The Coming Global AI Conflict w/ Gilman Louie | EP #54

How do you see these two superpowers?

They actually have a timetable. They want to be in a leadership position as the global AI leader. We got there first. We now face a competitor nation.

Are we moving fast enough?

The answer is no, we're not moving fast enough. Their view of alignment and my view of alignment are very different. Is it going to be fundamentally Chinese, American, fundamentally European?

It scares me sometimes.

It will change everything. The way we live, the way we play, the way we have relationships, the way we learn.

Right, Everything.

The goal is to lead, not to win, because winning defines a finite game. We're in an endless game. I think this is our AI moment.

Welcome to Moonshots and Mindsets. One of the conversations that is going on right now is, is the United States prepared for the speed at which AI is going to transform everything, including governance? And is China better situated to utilize AI? I brought on stage this year at Abundance 360, the Honorable Gilman Louie to have that conversation. A look at AI in the US Government versus AI in China. Gilman is the past CEO of In Q Tel, the venture capital arm of the CIA. Today he's the CEO of America's Frontier Fund, which is effectively our sovereign wealth fund here in the United States. It's an amazing conversation and one that we need to be thinking about as the speed of AI is accelerating. And it's going to have some extraordinary impact on who and how we run this planet. All right, check out this conversation from March 2023 at my private summit, Abundance360, with the Honorable Gilman Louie on the future of AI and governance.

I'm thrilled to have you here and thank you for your willingness to take on this conversation because it's a very real conversation in the world today. Right. You have been overseeing so much, and we're in the midst of these tumultuous times and, you know, we don't talk about the downside, dystopian side here and a360 too much. But we need to talk about reality. And part of reality is the fact that AI is going to drive a lot of change in the world. It was Alex who said. I think it was Alex who said, whoever's the dominant AI players in the world in the next five years will remain the dominant player. We're playing for keeps in some ways here. So if you don't mind, paint the picture for me. Where are we in the AI USA and AI China relationship here? How much is the US Government really leading and how much is the Chinese government really leading? How do you see these two superpowers?

Well, I mean, that's a great question. I think first of all, if anybody spends any time understanding AI, it is not because of what either governments are doing. Most of the AI innovation work is happening in academia in our companies though both countries have seeded a lot of AI research. We do it through our organizations like darpa, iarpa, our National Science foundation, our national labs. China does it with its national champions, with its university systems and its regional organizations. I think the real question and the challenge for both countries is it is clear that both countries view that this is a competitive race. It's the issue of winner takes all and what that really, really means. China kind of led first in the world to go out there and said as a nation for where they wanted to go, that AI was going to be critical. They actually have a timetable. They said by 2030 they want to be in a leadership position as the global AI leader.

Right.

I used to take a group to China every year up until things started devolving. But it was clear that the party and the mayors and governors were making huge capital investments. They were recruiting, they were creating policies, they were creating cities that were focused on AI. Do we see the US doing the same?

Well, I think right now because of the current situation, it's not just because of ChatGPT, the, that has accelerated things. The US kind of woke up and said, you know, we now face a competitor nation.

Right.

That's committed, that has integrated this as a part of a national strategy.

Is this the equivalent of the US versus Russia in the space race?

Absolutely. They had their Sputnik moment when, you know, AlphaGo beat the best gold champion.

Yes, right.

Lisa doll.

Right, right.

So that was their moment. In fact, even though I participated in this great study in this commission, a 750 page report that we put out two and a half.

I'm going to have ChatGPT read.

It's just that thick and chat can probably compress it down to a few pages. But I think this is our AI moment. This is our Sputnik moment. People are waking up in Washington and looking at ChatGPT and are saying to themselves, what does this mean for our economy, for national defense? How should we look at these particular technologies in a much more profoundly different way than we looked at narrow AI?

Yeah. By the way, once again, please put your questions into the chat and upvote them. I'll be going to your questions here in a little bit. I mean, listen, I'm thankful I'm a libertarian capitalist. I'M thankful to be American. I love this country. But when I think about linear sclerotic systems, I think about the government. And so the notion, it scares me sometimes. I'm super happy that we have amazing entrepreneurs like we just saw on the stage here who are able to do that. You're investing and I'll get to that in a little bit. But do you think that are you seeing from within the President's AI commission and so forth, are you seeing the internal attention and awareness and significance here is that are we moving fast enough?

The answer is no, we're not moving fast enough.

Okay.

