is a community of technology of
00:05:43
builders and policy makers who believe that technology is an engine of wealth creation and is an indispensable pillar
00:05:51
for American national security. And it was incredibly exciting to have the opportunity to engage in this event
00:05:56
which is actually going to cover um a lot of the topics that everyone in our community cares about. Um, ultimately we
00:06:04
believe that the and I actually said this in my confirmation hearing not too long ago that uh we're at an inflection
00:06:10
point. We are in an AI race and uh so the different parts of the programming
00:06:15
today will be a series of conversations that will cover the different facets of how technology will actually create
00:06:21
wealth for our country. I think is like very important as we talked about who do we want to have on stage and how do we
00:06:26
want to talk about the uh the president's action plan that David shared with us was to highlight that
00:06:32
there are new industries being created because of AI industries that couldn't have existed a decade ago and so we've
00:06:39
got a couple of those examples and then there are these industries that are enabling and accelerating AI and that's
00:06:46
mining energy chips fabs and data centers so we've got conversations
00:06:51
across each of those that's kind of this enabling conversation and then fortunately uh we've been able to get a
00:06:56
lot of folks from the administration to join us today to talk about the government's role in enabling this economic transformation
00:07:02
that's already underway and I just want to say one point I think what what's become apparent to me and I think is
00:07:08
wrong in the press narrative today is that AI is destroying jobs I think what we are seeing on the ground is an
00:07:14
incredible job creation engine that's underway and so I think it's very important to highlight that and share
00:07:19
that those stories because they're not told enough and I think there's a real opportunity to kind of bring them to light and that's hopefully what we can
00:07:26
kind of get through today. And Chimath just uh coming around the horn here. Doesn't matter if you're a Democrat, Republican, independent,
00:07:33
moderate, this issue transcends party. Uh this is the issue of our lifetime and there's a lot of hard questions and a
00:07:39
lot of hard debate. maybe you could just speak to this administration's ability
00:07:44
to bring in a lot of disperate opinions uh and work together across the aisle
00:07:49
and with all members of the industry. I mean, look, I think historically you've had a fork in the road where you
00:07:56
can view technology as either optimistic and glass half full or pessimistic and glass half empty. The optimistic glass
00:08:02
half full view says that the country that can harness AI or any of these leading critical edge technologies is
00:08:10
able to garner most of the gains and then those economic gains can be spread. Then it's a debate about how to spread
00:08:17
those gains within a country and within economy and then from there with economic supremacy you also have
00:08:24
military dominance and now you're a superpower and you remain strong. The problem is that historically we've gone
00:08:30
in the other direction where there has been this mistrust and in that mistrust you've had global competitors emerge and
00:08:37
create I think real fundamental existential risk for our place as a superpower.
00:08:42
Well said. So the great thing over these last you know frankly six months has been a massive pivot back
00:08:51
into this idea that America is the best. We should not be ashamed of the things that we've created and these incredible
00:08:57
technologies and these incredible people should be celebrated. Yeah. And let's go and win the race.
00:09:03
Let's win. Okay. And by the way, how we're going to win the race. The action plan. Uh Sax, I know you
00:09:08
invited Michael Katios to join us here today. Director of the Office of Science and Technology Policy. Should we have
00:09:14
Michael come out? Yes, Michael come out. Yeah. Michael, come on out. Please welcome Michael.
00:09:19
[Applause] All right. How are you guys? see it. Oh, thank you. So,
00:09:25
so I'll I'll kick this off. Um, so President Trump in his first week in office uh uh signed an executive order
00:09:34
that directed us to create this action plan. Uh Michael and myself and and the national security adviser and the
00:09:40
objective was to figure out how the US would dominate in AI. um from his first week in office, President Trump has made
00:09:47
this a priority and we see it we do see it as this global competition or global
00:09:52
race and the consequences of losing that race would just be unthinkable because AI is going to have such huge
00:09:57
ramifications for our economy and also for our national security. So, the United States has to win it. And um
00:10:04
working with Michael and the Office of OSTP, we we put out a plan today that has 90 concrete actions that at least
00:10:12
the executive branch can take to help us win the AI race. And I want to call on Michael in just one second, but I'm just
00:10:18
going to outline the three big pillars of the plan. So, number one is innovation. There's just no substitute
00:10:23
for innovation. You have to out innovate your global competition. You can't regulate your weight just to winning the
00:10:28
the uh the AI race. So number one is we we have a lot of things in the plan that are going to help our private sector,
00:10:34
our startups or our um tech community uh in out innovate the competition. Number
00:10:40
two is infrastructure. We have to have more and better AI infrastructure, data centers, energy, manufacturing in the
00:10:46
United States. And number three is the AI ecosystem. We want to have the the biggest ecosystem. We know from Silicon
00:10:53
Valley that the companies that create the biggest ecosystems are the ones that win. You have the most developers on your platform. you have the most apps in
00:10:59
your app store. Those are the companies that create, you know, uh those are the companies that dominate industries. In a
00:11:05
similar way, the United States has to dominate by creating the the AI stack for the entire world. So, those are the
00:11:11
three big pillars of this plan. Let me call on Michael. Can you I guess tell us how you know I guess maybe speak to the
00:11:17
process of how this plan was created the last 6 months. I know your office did a ton of work on this and then I guess if
00:11:22
you want flesh out some more of the important details as you see it. Yeah, absolutely. Once the president signed the executive order assigning us
00:11:28
this task, the first thing we did was actually an issue issue an RFI, very exciting government activity. And we
00:11:34
asked the country, hey, what should we include in this plan? And I think to be honest, I think we're all surprised with what came back. We had over 10,000
00:11:41
responses come from all corners of the country. We had Hollywood actors sending us responses. We obviously had tech
00:11:47
companies. We had everyone you can imagine. And I think it really showed how impactful this particular technology
00:11:52
is to everyone in every industry in the US. So we ingested a lot of those comments went out to all the uh agencies
00:11:59
that work with and and in some ways touch AI and came together with this with this plan. Now if you think about
00:12:05
it um you know there's been a lot of national strategies that companies put that countries put out there over the
00:12:10
last five or six years and what we really wanted to focus on um is in the title itself an action plan. We wanted
00:12:16
things that we could accomplish in the next 6 to9 months to accelerate and ensure that we can win this race. So if
00:12:21
you think about the first pillar which David talked about which was the innovation pillar, what's really key about innovation is we want the next
00:12:28
great AI discoveries to continue to happen here in United States. We have to create an environment that allows that
00:12:33
to happen. And when we talk about deregulation, the way I like to think about it is, you know, you can't create,
00:12:40
there's never really going to be a law that says, hey, this is how we regulate AI. What ultimately is going to happen is these AI technologies are going to be
00:12:47
built into so many other technologies. Whether it's drones flying self-driving cars, whether it's FDA approved AI
00:12:54
powered medical diagnostics, all these different agencies are going to be touching technologies that are powered
00:12:59
by AI. And it is incumbent on us to create a regulatory environment where these technologies can thrive and not be
00:13:05
hindered by the government. The next piece of innovation I think is really key is using the power of and the data
00:13:11
that the government has to drive scientific discovery through artificial intelligence. You know we have seen in
00:13:18
this first wave of AI great great progress in the way that LLMs are able to handle coding for example but we can
00:13:24
do so much more than that. There's incredible data sets department of energy has for example at their national labs that can help power a lot of next
00:13:31
generation discoveries and things in material science in medicine and that's what this AI plan calls for and and
00:13:37
drives. Um the next pillar which is about infrastructure, people talk about this all the time and
00:13:43
it's it's about how do you create a regulatory environment that encourages and actually accelerates the ability of
00:13:50
our power generators and our chip builders to be able to do what they need to do here in the United States. We can
00:13:57
uh plan calls for categorical exclusions for AI related activities which can
00:14:02
allow data centers and other power generation to happen on federal lands. And that's going to be coupled with all
00:14:07
sorts of other efforts to really accelerate the velocity that we can build power and ultimately run these data centers. So let's talk about before
00:14:15
we run out of time one of the most important issues which is the talent wars. Yeah, we um going to stay focused on AI
00:14:22
here. We'll leave the border and deportations off the table, but we'll talk about something super important
00:14:27
which is recruiting talent from around the world. This administration, we've gotten different signals and obviously it's a
00:14:32
very controversial issue here in the United States. What do we have to do in terms of immigration and let's just call
00:14:38
it recruitment because that's really what it is recruiting the best and brightest from around the world to come work on our team as opposed to say team
00:14:45
China. What do we have to do? What is the administration's philosophy on recruiting the world's best AI talent
00:14:52
in the action plan? I think what we bring to light and I don't think it's talked about enough is to power and
00:14:57
successfully drive continued American leadership in this domain. It is not simply about having the greatest a
00:15:04
engineers but is also having all the other parts of the workforce which needs to drive this forward. You know we
00:15:09
talked to some companies like Caruso and others who are building these large infrastructure builds around the US. The
00:15:15
challenge that they're facing is in electricians and HVAC talent and the and the AI plan itself spends a lot of time
00:15:20
and energy directing um various agencies whether it's department of labor and others who have these reskilling and and
00:15:27
programs to sort of train these people up to be able to fill that void. So for us it's about attracting here to the US
00:15:33
the greatest scientists and engineers but it's also to be able to train the American workforce to be able to do the
00:15:38
necessary jobs put that forward. Michael um what's the philosophy going forward on the thing you mentioned just
00:15:44
before this which is there's these enormously valuable data sets that sit inside the DOE that sit inside of FDA
00:15:51
where presumably if we made them available to private industry particularly American private industry
00:15:56
the gains could be incredible. Is that an open-source philosophy? Is that a licensing philosophy? How do you think
00:16:03
it should best serve the American economy to get this stuff out there? Generally, the government has taken an open- source approach to this. And the
00:16:10
general challenge that we've seen over the years is there's been a lot of lip service to, hey, let's unlock data for
00:16:15
the American people. And the main challenge is, and for all of us who are in AI, the format of that data itself
00:16:21
actually matters a lot. if it's like dirty, nasty data that isn't homogenized
00:16:26
in any way, it's not particularly helpful. And I think that's going to be a big effort that the DOE is going to
00:16:31
Department of Energy is going to try to do to make this better and possible. And what was great in the uh recent
00:16:37
legislation that was passed in BBB was and actually $150 million ticket to the
00:16:42
Department of Energy to build an AI for science program that very much is going to be working on this exact problem.
00:16:47
Should there be federal preeemption on AI regulatory um schemes? So, there's
00:16:53
been a conversation about doing this to ensure I think right now there's over a thousand state laws that have been
00:16:59
proposed or passed that have some regulatory effect on AI and tech related technology. Should the federal
00:17:05
government preempt all of that and raise it up? I think generally preeemption is an issue that comes up very often broadly
00:17:10
in technology. You have this issue with uh privacy for many years. What we're trying to face today and what we talk
00:17:15
about in the plan itself are actions that the executive branch can take itself. And a lot of preeemption discussion revolves around what Congress
00:17:22
can or can't do. So we don't necessarily lean hard on that because we focus on things we can accomplish. Right. And just to just to add to that, so it's
00:17:28
true the the action plan doesn't speak to that issue um Freeberg very much. But I I do think there is a real threat to
00:17:35
national security that's brewing by virtue of the fact that like you said, we've got a thousand bills going through
00:17:41
state legislatores right now, all regulating AI in different ways. If this continues, we're going to have a
00:17:46
patchwork of 50 different state regulatory regimes as opposed to one seamless national network. And look,
00:17:51
China is they've declared that AI is a national priority for them. It's they understand how strategic it is. And I
00:17:58
think if we hobble our AI innovation with a patchwork of 50 different state regimes, I think it's going to hurt us.
00:18:04
So I don't, you know, we weren't ready to declare a policy yet in the action plan, but I think it's something that's
00:18:09
going to have to be looked at uh over the next year or so. Thanks for thanks for joining us, Michael. Everyone, the director of the
00:18:15
Office of Science and Technology at the White House. Well done.
00:18:21
Thanks. Great job. Pleasure. Thank you everyone to the besties in the Hill Valley Forum for the warm welcome.
00:18:28
Uh I'm Chris Power, the founder and CEO of Hadrien and I'm here to talk to you today about our company. The mission is
00:18:34
to re-industrialize America. We do this by building AI powered factories in the United States.
00:18:41
So you might ask why is this important and why should you care about manufacturing in the United States? Well, what I realized before I came to
00:18:46
this country is that we're in a global race. So every great nation gets built by
00:18:52
having the best industrial power first. That gives you the best military,
00:18:57
usually the navy. Then you end up with the reserve currency after a conflict and you kind of rule the free world in
00:19:02
what we've called Pax Americana. Like all great companies, you kind of get lazy through that success and you
00:19:09
end up offshoring all your heavy industrials to the developing country and then when a conflict comes around, you're kind of in real trouble because
00:19:14
you offshored the thing that gave you the power in the first place, which is heavy industry.
00:19:20
The last three times this happened, it was a pretty good trade for the West. It went from the Dutch, the British, the American Empire when we won World War
00:19:26
II. This time around, in this kind of two decade period where we're fighting the AI race, the climate, settling the
00:19:33
stars, it's really the United States versus the CCP.
00:19:38
And bear in mind that we won World War II not because we had a defense industrial base necessarily, but because
00:19:43
we were the industrial powerhouse of the world. And when there was a time of crisis, we had all our commercial
00:19:50
manufacturing companies pivot to defense when we really needed them the most. And you had, you know, watch makers making warship navigation equipment. Ford was
00:19:57
switched from building cars to building bombers. And it was because of this industrial power, you know, our tanks
00:20:03
weren't so great. We just had tons of them. This is how we won. Unfortunately, since the 1970s through
00:20:09
to the 2020s, we've basically hollowed out the middle of America and offshored every bit of manufacturing we possibly
00:20:15
can. This started with Nixitting opening up China, let them into the WTO. They were the world's factory. This is like a
00:20:21
huge strategic mistake and it's completely hollowed out good jobs in America as well as left us in a very
00:20:27
strategically dangerous position in terms of our industrial power. So, while China de-industrialized us,
00:20:34
they industrialized themselves and they treated manufacturing not as economics but a national security priority. And
00:20:39
now we're in this 20-year window where staring down the threat of Taiwan, we're in real trouble. So just how far behind
00:20:46
China are we? Well, in munitions, China has automated factories that can produce
00:20:51
a thousand a year, whereas we run out of missiles in the first seven days of any war game conflict. And then we can't
00:20:58
reproduce that ammunition for like 3 years. In ship building, they're 200
00:21:04
times greater than us. We produced a grand total of five ships last year. Pharmaceuticals are all offshore.
00:21:10
Drones, iPhones, we don't make any of them. And bear in mind in pharmaceuticals the CCP makes all our
00:21:15
antibiotics. This is why industrialization is so important. And more importantly, this gets back to the
00:21:21
AI race for talent is while the US is still the global powerhouse in software and AI talent, we made China into the
00:21:28
global powerhouse for manufacturing talent. And what we realized through building this company is that while US
00:21:34
defense manufacturing, which was all we have left because we offshored everything else, is really important
00:21:40
because we let all those jobs go, the entire base is basically a bunch of patriotic Americans that still know how
00:21:46
to do skilled trades that are in their 60s are retiring faster and faster and faster. So the underpinnings of our entire defense industrial base is this
00:21:53
American talent that knows how to do the job, but the rest of the country forgot how to manufacture.
00:21:59
This is a screenshot of one of China's munitions factories. You can Google this online. And it's a myth that it's just
00:22:04
lowcost labor in China anymore. They are very advanced at production. Whereas in the United States, underpinning all our
00:22:11
defense primes and our industrial base, we basically have skilled Americans that are retiring faster and faster and faster, supporting 100200 billion
00:22:18
industries across all these different ways to bend, cut ship metal that you need to then put it into drones, ships,
00:22:25
satellites, rockets. So while China is racing ahead of us,
00:22:31
we're really falling far behind and we forgot how to manufacture. So what we realized was we have to build full stack
00:22:37
AI powered factories to solve this problem. Secondly, the number one issue is this massive skilled talent shortage.
00:22:43
Remember if you look at ship building or any of these other industries, we are begging for millions and millions of welders or machinists that we you could
00:22:50
give me a billion dollars and we can't hire them in this country anymore because we lost that skill. the production not having inventory is real
00:22:56
deterrence and that you've got to do this by re-industrializing the country to create more jobs not replace them or
00:23:02
automate away and that it's always about national security not economics so we set out to solve this problem by
00:23:07
building automated factories driven by AI in the US 3 years ago when we started this journey
00:23:13
we figured out how are we going to do this well the answer was just start running a factory and build all the AI software at the same time which was a
00:23:19
hilariously painful journey in the early days of the company this is what factory 1 looked like We partnered with some of
00:23:25
America's greatest aerospace companies to really beta test this for a good 18 months. What can we automate? What can't
00:23:30
we? This is one of our first tiny parts that we ship to America's greatest rocket provider. And now we're up to the point
00:23:36
where we're building whole products. We built Opus, which is a full stack
00:23:43
platform for AI autonomy of factories that does a couple of really important things.
00:23:50
In 2024, we launched Factory 2, once out of this beta phase, scaled 10x in a single year with the fastest growing
00:23:56
manufacturer in the country, and now lucky enough to support America's greatest companies, uh, both startups,
00:24:02
defense primes. And this is what the most advanced factory, in our opinion, looks like in
00:24:08
America today. This is our scaled factory 2 in LA. Here you see cutting metal coming from
00:24:14
raw material, shaving this down into micron precision tolerance components that go on rockets,
00:24:20
satellites, jets, and drones. And what you see as you go through this is in legacy industry, in a de-industrialized
00:24:26
nation, you've got really skilled people on every machine. Hadrien's advanced factories look and operate more like a
00:24:32
data center. We're really proud of having pulled this off, but the journey is not over yet because again, this is a whole of nation
00:24:39
100 to200 billion problem. So where do we actually get to and what sort of productivity gains can you get
00:24:45
in AI and manufacturing and are we creating more jobs? So firstly, most factories in the US run at only a 20%
00:24:52
uptime. It's not really that productive. We have a four times jump in manufacturing productivity.
00:24:57
Secondly, and more importantly, we have a 10x jump in workforce productivity. And again, because we have a such a
00:25:04
scarcity of skilled talent in this country, you actually need that AI powered jump to even create the capacity
00:25:09
in this nation to be able to build ships, drones, and rockets. The second important thing is speed to
00:25:15
get people in these jobs. So, if you're an advanced manufacturer, it can take you up to a decade to get really good at
00:25:20
what you do. Whereas at Hri, we've managed to make it so that we can train anyone in 30 days. And most importantly,
00:25:27
100% of our workforce are from non-factory backgrounds. They've never set foot inside a factory before. These are folks straight out of high school.
00:25:34
They're retired from the military. They had a desk job. They were a nurse or bus driver from, you know, 18 up to 40. And
00:25:40
this is the most important thing that people have got to realize about the power of advanced AI in manufacturing is that we need this productivity boost to
00:25:48
just be able to compete with China and catch up on these skill trades that we lost. And this is the most important thing that AI is doing for us is
00:25:54
enabling huge, huge workforce growth. So, where are we at? You know, we've
00:26:00
been on this journey. In 2025, we're going multicategory, multiactory, and I'll show you our new factory that's
00:26:05
launching AI powered in six months in the great state of Arizona, as well as launching a dedicated gigafactory. And
00:26:12
you can think about this as like everyone in defense and aerospace needs a Tesla Model 3 factory.
00:26:18
This is our beautiful uh new facility. It's about four times the size of the one in LA, launching by Christmas. We signed a lease a couple of weeks ago.
00:26:24
It'll be online in 6 months. And the most important thing is we'll be creating 350 plus new AI powered jobs in
00:26:32
scarce talent industries where America just needs this leverage to get ahead.
00:26:37
The other thing if you listen to the secretary of the Navy and re-industrialize what is the number one problem in ship building, submarine base
00:26:44
and munitions, it's actually that there's millions and millions of jobs that we need to fill because we don't have skilled trades anymore. We don't
00:26:50
have the volume of the people. So we need this productivity boost. So in 2026, we're launching advanced factories targeted at America's greatest
00:26:57
production challenges, submarines, ships, and munitions. So by the end of this year, we'll have
00:27:02
three facilities up and running. Our headquarters, Factory 2 and Factory 3 in LA. But as we re-industrialize the
00:27:07
country powered by AI, like where is this really going to get us to? Well, to solve this problem for the country and
00:27:13
fulfill the mission, we need to have factories in every state. And you got to remember that AI in manufacturing is
00:27:21
creating thousands of jobs because we offshored everything. And we need this productivity boost to give our nation the capacity it needs. Reshore all these
00:27:29
jobs, pull them back into the middle of the country, and make sure that we're creating millions and millions of jobs
00:27:34
along the way. Thank you for having me. It was a pleasure to be here.
00:27:42
Chris, I think we wanted to kick this off. We have a couple minutes to just um cover what you've introduced which is I
00:27:48
think like a really important opportunity. China has roughly 3 million factories. US has 250,000. The
00:27:54
assumption is they've got cheap labor. Looks like they've got automation. Things are very different on the ground than what folks read about as we try and
00:28:02
compete. What industries are going to be first from a manufacturing perspective that we can actually compete
00:28:07
successfully and do we need trade tariffs in order to succeed on the competitive landscape? So I think
00:28:14
there's two really important points. One is there's industries that we have to reshore. Specifically in defense, we
00:28:19
have to produce submarines and ships and munitions. Uh we have to produce things like rare earth magnets and drones. We
00:28:25
just have to do it. Um the tariffs really help. And this trade policy is really important because you got to
00:28:31
understand that yes, China is more competitive than us, but the CCP also nationally subsidizes the cost of
00:28:37
energy, the cost of raw material. And because we've kind of degraded this capacity like not having nuclear in the
00:28:42
US like we can't compete on those raw inputs. So it'll start with our most critical industries first. But I think
00:28:48
as AI goes through manufacturing you'll create millions of jobs and that will allow us to reshore more commercial volume not just in defense. And I think
00:28:54
that's the most important thing. And you've talked about this like degrading you know sort of infrastructure and what that means in terms of workforce but then how
00:28:59
reshoring also requires this like upskilling. I know you guys have this like associate named Owen that you guys took I think like literally straight out
00:29:05
of Home Depot. Can you like give us a little bit of his story and just like what that represents in terms of you guys upskilling labor?
00:29:10
It's it's really incredible. So, as I said in the presentation, 100% of our people never set foot inside a factory
00:29:15
before. And I think we really didn't do a great job as a nation by convincing everyone they needed a four-year college
00:29:20
degree to have a really good job. And we've hired people that, you know, packing shelves at Home Depot. Now they're running 10 machines at once. And
00:29:27
actually, what we are seeing is that most of those people when they're exposed to software or AI, they're very
00:29:32
smart. And we've promoted a lot of those people into leadership management or software engineering roles. And I think
00:29:38
re-industrialization with AI is about creating new jobs, but also reattaching people to the Silicon Valley economy and
00:29:45
not just having it on the coast and the cities. H how um are you going to compete with
00:29:51
people having gig labor and making 30 40 bucks an hour being a door dasher? Um
00:29:57
and we have the lowest unemployment in our lifetimes, 4% right now. Is it
00:30:02
realistic to find all this labor out there or do we have to have some people
00:30:08
immigrate to this country in order to fill those jobs? For for us specifically in defense, um
00:30:14
we we can't we have no choice in immigration because it's a regulated environment. So we have to upskill Americans.
00:30:20
Secondly, what we see maybe not in LA or the coastal cities, but across the country, there's lots of underemployment. Some of our favorite
00:30:26
people have desk jobs where they're a parallegal and they were filling out forms and they hate it and they want to come in factories and work on the
00:30:32
national mission. And I think for us it's a lot about getting people inspired. And then secondly, with this level of productivity jump, we can
00:30:38
actually afford to give people incredibly good healthcare and incredibly good pay. And I think a lot of Americans want to go back to work in
00:30:44
a real environment that's for the national mission. You uh you showed some incredible images
00:30:50
and video of these very intricate machines. are do you make the machines that then make all the machines or is
00:30:56
there a supply chain risk as there is a huge supply chain risk so we
00:31:01
actually invented via the air force a lot of these advanced machines um and we
00:31:07
forgot how to make them so the main sources of supply are actually our allies in you know China is number one we don't buy for them because they've
00:31:13
got cyber security holes all over the place Germany South Korea Japan the insight that we had was they're actually
00:31:19
just really dumb computers and software and AI can actually upscale and overpower them um and really have a
00:31:25
lead. But it is a huge supply chain risk of not building the machines that build the machines in the country anymore. Right. To economically compete though,
00:31:32
do you I I was trying to parse if you were asking for the government to give you support since the Chinese government
00:31:38
is underwriting their companies with free energy. Are you explicitly asking the government to help with say paying
00:31:44
for reskills training or um maybe in some way deferring your energy cost or
00:31:50
can you make this economically work? We make it economically work because in the US there are really two markets. There's
00:31:56
the stuff that has to be onshore for defense and aerospace and then there's this offshore market that's 10 times
00:32:02
larger. Um you know commercial aircraft, a lot of that is in China. For us we can compete in the US because we've got to
00:32:08
create all these new advanced jobs because we just don't have the skills anymore. If we want to reshore the commercial volume that is not regulated
00:32:14
to be onshore, we have to do tariffs and economic policy because it's not an even playing field. It is right now companies
00:32:21
versus the CCP. What would that look like in in terms of execution? You would want uh them to
00:32:27
pick up the retraining, the energy cost, part of their salaries. It's it's really three things. It's the
00:32:33
cost of energy. It's the cost of raw materials, aluminum, steel. 90% of the cost of that is actually energy. And if
00:32:39
we level that playing field, then we can go compete in what we're great at, which is the, you know, American software and
00:32:44
the American spirit and AI powered workforce. So the silver bullet is energy. Yeah. And then tell us about the actual
00:32:50
software. Do you have a team that's writing a lot of control systems and or AI models themselves or you're taking
00:32:55
things that are off the shelf and you're fine-tuning them? How are you how are you doing it? Unfortunately, because American manufacturing software is 30 years
00:33:01
behind Silicon Valley, we had to build everything ourselves from scheduling systems to the deep tech. And the key
00:33:08
insight that we had is the faster we grow, the more data we are labeling, right? So we always do things 80%
00:33:14
automated with a human in the loop. And as we label this complex manufacturing data, you know, this is where our AI
00:33:20
models actually kick in because manufacturing has been offline for 30 years. So there is no stack overflow.
00:33:25
There's no GitHub code base to train a model on. We had to train it ourselves off our own label data as our experts
00:33:30
were ticking and tying the automation. I mean histo like traditional automation is purpose-built, does one thing. A lot
00:33:36
of engineering goes into making it do that one thing really well. Are you leveraging things like to Chimat's
00:33:42
question vision action models that allow you more extensibility with one particular piece of machinery and like
00:33:47
or when does that start to happen from a tech perspective in your view? Right. Right from the start. So the way
00:33:54
oddly that customers translate data to their supply chain is by giving them 20page PDFs full of hieroglyphics. So we
00:34:01
actually have to train huge vision models on interpreting that. What does that mean? It's it's very complicated and it usually takes an expert 50 hours
00:34:07
to pour over that. So, it's vision models, it's training engines on the data, it's also training engines on
00:34:13
reinforcement learning of, hey, we we made a part with automation. Was it high quality or not? Did it actually work?
00:34:18
And embedding all of these in the workflow real time is is the magic trick here with AI. And you're not doing this stuff just
00:34:24
like you know on the coast, right? Your next factory is sort of more middle America. Like how do you end up choosing where to put that? The most important reason why we
00:34:30
selected Arizona was because of permitting, energy, and regulations. You know, we got to go fast, right? We got
00:34:35
to build this in 6 months and then we will expand into the middle of the country kind of left to right on the map
00:34:40
and I think that's the most important thing is we're going to be able to expand into all these cities and states where the manufacturing jobs were
00:34:46
destroyed and we're going to bring him back. Are you guys investors? Oh yeah, just led the uh series C which we just announced last week and uh join
00:34:52
the board much to uh Chrisin. You're on his board? I'm on his board. Yeah. Is that terrifying? It is very terrifying.
00:34:58
How long have you guys known each other? Uh too long. Yeah, too long. Yeah, I was board observer for a while and uh you know tried to avoid getting the official
00:35:05
seat. Uh you got a date for a while and then now we got married. Well, Chris, thanks for being here today. Thanks for hosting me, guys. Pleasure.
00:35:10
Yeah, thanks for the education. Thanks, man.
00:35:16
There is a huge fire going on right now at Philadelphia Energy Solutions. Oh my gosh. Again, look at this guys. Look at
00:35:23
this video right now. Today, the Navy remains a formidable fighting force, but
00:35:29
even officers within the service have questioned its readiness. Developing right now, gushing for hours
00:35:34
with no end in sight. Thousands of barrels of crude oil spilling from a tank. Uh the report does an estimate of
00:35:41
what the need is to bring the overall grade up to a B, which is what the society sort of determines to be
00:35:47
adequate, and it's like $4.59 trillion.
00:35:53
Heat
00:36:02
[Applause]
00:36:10
up here.
00:36:18
[Music]
00:36:30
A company that started in my college dorm is now a company that manages over 500,000 of the world's most critical
00:36:36
pieces of infrastructure. Now, at Gecko, we build robots and AI
00:36:42
models to help unlock physical the physical world. Now you see we build when we rebuild
00:36:48
robots we wanted to build them that could fly, swim, crawl and walk on any surface to gather the most amazing
00:36:55
information and data sets that have been forgotten about the physical data layers. Now all those data layers are
00:37:00
incredibly valuable when you're able to unlock and use AI models to drive
00:37:05
incredible and important outcomes. Now, I started the company in the energy
00:37:11
sector, deploying the technology to help prevent catastrophic failures in downtown power plants. Now, we've to
00:37:18
expand into mining, metals, and manufacturing as well as for the defense. And so, we're helping to deter
00:37:24
conflict by getting ships out of dry dock on time and not uh and patrolling this the the borders. Also, we are
00:37:32
helping the air force ensure that planes are in the air and not in hangers. And then just last week when the president
00:37:38
was in Pittsburgh, my hometown, we just signed an amazing deal that ensures that we can help revitalize manufacturing in
00:37:45
the United States again by helping to build ships and subs. Now, the energy sector has been
00:37:51
incredible and we in a lot of other sectors as well. But what I begun to realize is that the most impactful thing
00:37:57
that Gecko Robotics can do to help ensure that we uh deter conflict and our
00:38:03
most impactful for national security is actually in the energy sector. You see, President Trump is absolutely right and
00:38:10
his executive order today calls out an extremely important um extremely important reality that the companies
00:38:16
that can unlock energy are going to be the ones that can dominate in the AI race.
00:38:23
However, as you can see from the graph here, China is on pace by 2030 to 3x the
00:38:28
amount of generation by against the US. But this isn't the whole story. You see,
00:38:34
we constantly think about AI as an energy consumer. However, I'm here to tell you that artificial intelligence
00:38:41
can actually be used and unlocking energy production in ways that you never seen before.
00:38:48
Now, inputs really matter to being able to unlock this potential. And CEO after CEO that I talked to in
00:38:55
the energy, mining, manufacturing, and defense sectors will tell you that we're trying to figure out how to unlock
00:39:00
artificial intelligence to supercharge everything. However, the value is just not there. And it's no wonder the
00:39:07
consistent common factor between each one of these sectors is Joe. Joe is out
00:39:12
there gathering information by hand, trying to diagnose and get physical data
00:39:18
to drive really impactful decisions. But it's important to understand that Silicon Valley uh Silicon Valley
00:39:25
artificial intelligence researchers and software engineers, they can't do much with data sets coming off of the backs
00:39:30
of Joe and Joe's been armed with the same technology for the past century. So, it's no wonder that impact isn't
00:39:37
being unlocked in these sectors. And unfortunately for Joe, it's a very dangerous job as well. And someone dying
00:39:44
doing this job was actually one of the things that inspired me to build Gecko. We have to give Joe better tools in the
00:39:50
new century. So, what I'm going to walk you through right now is an example of exactly how we do that for the power sector. We send
00:39:56
in robots. Robots that are gathering information and data sets about the physical environment. In this case, a
00:40:02
natural gas power plant. We're understanding what the physical environment looks like. And then we send
00:40:08
in other robots like this dog over here. Now the robot dog is gathering operational data sets to help
00:40:14
supercharge cantaliever our AI powered platform where all the data sets are coming into. You see we sell an
00:40:21
operations platform and data sets gathered in the physical world is what's enabling that.
00:40:27
We also send in wall climbing robots and you can see the wall climbing robots to your left and to your right. Now these
00:40:33
robots are going into the physical environments and gathering health data all while the app while the uh power
00:40:39
plant is actually online. Now the health data is really important because we have
00:40:44
to understand process health data to be able to optimize and feed into AI models. But again this data set just
00:40:50
never existed before. So we had to go out and actually get it physically in the real world.
00:40:56
So robots like this supercharge our ability to be able to drive models to
00:41:02
create largest amount of efficiency gains. So this power plant for example is supposed to be operating at 620
00:41:08
megawatts but it's not reaching its capacity. It's only operating at 580. So how do you unlock that? Well, when you
00:41:13
have all this information and data sets that we've captured with robots, plus all the data sets that customers have,
00:41:19
you're actually able to drive optimization to see how to impact efficiency and production. And so what
00:41:26
this AI model is doing is looking at the data sets from the robot dogs as well as data sets from the health the health
00:41:32
data from the robots to pinpoint that there's actually a steam issue going into the turbine. Now, an ability to fix
00:41:39
these things has actually been able to unlock for this site and for many others that we work on a 1% improvement to
00:41:46
efficiency. And this is just the first place that we looked. Now, it's also important to understand
00:41:51
that the assets that power the grid are failing at a really fast rate. Now, this power plant had assets like this tank
00:41:58
that was decaying at an at incredible rates. It was supposed to be reaching retirement pretty soon. But we were able to determine predictively how to extend
00:42:05
the useful life of this asset by 10, 20, and 30 years from all this data set.
00:42:11
It's really important to understand when you culminate all the kinds of impacts that you can have from this this kind of
00:42:17
technology, you get things like this efficiency gains on the dozens of power
00:42:22
plants that we've been able to work at. If you extrapolate that across the thermal fleet in the US, that'll give
00:42:28
you 11.9 gawatt of new power without putting a shovel in the ground. The
00:42:33
energy is able to be unlocked using artificial intelligence. It's really important to understand the statement.
00:42:38
AI shouldn't just consume, it should create energy. And that's what we're showing here. And not to freak anybody
00:42:45
out, but the DOE just came out with a study that showed we have about four years left of useful life on the assets
00:42:52
that power our grid. Now in this trend it means that a 100 there's going to be
00:42:57
a 100 times the amount of blackouts by 2030 if we don't reverse this trend. But what we're able to show not just with
00:43:03
power plants but mining metal manufacturing as well as defense assets you can actually extend the useful life
00:43:09
of infrastructure in some cases by 30 years and on average it's been about 35.
00:43:15
This is extremely important in ensuring that we're able to reverse that trend
00:43:20
and ensure that America is well positioned to ensure that we lead on the
00:43:25
energy and the in the energy race to enable and unlock artificial intelligence. Now, let me summarize
00:43:31
this. We spent so much time and I think JD Vance has done a great job at at highlighting how much effort and how
00:43:38
much data set have been gathered to power AI models in the digital world. And it's what makes chat GPC so
00:43:44
addictive. But remember, the physical world has been forgotten about. And our
00:43:49
robots are going in to the fog of war to try and decipher and unlock massive amounts of information and data sets
00:43:56
that gives America and our allies unfair advantages. Unfair advantages to unlock
00:44:02
things that we didn't even realize were there. And if you build software with an
00:44:07
ontology based on first principles, gathering the data and building software up from there, you're actually able to
00:44:13
deliver impactful things for Joe, turning Joe into a PhD scientist or
00:44:19
engineer instead of forgetting about him like a lot of Silicon Valley companies have in the past. Unlocking potential
00:44:27
physical intelligence data drives um it drives artificial intelligence and
00:44:32
that's how you're going to win the AI race. Thank you.
00:44:38
[Music]
00:44:44
My name is Laura Deerdinas and I'm a registered nurse here at Tampa General Hospital. The last 17 years I've been
00:44:51
able to serve the neurointensive care unit where we care for the most vulnerable and critical care patients.
00:44:58
So before utilizing AI, it would take
00:45:03
hours to gather information, looking in chart reviews, uh talking to nurses,
00:45:09
talking to physicians. We relied on paper, pencil, um a lot of paper stapled
00:45:15
together with sometimes outdated data by the time I was done going through uh 32
00:45:22
patients. and this is how we would try and give reports. Bringing in AI, it has
00:45:30
significantly changed the culture on the unit. I had a
00:45:36
charge nurse who never gave um a multid-disciplinary round or report out.
00:45:43
She came on board. She said, "This is an amazing tool. Look at this. It has all my information already gathered and
00:45:50
collected." and she was able to report out on the patients. It was completely
00:45:55
user friendly. She's like, "Laura, what is this?" It is creating excitement
00:46:01
throughout the nursing community. Using AI has provided more time to be with you
00:46:07
or your loved one at the bedside where nurses should be. We are the heart of healthcare.
00:46:13
Matt Troutman. I'm the vice president, general manager for PRL industries, supplier of components for nuclear
00:46:18
submarines at our service men and women lives depend on. We are a fully integrated foundry pouring metal all the
00:46:25
way through finished machine components. 2 months ago, we weren't getting after any of the problems on the shop floor.
00:46:31
engineering director told me all his team was doing was quoting a 3-day
00:46:36
process to quote apart paper files, old archives, data tables, emails, side
00:46:41
communications, all this which ends up getting lost in the fray. Now, using an AI tool, they are getting halfway
00:46:48
through that process in minutes frees them up to get back out in the floor and do what an engineer does best, which is
00:46:53
solve problems to provide the Navy with the best quality products in the
00:46:59
shortest amount of time. And this is what AI is going to help us do. Understanding part location and status
00:47:06
that is a gamecher. We can now talk very clearly with the customer if that part is now to become the primary focus of
00:47:13
the business cuz it's a critically needed part for a ship construction. You get notified. It's an automatic notification. We can see the exact
00:47:20
status. Here is the impact. How can we be better? How can we do more? And this is how we're answering that call. With
00:47:27
AI, we match the speed of the quality management process to the speed of the
00:47:32
workforce and the machine capabilities and we will truly see a multi-step change in the amount of product that can
00:47:39
come out of any company in this supply chain. More jobs for American workers here at
00:47:45
Pro. My name is Julie Nordberg. I'm a registered nurse leader here at UP
00:47:51
Health System Marquette in the heart of Michigan's upper peninsula and we are really the only game in town as we like
00:47:57
to say it. The next closest hospital to us that could service us is downstate which is about a 4-hour drive. Prior to
00:48:03
using AI, it took a lot of time to go through the patients charts to see where they need to be. It took a lot of time
00:48:09
just to try to communicate with people. I think that's a fear that everybody has is that AI is going to replace people.
00:48:15
But AI in the way it's being used here could never replace our frontline staff.
00:48:21
You know, that the vibe is I think it's just one of excitement that everybody's just proud to be part of this and to say
00:48:27
that we're doing it here and we're honing it in and tweaking it and and using it to enhance our care and using
00:48:33
it to help our staff. Having this kind of communication hub and facility snapshot has helped has helped
00:48:40
everybody. For the nursing staff, I think being able to see everything in one spot has just revolutionized kind of
00:48:47
how they are able to provide care. I don't think anybody is sad to get rid of a meeting.
00:48:54
The impact on patients is earlier detection, which means earlier treatment, um, which is a better outcome, life-saving for some of them.
00:49:02
That's where I think this is going to help us a lot is because we we don't have as much manpower as those big academic centers. So having the AI in
00:49:08
the background doing some of that leg work for us is is huge. I joined PCNA in 2018. We have built
00:49:16
over 11 billion batteries in the last 8 years. I walked out onto their massive production floor for the first time. I
00:49:23
knew right then and there I wanted to make this technology accessible for anyone who wanted to learn it.
00:49:30
people coming from the tourism industry and the hospitality industry. Quite a few technicians that have fixed slot
00:49:37
machines in a past life. People from automotive companies, people who are used to repairing cars, however, have
00:49:42
never seen equipment, you know, at this scale and with this complexity. You know, we don't really um have to pick
00:49:48
and choose what people's backgrounds are because we do have this very powerful learning tool that makes it easy for
00:49:55
anyone to be able to enter this industry. It is taking our historical maintenance
00:50:01
records, pairing it with our machine data, and is now starting to understand
00:50:06
early warning signs of a breakdown, and deploy our technicians to equipment before it ever actually breaks. This
00:50:13
helps minimize our production losses, keep our technicians safer. We're taking reactive events, turning them into
00:50:20
predictive events. We used to honestly lose a lot of technicians because they would lose
00:50:26
their confidence, think, "Hey, maybe this isn't for me." I pulled the supervisor off the floor and said, "Hey,
00:50:31
you got to come listen to this idea and you have to help us make it better because you're the one who lives it every day." And they immediately started
00:50:39
suggesting new features. They were telling us what was wrong with the old systems and we were coming up with
00:50:44
solutions on the spot. So, this is really helping people feel like they belong here. We don't believe AI should
00:50:50
replace human talent. We believe it should elevate it. Our workers are very excited. They have a tool that they can
00:50:56
turn to to help them learn at their own pace. It really puts the power back into their hands.
00:51:05
[Music]
00:51:16
All right, Christian's joined us from Hillen
00:51:21
Valley and 137. And Sean, welcome.
00:51:27
Thank you. Great to be here. Christian, you want to kick us off? Yeah. Uh, thanks for having me. It's It's nice to be here. I'm definitely uh
00:51:33
feel for the first time like a guesty of the besties. Um, don't [ __ ] it up. And, uh, yeah. Uh,
00:51:40
this is great. So, Sean, thanks for coming. We were talking a little bit earlier. maybe this is a great place to start. Um, obviously we have the good
00:51:46
fortune to be investors in Paler for 15 years. We've seen the growth of the company, but particularly lately you've
00:51:52
been pushing this messaging. I think it's been incredibly exciting of how AI is not a force for job destruction. It's
00:51:57
a force for job creation. It's also a way that you can give superpowers to the average American worker. And obviously, we've seen a little bit of content here
00:52:04
and how it's already doing that today. I want to start by saying many of the workers in the video are actually here today joining. Uh Laura, the nurse from
00:52:12
Tampa General, actually brought her 12-year-old daughter. So, I think the ultimate litmus test is not just how excited are the American worker to
00:52:18
leverage AI, but how excited are they for their children to exist in an America that's really embraced AI. And
00:52:23
Julie has four kids, and she would tell you how how much this has not only transformed her view of her job, but the
00:52:29
view of her children's future. Uh I I think the the right frame here really is how do we give the American worker
00:52:35
superpowers? You know, we should not be aspiring to build things that make them 50% more efficient. They're really 50
00:52:41
times more productive and to use that as our our asymmetry in the competition
00:52:46
here. You know, our strengths are not only AI, which is clearly an American phenomenon, but also the ingenuity of
00:52:52
the American worker. And if you spend time on the factory floor on the front line, you see a very different narrative
00:52:58
emerging where you see people are actually excited about these tools. Every single one of those workers to a tea said AI is giving them more time to
00:53:04
do what they do best to spend time with the patient delivering care to actually build the parts as an engineer to solve
00:53:10
the problems not to be caught up in all the coordination and the paperwork that's around these things. Uh that's
00:53:16
the future we should be unleashing. Can you generalize the adoption curve? What is it about a particular industry
00:53:23
or use case that makes it an early adopter versus mid versus late that you're seeing because now that you're
00:53:29
touching all these different industries. Uh you probably have a good point of view on this. Yeah, my my take is actually a different
00:53:35
dimension of slicing that which is where do the where does the institution liberate their worker to drive the
00:53:40
adoption versus where are they trying to forcefitit some sort of solution top down. You know, AI is a method of
00:53:46
unleashing the agency of the worker, the creativity of the individual, and they're the ones coming up with these
00:53:52
use cases. I mean, Chris was talking about it from Hrien where you'd be surprised at how people with deep
00:53:58
mechanical intuition, traditionally considered bluecollar workers, are the ones who are able to pick up the skills, build the applications, innovate on
00:54:05
their own processes, and have that spread through the organization. Mhm. And are you seeing that you have to
00:54:10
build vertical tools or generalized tools for some horizontal kind of set of
00:54:16
users somewhere in the organization? Well, I think that the opportunity with AI is really that you can unleash what's
00:54:22
different about your business than all the others. So there there's a degree to which you can have uh generalized solutions, but there's a lot of alpha to
00:54:28
be captured by understanding what's unique about how we do things. How do we lever up human taste? Everyone is afraid
00:54:35
of AI replacing the human. That's not what I'm seeing. I'm seeing it it make the the most the the person with the
00:54:41
greatest taste more valuable and an ability to spread that to the breath of the organization. Let's talk about um something beyond
00:54:48
taste which is also like knowledge and skill and tell us about AI inside of
00:54:53
healthcare. Um I think that a lot of people probably think that we have an
00:55:00
incredibly cutting edge system of tools and software that helps doctors and nurses actually provision great care.
00:55:06
What's the actual reality that you guys are seeing? Well, sadly, I think with the forced adoption of EHRs, what we saw is roughly
00:55:14
a havinging in the productivity of how many patients you can see per hour. A having a having. Yes. So, we became half as
00:55:20
productive. And we really need to, you know, the opportunity is to work backwards from what is the care that
00:55:25
needs to be delivered? How do we build the tools around that? How do we help the the the nurses, the care staff spend
00:55:31
more time with the patients and less time with the computer? And do you guys see a world where in order to facilitate
00:55:38
that end market versus a different end market you have a ensemble of many many
00:55:43
many different techniques and approaches in AI or do you think it all sort of gets form fit into this one trillion
00:55:50
parameter huge ginormous thing that kind of tries to do everything? I think the cardality of agents and
00:55:55
models is very high. I think there will always be alpha to be achieved you know improved differentiation improved
00:56:00
outcomes by specializing to the use case. Now, it's great to start with the general models, but you will specialize
00:56:07
over time. And do you feel pressure to do that now or do you think that'll just be a natural evolution over time as
00:56:12
Yeah, I think it's a journey that people kind of get on like you realize like, wow, look how much better things have gotten with this. Now, how do I go get
00:56:18
the next incremental piece of performance out of it? You know, I'm just having this thought as we sit here and discuss this. If you
00:56:24
think about any experience we have in service that has a long wait time where
00:56:30
we feel like we got more time with the practitioner, it's the perfect place for
00:56:35
AI to create more abundance and healthcare and education are the two that that come to mind uh where people
00:56:42
could just offload their chores and the people who are getting the service can use AI to maybe start the conversation
00:56:48
on second base or third base. What other industries are you seeing after healthcare education where AI can have
00:56:55
that dramatic of an effect where the six week wait time to see a doctor the three
00:57:00
or four other students who are getting tutored are ahead of you and maybe you don't need as much help so you don't you
00:57:06
never get the tutoring. The the place I'm most excited about it is really in reindustrialization. So it
00:57:12
because there's so much dwell time in the value chains around what does that mean dwell time in
00:57:17
industrial where you're just waiting for someone else to figure out how to approve something or the coordination costs mean
00:57:23
that it's essentially dead weight loss. Give an example there. Yeah. Uh you saw it with the the submarine uh
00:57:28
industrialbased partners there where they're working on quoting a part to the Navy. That means you have to go gather
00:57:34
all of this data. You have to look at historical archives. All of that is time. You're not making a part or
00:57:39
solving problems. that's just it's just sitting there. The factory floor is idle, right? So, how do we get rid of that
00:57:45
dwell time so that you can be utilizing the the capex that you actually have to the maximum extent possible and then if
00:57:51
you start if you zoom out that's like one part manufacturer. You're you exist in a massively complicated supply chain
00:57:57
and you just end up with all these busy weights along the way here. Yeah. That's so profound. A friend of mine said who's in that industry, you're
00:58:04
only as efficient as your worst supplier. Exactly. Yeah. And a second part of that which uh the Panasonic energy example
00:58:10
really touched on is how do we train our workers? You know, so here you have exquisite Japanese technology. It used
00:58:16
to take three years to train a worker on it. Now with an AI assistant, the workers who are prior casino workers,
00:58:22
you know, they're not not from this industry are able to get up the curve in three months. So you think about how we
00:58:27
can use that to more quickly absorb the slack that's happening as we as we adopt AI and and democratize opportunities. So
00:58:33
much so I have so much conviction as we've launched the American Tech Fellows program at Palunteer to find blueco
00:58:38
collar workers at our customers in in the heartland overlooked folks who have a natural proclivity.
00:58:44
How do you find them? How do you find them? Well, some of them beyond just saying apply like yeah some of them are at our current
00:58:49
customers. The idea really came from us organically where it's like wow who is building the most compelling applications? It's the guy on the
00:58:55
factory floor, not a formally credentialed computer scientist, mostly an autodidact, but there's immense not
00:59:01
only grit, but ambition. They have the drive to reshape their own organization,
00:59:06
to reshape the processes. Let's bet on that person. going earlier, does that mean, and I'll ask the same question
00:59:12
many times today, that college education, the traditional four-year liberal arts degree, doesn't matter as
00:59:18
much, that kids can go from high school or earlier in their careers into a new
00:59:24
workforce and get well trained and well suited to to make money and and succeed in life?
00:59:29
Yes. I I think I think the traditional college degree is is dead and uh we should be betting on the American
00:59:34
worker. Well, on that point, can you talk about the tech fellowship? uh I got to recently see a bunch of demos from the first cohort with you and is really
00:59:40
incredible what you guys are doing there. Maybe give a little bit there and then maybe also talk about the opportunity for other companies to follow this trade school framework as we
00:59:47
uh end here. Yeah, I mean it's really kind of an elite trade school. So like finding people with me mechanical intuition who
00:59:53
have done things, some of them are right out of college, some of them are 20 years of experience, but they're really your first trade school that you guys
00:59:58
have done, right? Yeah, that's right. Uh and we have just enormous demand from our customers.
01:00:04
We're like, who are people who have these skills, you know, and it's it's not classically trained, college educated people. They don't have these
01:00:10
skills actually. So, the market's not meeting and they don't know how to source these folks. So, I can credential
01:00:15
them. I can put them through the boot camp in four weeks and place them with my customers to go unleash AI within
01:00:21
their organizations. It's incredible. Sean, thank you, Sean. Thank you. Thank you so much.
01:00:26
Well done. Thanks, man. Thank you. That's great. Thank you so much. Cheers. All right.
01:00:34
Next up, uh, we have Paul from Y Combinator. Please welcome
01:00:40
Oh, Paul Buhight. Blue Height. Paul Buhight. Paul Blue.
01:00:46
Paul, you created Gmail talking about efficiency and making the world more efficient.
01:00:51
And also, I believe we work together. You came up with the slogan. We worked together, too.
01:00:57
The slogan don't be evil. Yes. Yeah. Yeah. How'd that turn out? I I don't know what
01:01:03
she It's an attempt at alignment, right? Like we we worry about AI alignment. What do you What do you tell the super
01:01:08
AI once you've built it? Yeah. Um you're at Y Combinator now.
01:01:14
Although you recently said you're stepping down, right? Or you're uh partner ameritus. We're starting a
01:01:19
new firm uh Standard Capital. So Got it. Oh, that's exciting. Yeah. Wow.
01:01:24
Let's talk about the game on the field with startups. You get to see startups in year zero and year one. Um, and one
01:01:32
of the primary thesises I think we all have is vibe coding and making coding not a
01:01:39
roadblock. I think Paul Graham's great innovation at Y Combinator was saying I'm just going to accept two or three
01:01:44
people who actually build the product. In fact, in the YC application, it says who wrote the code for this? Who's
01:01:50
writing the code? Just so you can make sure that you're actually hiring coders. What are you seeing on the field in terms of vibe coding? Because people are
01:01:57
now great question. the you know English is the new programming language. It's only
01:02:03
two or three% of the country knows how to code probably half that code well enough to do a startup. So here we are
01:02:10
um could we be on the precipice of 10 times as many startups 100 times as many
01:02:16
startups? Absolutely. I mean that's that's the dream. Um that was actually you know YC was started 20 years ago uh based on
01:02:23
PG's insight that actually it's getting easier to start a startup right. It used to be you had to have a big mountain of
01:02:29
money, you you know hire a big team, etc. And his realization was you can
01:02:35
start a startup with just a couple of people and um you know basically ramen
01:02:40
few few kids live living off of of ramen. Um and that's proven to be true. Uh and our belief is with AI that
01:02:47
actually just goes that much further, right? Because the universe of people who are able to create uh apps using
01:02:53
something like Replet uh is enormous. And so, uh, my, I think maybe most
01:02:59
optimistic vision of what we're doing with all the AI is essentially putting all of these tools of wealth creation in
01:03:05
as many hands as possible. Do you think that English is I think it's Andre Carpathies that said this,
01:03:10
right? Like, do you think English is the ultimate destination language that everybody will use to code or do you
01:03:16
think it gets abstracted even further beyond that where you sort of think things and they just kind of appear? Uh
01:03:22
I think it might be a little while until we can just think them. But clearly uh that's the direction right is that you
01:03:28
have a a dialogue with the AI and so you describe okay not quite like that more
01:03:33
like this. Um and the direction is essentially just that it becomes easier and easier for us to realize our visions
01:03:40
and for everyone to realize our visions not just people who are well let me ask you this question. You I mean that clearly grows the funnel
01:03:47
right? So now we have 100 million, 500 million, a billion people, two billion people, whoever can speak English can
01:03:53
now code. How does it, how do you think about that as one of the best computer
01:03:59
scientists that America's ever created? Mhm. How do I think about all those people having the ability?
01:04:04
Yeah. I I mean, I think it's great, right? Anything um you know, our philosophy is that I don't want to see
01:04:11
all of the power concentrated in a small number of large organizations. I think that's bad for everyone. It's bad for
01:04:19
freedom. Um, and so what we want is to give that power to as many people as possible so that everyone can create um,
01:04:27
you know, apps and it might just be something for their own local community. It it's not not every one of those apps
01:04:32
is going to be the next Google obviously, but the more people can create wealth in their own community and
01:04:38
in their own lives, we spread the prosperity everywhere. Are you seeing in the applications you get to YC or that
01:04:45
you've heard of more physical AI, robotics, automation, those sorts of
01:04:51
tooling? Because as this becomes easier, it actually leads to the leap. Hey,
01:04:56
maybe I could do this as a robot and I get a robot to do a particular thing and that creates an opportunity for new
01:05:01
business. Has that become a big kind of growth curve right now is physical AI and absolutely that number of robot arms at
01:05:08
the most recent demo day was was striking. I think everyone is starting to work on that and again as as the
01:05:13
things that used to be difficult get easier um we just start doing more difficult things but absolutely I think
01:05:20
and I think robotics all technology curves yeah exactly and and I think that's going to open up you know whole new
01:05:25
realms that were previously impossible or impractical so so does that create new industries is
01:05:30
like I think a key absolutely point which is like what I think is most under misunderstood about AI is it's not
01:05:36
about the displacement of doing old things but it's about activating new things that are complex and historically
01:05:42
not tractable, but now they're tractable. Right. Exactly. So, I mean, if you think about just the fundamentals of wealth
01:05:47
creation, the inputs are essentially energy and intelligence, and we're about to unleash essentially an abundance of
01:05:55
intelligence where like the total global intelligence is going to 10x, right? And so, that will enable us to 10x our total
01:06:02
wealth. And that's going to come in a lot of different forms like you know as we start to have AI science labs for
01:06:08
example where the AI can actually start running its own experiments producing its own data um I think our
01:06:15
understanding of biology is going to be incredible. You know in 20 years we'll be able to know how a drug affects the
01:06:23
body without ever actually testing it. And my prediction is actually our our AI
01:06:28
models will be more predictive than today's clinical trials. You know what's interesting hearing you
01:06:34
talk about this Paul is and it's really the power of great conversations. There was a troll over the last couple of
01:06:40
years when somebody lost their job in journalism like learn to code learn to code and now you think about it there's
01:06:46
multiple types of intelligence. startups were limited or you know gatekept in
01:06:51
some ways by mathematical intelligent intelligence the ability to to write code. Opening up that to people who are
01:06:59
high intelligence or high design, high emotional intelligence could lead to many more beautiful interesting products
01:07:06
that maybe people who are math intelligence uh you know uh focused just would never get to.
01:07:12
Absolutely. Uh and this is an abundance that I think people are maybe not even realizing yet is that a whole group of
01:07:20
journalists, writers who are being displaced or you know Uber drivers or people working in factories, well if
01:07:26
they can embrace this technology and we saw it with um no code. Remember the no
01:07:31
code kind of ghetto that was you know emerged for a couple years. Oh startups are going to be no code. It was kind of
01:07:37
like the false start, but you did see a bunch of new entrance applying for Y Combinator or other things. Uh this this
01:07:44
could really be um accreative to humanity. Yeah, absolutely. And it reaches people
01:07:50
who are perhaps otherwise left behind, right? Like it it shouldn't be just people in Silicon Valley who can create
01:07:56
apps. Like there's a whole country full of people who have ideas. Um and the same thing goes, you know, not just for
01:08:02
apps but for media. Like um I think a lot about uh you know again when we look
01:08:07
at where the generative video models are going it's pretty amazing right in a couple of years that means a kid
01:08:14
in wherever middle America over country who has like a vision for their own Disney movie can actually just create
01:08:21
the Disney movie. You don't need the $100 million budget. And so that's going to give, you know, a lot of voices that
01:08:28
are currently not represented in media because they don't have access to the capital or or Hollywood.
01:08:34
The elite version of this would be, oh my god, we're losing this job creating at Netflix, but you're creating a
01:08:41
million other jobs for people to create their own superhero that represents them, that represents their country,
01:08:47
represents their sensibility. Exactly. Let me ask you a question as a as a technologist for a second. When you see
01:08:53
the landscape of these foundational models and how good they're getting, is your belief that the number of those
01:08:59
will grow or do you think that they'll consolidate and they'll just be fewer but better? How do you see all of this
01:09:06
investment that's happening now? Yeah. Play out and feel free to name companies while you're doing your analys which ones will
01:09:12
go away. Uh yeah. No, I I mean I expect that it'll probably stay relatively
01:09:17
stable honestly because I the cost of building these foundation models is astronomical, right? We just saw XAI is
01:09:25
raising another 20 million, something like that. Um and and so just the
01:09:30
capital requirements are going to limit how many there are. Um but I certainly hope that it doesn't consolidate down to
01:09:36
just like one or two because again I think part of what's important for preserving freedom is just that we have
01:09:43
many options. And so actually a lot of people don't know we started OpenAI at Y Combinator uh 10 years ago in 2015. We
01:09:51
saw that AI was on the rise. We saw that this was happening. Um but at the time we were concerned that it was
01:09:57
essentially all locked up inside of Google. And so that would be bad arguably for the world, but certainly
01:10:03
for our companies, right? We we have thousands of companies. If our companies don't have access to that next wave of
01:10:10
technology, like we're going to be out of business. Um, and so open AI was kind of like a moonshot project that we were
01:10:16
actually going to take this out where it's not just locked up inside of How did you feel when they made it closed AI and and
01:10:23
you know I there was never specifically promised to be open source but I sure it was it was
01:10:29
if you go back it's it's a little bit but again I think what's most important is that we actually just have a lot of
01:10:35
choice right and and I certainly support open source because I think open source is the thing that you think open source wins. I think
01:10:41
we'll have both. Um, it seems it seems like the balance is that is that there's reasons to have both, but the importance
01:10:49
of having open source as an option forces all of the closed source vendors to be uh honest, right? Like if they
01:10:56
start if they start censoring the models, they start, you know, disabling too many abilities, then people will all
01:11:03
switch to the open source. Well, you worked at Google, you worked at Facebook. Oh, this was my question. Um,
01:11:08
Google has done an incredible job with their um you know ensemble of Gemini apps. I
01:11:13
mean Gemini models. Facebook has had some missteps with Llama. I'm just curious if you were the CEO of Facebook
01:11:21
today. Are they making the right bet or Google?
01:11:26
Well, I'm actually more curious about Facebook. Are they making the right bet with respect to just the talent war that's been created or is there a
01:11:33
different technological approach? You know, for example, the one thing that we talked about before was this concept of
01:11:38
the bitter lesson, which is always that compute overpowers humans. I don't know, how do you think about that or what
01:11:44
would you do if you were running that business today? I mean, I think he's he's doing what what needs to be done, right? Like Facebook has clearly fallen behind. Um,
01:11:51
and that's a real threat, right? Because Facebook actually competes with AI. Like people are switching from Instagram to
01:11:57
chat GPT. Like my kids are not on social media, they're talking to the AI. Um and
01:12:03
so if they it's fundamentally cannibalistic is what you're saying. Yes. Yes. Um so that's an interesting concept like it's
01:12:10
a finite amount of time and which is forget about the categories we put on them.
01:12:15
I mean the compound question ask a great agent is incredible. You know the way that you can speak. Yes. And that they're actually now with
01:12:22
Grock having the avatar kind of leaning into this concept of personality. We as old people in Gen Xers might be totally
01:12:29
missing the script. Right. Sure. Well, actually um so uh character AI is an example that actually gnome
01:12:36
made that bet and and uh Noom is a friend from Google who actually basically invented Transformers, right?
01:12:42
Um and then got frustrated that he couldn't launch anything at Google. So started character AI, but that was the
01:12:48
entire thing is making um characters that people want to talk to and so the usage on on characters. Yeah.
01:12:55
Well, thank you for being here. We're over Well, to be continued, we have to have you on the pod.
01:13:00
new fund. That's amazing. Yeah. Yeah. Congrat Congratulations on the new fund. Thank you. Yeah.
01:13:06
Thank you, Paul. Appreciate it. Oh, Keith. Hello. Keith is back here.
01:13:12
Look dragged in. Guys, boy, how are you? It's great to be here live. Everything
01:13:17
we've done has been remote over Zoom. Zoom. This is what you look like. Exactly. You look great. This is what you
01:13:23
been going to. Yeah, clearly. It's what 8% body fat looks like in I
01:13:28
know everything. Who's counting? Apparently the both of you.
01:13:34
[Music]
01:13:45
How are you, David? Good to see you, Keith. Great to be with you. Yeah. Hey, Kelly. Nice to see you. Oh my gosh. Jal, how are you?
01:13:51
Good. Good to see you. See you, Kelly. Thanks for being here today. Um, not sure you've been following the
01:13:57
panels, but a lot of conversations going on around AI, particularly around job
01:14:02
displacement. You're the 28 administrator of the SBA. Um, I
01:14:08
think more than half of the American workforce is employed by or are small business owners. You and I had a
01:14:15
conversation a week or so ago about what you're seeing on the ground with small businesses. In an AI workplace setting,
01:14:21
the conversation is always, are they going to get out competed? Are they going to get displaced? What's going to happen to American jobs and to the small
01:14:28
business? But what are you seeing on the ground and how does the SBA kind of associate with the the transition
01:14:33
underway? Yeah, Dave, first of all, great to be here. Look, a small business is big business in America, but small business
01:14:40
is big business for AI and I have been walking hundreds of factory floors for
01:14:46
the last 6 months. Most manufacturers in America are small businesses and without
01:14:53
AI, we would not be winning back these industries. And I will just tell you a
01:14:59
case in point. I actually bought a slide to show you workforce development in
01:15:04
action. Modern workforce, we call it the new collar boom. I don't know if they can put it up, but it's a factory in
01:15:11
Seymour, Indiana. It's a bike factory. We had lost the bike industry over the
01:15:16
last 30 years. Thousands of jobs, 98% imports. We're now for the first time in
01:15:22
this country building bikes in America because of AI, advanced manufacturing
01:15:28
techniques. Imagine we replicate this industry after industry and these are small business. This is a 60 person
01:15:34
factory in Seymour, Indiana, where they have no jobs. So it's a AI is a job creation machine
01:15:40
for reshoring, onshoring and advanced manufacturing. So manufacturing you're seeing a big heavy um influence
01:15:47
potential for kind of redefining. What about in the services businesses? What do you see there?
01:15:52
Across the board we have 7 and 12 million jobs open in America. Most of
01:15:57
them are open at small businesses. Number one concern of small business is a skilled workforce. That's because
01:16:03
President Trump solved inflation, regulation, taxes. Now they're saying, "Okay, we're booming. We've got $15
01:16:10
trillion of investment coming in. A lot of that's going to trickle down to small business. We need the skilled workforce.
01:16:16
So, President Trump is ensuring that we have that skilled workforce uh through some of his workforce initiatives, but
01:16:23
small business is going to be driving the AI boom from the bottom up.
01:16:28
And what um I guess what is needed for workforce training and transition.
01:16:33
Yeah, technology is going to be a big part of it. So when you think about go back to 1940 our workforce size was 56
01:16:41
million and people say well as technology advances our workforce gets competed away. Today our workforce is
01:16:48
170 million and compute power has been asmmptoic. So essentially 85% of the
01:16:54
jobs that exist today have been driven by advances in technology and only 40%
01:17:00
of the jobs that we had back in 1940 still exist today. So we are relying on
01:17:06
innovation as a job creation engine. It's just that people have a fear of the unknown and they're saying I I can't
01:17:12
envision what it is. Well, I can't envision what my life would have been like when I started a small business if
01:17:19
I could have had Figma or Canva instead of PowerPoint. Oh my gosh. Uh so just
01:17:25
these are uh we're going to create millions of solarreneurs who are going to have massive software companies or
01:17:30
manufacturing companies thanks to AI. Is there something the government can do the SBA for and what is the role of the
01:17:38
SBA? I mean I know one of the big focuses of this administration was to make government smaller. So is that a
01:17:43
goal you have to make government smaller and then maybe give the ability to give loans to the state. What what is the
01:17:49
role of the government in getting one and two person companies up and running if anything?
01:17:54
Well, the the mission of the SBA is to grow the economy and to support small
01:17:59
businesses and that's that's what we're doing. And the the last four years it had not been doing that. In fact, with
01:18:05
regard to AI, the Biden administration banned the use of SBA based loans for
01:18:11
use of purchasing technology in AI. I had the rules rewritten. So now small business entrepreneurs, solarreneurs up
01:18:18
to 500,500 person factories can use the proceeds of their loan toward AI
01:18:24
implementation, advanced manufacturing. Our goal is to get out of the way. Yeah. But but educate us on the loans
01:18:29
because we hear about that. But we're in venture capital where we have an incredible ecosystem of angel investors
01:18:34
doing this. How do SBA loans work? Who are they for? How much do do the American taxpayers, you know, uh put
01:18:42
into this? And and what's the result? Yeah, I'm glad you asked. So the SBA does not do direct lending. We span out
01:18:48
across network of thousands of banks in this country that offer SBA which are governmentbacked loans but we also
01:18:55
operate the small business innovation uh company guarantee that has been
01:19:00
responsible for backing many massive startups uh SBIC money was in Tesla for
01:19:07
example. So, we have an equity piece as well as uh the the SBA loans, but those
01:19:13
loans have to be repaid over 30 years, but they simply give small businesses that banks wouldn't normally lend to
01:19:20
that government guarantee that gives them the confidence. We do about 2,000 Main Street loans every single week. So
01:19:28
far this year, we are on pace for a record year because we've made the SBA
01:19:34
rights sized, which means we've taken it back to the preandemic size. It had doubled during the pandemic. 90% of the
01:19:41
employees were working from home, not focused on small business. We took it back down and the spending had doubled.
01:19:48
So, we took the spending down. We took the headcount back to pre- pandemic and now we have record level. You
01:19:54
have people showing up at the office. Oh, yeah. We're back every day. Wow. So the American taxpayers are paying people for a job and they're
01:20:00
doing it in an office. Not only that, outside of Washington, we sent them out to the field. Do you think that at some point you will
01:20:07
look at um either adding new types of SBA back loans or changing some of the
01:20:12
conditions to do as you said even further incentivize the investment in AI? Yes, absolutely. We are looking right
01:20:19
now at critical industries like medical metals, minerals, medical device
01:20:24
reshoring and onshoring. We have a massive at the SBA we're leading the make uh onshoring great again portal
01:20:31
which is a on the SBA website it's a resource of 1 million onshore manufacturers uh we're leading the made
01:20:38
in America charge so we focusing on smart manufacturing and looking at loan
01:20:43
types and we're trying to double the size of SBA loans so that for buying
01:20:48
advanced technology equipment CNC machines training that there are many
01:20:53
more resources available for how do you think about um energy. Yeah. On top of that,
01:20:59
yeah, I I was just talking to Secretary Bergam and Wright last night at the White House and we were talking about
01:21:04
the convergence of small business uh with the physical and the digital and
01:21:10
energy is going to be a big part for small business there because the innovation is going to be coming from
01:21:15
smaller businesses. And in manufacturing, you can be a small business and have 1500 employees, but
01:21:22
frankly, I'm seeing a lot of energy companies and others with 300 people.
01:21:27
So, small business is going to drive it. If you stipulate stipulate that there are 34 million small businesses in
01:21:34
America and 20,000 large companies, this is a small business-driven energy and AI
01:21:41
boom. Well, your your vision is something that some of the leading entrepreneurs in Silicon Valley have been pushing for as
01:21:46
well. this idea that there is an entire boom that will happen of soloreneurs,
01:21:52
the two and three person companies that are vibrant, successful, profitable, growing. Um, and what they just need is
01:21:58
a little bit of help at the edges potentially on maybe paying for some compute resources or whatever and then
01:22:04
they're off to the races. And that's certainly backed up by the data we have at the SBA. So 60% of the
01:22:10
40 of the 21 billion that we've lent this year have gone to companies with
01:22:15
one to five employees. So that's where the growth is coming. Uh certainly we
01:22:21
know that they're going to scale from there. Um but we're seeing all the trends say that putting more technology
01:22:27
into the hand of small businesses is growing the economy and and small business is still growing the jobs boom
01:22:34
in America. 720,000 jobs created this year led by small businesses.
01:22:40
Keith, I'm curious. You're a free markets guy. What What are your thoughts on the government's role in maybe
01:22:47
juicing up the this onshoring specifically in categories where maybe China has dominated for a couple of
01:22:53
decades? Well, as Kelly pointed out, the government's actually not extending the loans. The community banks in America are extending the loans. So, it really
01:23:00
isn't a deviation from free market principles. If you think about it, AI is really this rocket fuel to turbocharge
01:23:06
small businesses and entrepreneurs at least in three dimensions. First, f access access to information. Typically,
01:23:11
if you're starting a business, you have to compete with very large incumbents that have expertise in market research, marketing, legal, accounting. Now, tap
01:23:19
of, you know, tap of your fingers or your voice, you have the same expertise that all these large companies have. So, you've leveled the playing field.
01:23:25
Secondly, you have access to products like building an app. Like, everybody can compete with a large company.
01:23:31
anybody can code an app. So you're like air HVAC repair person, you have an app that's on par with a Shopify store or
01:23:37
better. Like that allows you to compete. So we're going to see more propellant there. And then third, you can save money. Like you used to have to have a
01:23:43
GNA team. Like you have accountants and you know bookkeepers and HR. AI can do all that. Maybe even do it better than
01:23:49
humans, but certainly at zero cost. So the economics of running a small business are going to be much better. A
01:23:55
the risk of running a small business, starting a small business is going to go down, which we're going to have increase. And then finally, you can save
01:24:01
money through things like RAM. You can use AI to audit your expenses and not waste 5 to 15%. Which will make you more
01:24:07
successful. So all these trends are going to combine and we're going to see in this administration an explosion of successful small businesses.
01:24:13
Does that mean that there's just more competitive forces in the marketplace? So big companies are going to now have
01:24:19
more competitors and it just ultimately drives net productivity gains long. Well, hopefully net productivity gains
01:24:25
and in so far as there some substitution, I suspect you wind up with a barbell. So the large largest players, the Nvidas of the world do benefit the
01:24:32
more people that run compute etc. But then I think that the smaller businesses actually eat at mid-market companies
01:24:39
because they can compete now and they've been at an economic disadvantage for decades. A and we're going to be in industries that
01:24:45
we couldn't have even imagined that we would be in. When people say why, you know, why do we need to make bikes in America America? Because it creates 60
01:24:53
great paying jobs in a tiny town in Indiana. People that want to do it. That's right. That's right. Uh PP PP and
01:25:00
E like whatever the you know during COVID pharmaceuticals we should be making that here. We can do that with
01:25:06
smart manufacturing with 100 people in the factory. You you must uh give the criteria or
01:25:11
some guidelines to the banks of how to pick and uh I'm assuming you take
01:25:17
diversity and inclusion and gender and all these important factors into uh account or do you do it based on merit?
01:25:25
Question is I was just trying to trigger the two of you. They they send that to your fund. Yes. I don't do any DI but jokingly. But
01:25:33
what's the criteria like when somebody comes and says I want to you know raise 100,000 and go to their local bank. How
01:25:39
do they get picked? Yeah. We have strict underwriting guidelines and we've stripped out the DEI that the last administration had put
01:25:45
in. They had a green lender initiative to preference where money went under the green new deal. I mean uh we've gone
01:25:52
back to saying if you qualify for these loans have at it. uh we're not going to pick winners and losers. We want
01:25:58
everyone to compete on a level playing field and have access to that capital. But what had happened under the last
01:26:03
administration, they'd lowered the underwriting guard rails. As a result, the loan loss portfolio the on the
01:26:09
portfolio went way up $400 million. We've reversed that, strengthened the underwriting standards to make sure that
01:26:15
the money goes to small businesses who are building these factories to onshore drones and pharmaceuticals and defense
01:26:21
and aerospace. What are the what are the target performance ratios in the loan portfolios? Oh my gosh. I mean our loss ratio should
01:26:28
be 3% or less and they are including on the SBA is one of the largest disaster
01:26:34
lenders in the country worthy recovery lender and they do well. very low very
01:26:39
low and in fact there's a secondary market for SBA loans because they perform well because of the strict
01:26:45
underwriting standards part of um oh zero subsidies sorry it it operates
01:26:50
at no cost to taxpayers when we enforce prudent underwriting standards which
01:26:55
we're getting back to that yes one of the things that um helps burnish entrepreneurship is imitation is the
01:27:01
sincerest form of flattery you must have so many successes is, but
01:27:08
they're not always well marketed or known, which would then pull other people to say, "Well, if they could do it, I could do it."
01:27:14
How do you think about that? In a world of social media and all of this, you've picked up on one of my key
01:27:20
problems. I run an agency that starts with the word small. Uh, small does not
01:27:25
mean insignificant. In fact, small business is significant. And President Trump and I talk about that all the
01:27:31
time. He loves small business. He knows the innovation starts there. The manufacturing is small business. So we
01:27:38
are working on a massive resetting of what the SBA does, but more importantly
01:27:43
what small business means to America. And I think people are waking up that Main Street is going mainstream and we
01:27:50
have to continue to push the understanding that if we don't protect our small businesses, our innovation
01:27:56
pipeline, our job creation engine is going to shut down. How do you interface with state agencies and state senators
01:28:04
and state governors who have 50 different views of the world but you know you're responsible for at least
01:28:12
supporting the underpinning of the business people that are there. How does that tension play out? It it's really important. In fact, we've
01:28:18
started an initiative where I'm meeting with governors across the country and their economic development uh
01:28:24
departments essentially because they know best what they need in their state. And if we can push more of this out of
01:28:30
Washington and say this needs to return to the states, they need to know the SBA is a resource for recruiting companies
01:28:37
into their state to create jobs in manufacturing like in my home state in Georgia that has done that. Uh so we're
01:28:43
going to continue to partner at the state and local and across the administration. I mean, having David
01:28:48
Saxs and this administration to be an ambassador for AI and crypto has been huge because it gives us a way to work
01:28:54
across the administration and then we can focus with the governors at the state level. Can I can I ask one question on that
01:29:00
because your comment was really striking that you guys have strong underwriting performance in the loan portfolio. There
01:29:06
are many other insurance programs across the federal government that do not have good underwriting standards and run a
01:29:13
terrible loss ratios and they're highly inefficient for the taxpayer and then they cause all of these market inefficiencies as a result and I won't
01:29:20
start to name them but you know who they are given your background financial services and fintech your experience
01:29:25
here is there an opportunity do you get drawn in and is there an opportunity to go in and try and address some of these other very very very large insurance
01:29:32
programs and underwriting programs that the federal government operates Dave I think there is because we've recruited to the SBA really an elite
01:29:39
group of financial services leaders who understand this. I I served in the Senate previously in the US Senate. I
01:29:45
was the only CFA to have ever served in Congress and I found out when I went to
01:29:50
Washington Congress ever ever. They don't like people with financial services experience in
01:29:56
Washington because we know how to read a P&L and um but yeah, so we're bringing that discipline. We're happy to share
01:30:02
it. Very open source. Please do. Um, yeah. So, like like you say, we've open sourced it and the fans have just gone
01:30:08
crazy. So, Kelly, we have a uh we have a word for small businesses in our community. It's called Startups. Maybe it's time to
01:30:14
rebrand the SBA. I'm completely open to it. That's right. Uh I was going to call it Main Street
01:30:22
Manufacturing, but I like I like startups a lot, too. I love that because by the way the point the point about China we made earlier they have there's
01:30:28
3 million factories in China but these aren't massive 100 400 acre facilities.
01:30:34
These are very often small warehouses that were turned into a small manufacturing facility and we could
01:30:40
recreate that in America across all of these great states where people are looking for economic expansion
01:30:46
over and over. Well David actually many of those opportunities exist but uh people don't know that they can find
01:30:52
local sourcing. So, what's now possible through AI, you can say, "I have this product, and historically, I've got it through China, or I've got it through
01:30:58
Indonesia or whatever. I want a US-based manufacturer." Yeah. And you can use AI to go across the
01:31:03
entire country and find local manufacturers. There's there's almost always a choice in the United States.
01:31:09
It's just people don't know where to find them and how to negotiate with them and how to get in touch with them even. And so, that's a solved problem now
01:31:15
through AI. Yeah. Yeah. Yeah. Do you think that there's um a place where
01:31:22
the SBA maybe in partnership with the White House says here are these industries that frankly are just a
01:31:28
little bit more important or kinds of companies that maybe are just a little bit more important for a bunch of strategic reasons where maybe in you
01:31:37
relax the underwriting criteria or you just try to get a lot more people on the field, chips on the table. How do you
01:31:44
how do you think about that? Well, first of all, I'm a taxpayer champion because as a small business person, I know that
01:31:50
small businesses are taxpayers, too. And we can't put some small businesses on the hook for other small businesses. So,
01:31:55
we've got to have an efficient market that discovers the right funding mechanism. So, we're looking at making
01:32:02
sure that we're we have the right underwriting standards for critical industries. Uh, as we're working on some
01:32:09
things with Department of Defense right now. uh we have our SBIC program that we're experimenting with some different
01:32:15
equity structures. So there's more to come on that. I think financial engineering is important but we have to
01:32:22
first and foremost not put taxpayers on the hook for it. I think that's a really interesting point. Uh the equity structures if you
01:32:29
look at Celindra Tesla and that cohort uh Tesla paid back their loan with
01:32:35
interest early. That's right. If the government had gotten just 10% of that in equity,
01:32:42
that would have paid for a hundred cylinders and mistakes. So some equity
01:32:48
component or warrants could could change the SBA into, you know, having and the
01:32:53
American taxpayer by proxy having some upside in these investments. Yeah. Yes. Taxpayers have all the downside and
01:32:59
none of the upside. You like you like the taxpayers having equity? It's tricky. It's more complicated than that. you know, you have adverse
01:33:05
selection issues and a b it it it's not a one-sizefits-all and it's all good, but having flexibility for certain
01:33:13
industries to have a different corporate structure or different investment structure is, you know, paro optimal.
01:33:18
You you don't do that at all today, Kelly, at SBA. Uh, not today. So, very plain vanilla.
01:33:24
Um, but we're continuing to have the conversations about how to be creative, particularly around defense critical
01:33:29
technologies. There's a lot to do there and we need to do it very quickly and there's some great success stories that
01:33:34
we can replicate and they may not even require massive re-engineering. In the last few minutes, Kelly, can you
01:33:40
give us a very quick contrast? Your life as a senator versus your life as a head of the SBA.
01:33:46
Well, I'd much rather be an executive than a politician. So, I I was humbled and honored to serve in the Senate. As a
01:33:53
kid that grew up on a farm and the first in my family to graduate from college, it was amazing. Um, but being able to
01:34:00
run this agency, which at 7,000 people is considered small, um, is amazing, but
01:34:06
I'm really an entrepreneur and a businesswoman at heart. So, I'm approaching this as a businesswoman, a
01:34:11
service to taxpayers, the government, and I'm incredibly blessed to be able to do it. So, I I love it. Thank you for doing it. Please join us
01:34:17
in thanking Kelly. Thank you. Great to be with you. Thank
01:34:23
you.