The governments have a very difficult time dealing with nonlinear change functions. Yes, Logarithmic change. They can't understand how to put that in perspective. And so. And it's not just the US government or the Chinese government. I just got back from Brussels, right? The EU is trying to wrestle with this now. All three countries at some level is afraid of particularly the new technologies around generative AI and other AI technologies. That's on the bubble because unlike narrow AI, where we can focus it on the problem, we can't necessarily control its outputs. And that's frightening for governments. So in Brussels, what they want to do, at least some in the commission, they want to regulate it, right? So the first thing they do, they want to regulate it.

That's the first reaction of government always.

On the US side, it's let's create tort law, basically sue people, right? It's the American way. So what we do. And then the Chinese are kind of go like this might come out and say things that are not on the party line. You have to understand that kind of cultural realities of government as they look at these technologies. And all three are wrong. I can give this example. I got this phone call, I won't say from whom, and said, we got to stop this thing, we got to do a timeout. It's like stem cells all over again. And here's what we need to do. We need to call a moratorium until we figure this thing out, right? Try to explain. It's already out there. There are these companies, the code is totally available, it's open source. You know, that strategy is not going to work. But they said, but it acts like a nine year old, right? It's like it says stupid things, but it can put together dangerous combinations. You can build weapons systems off of this. And I said, look, if your 9 year old acts out in your house, she throws food, right? Kicks, kicks things around, throws a temper tantrum, do you kill it.

Right.

No, you train it. You train her to be smarter. You put some rules around it. You get her to think about what are the ethical considerations of what it's doing. Our goal, all the countries, even our competitors nations, needs to use AI in a way and mature it in a way so that it gets smarter, not dumber. You know, one of the lessons we learn in the Ukraine right now.

Yes.

Is dumber is deadly.

Right.

So this idea that I know there's been a huge debate about how AI is used on the battlefield, but dumber systems leads to higher casualties.

Right.

And worse.

Non intentional.

And so, you know, generally smarter is better. Same thing is true with AI.

This episode is brought to you by Levels. One of the most important things that.

I do to try and maintain my peak vitality and longevity is to monitor my blood glucose. More importantly, the foods that I eat and how they peak the glucose levels in my blood. Now, glucose is the fuel that powers your brain. It's really important. High, prolonged levels of glucose, what's called hyperglycemia, leads to everything from heart disease to Alzheimer's to sexual dysfunction to diabetes. And it's not good. The challenge is all of us are different. All of us respond to different foods in different ways. Like for me, if I eat bananas, it spikes my blood glucose. If I eat grapes, it doesn't. If I eat bread by itself, I get this prolonged spike in my blood glucose levels. But if I dip that bread in olive oil, it blunts it. And these are things that I've learned from wearing a continuous glucose monitor and using the Levels app. So Levels is a company that helps you in analyzing what's going on in your body. It's continuous monitoring 24 7. I wear it all the time. Really helps me to stay on top of the food I eat, remain conscious of the food that I eat, and to understand which foods affect me based upon my physiology and my genetics. You know, on this podcast, I only recommend products and services that I use that I use not only for myself and my friends and my family that I think are high quality and safe and really impact a person's life. So check it out. Levels link. Peter give you two additional months of membership. And it's something that I think everyone should be doing eventually. This stuff is going to be in your body, on your body. Part of our future of medicine today. It's a product that I think I'm going to be using for the years ahead and hope you'll consider as well.

Fascinating. You know, I'm going to ask you the Question ultimately here of who do you think is likely to win the AI race between us and China? But before I do that, when I think about communism and socialism as experiments run, you know, over the last century, one of the reasons they failed is because they were. You couldn't run that system without having actual knowledge.

Yeah.

Could communism or socialism, where socialism is basically technological socialism, technology is taking care of you and communism is understanding the markets better than the markets forces could be. Are those systems potentially better structured for an AI world?

So it's pretty clear that authoritarian governments or socialist communist governments want to use AI to better serve its public.

Right.

I mean we have a very western point of view.

Right.

Which is we're kind of grounded in the American style democracy, which I think is a good thing. I think ultimately that's the better model. But if you're in China, by the.

Way, it could be made better with AI.

Oh absolutely. All systems can be made better with AI. But if you want to deliver higher quality, the things that any government, democratic or authoritarian worries at most is unhappy populations. Unhappy populations lead to revolt, revolution and revolt.

Right.

So China's putting a lot of effort of using their AI to deliver better services.

Right.

So they put a lot of energy in things like smart cities, you know, making sure their garbage is picked up on time, making sure that the public workers are actually doing their jobs. They're doing some other stuff that I find not necessarily great. The social scoring that, you know, the, the monitoring and tracking. But again, culturally, in China, this issue around using technologies for harmony.

Yes, Harmony is making sure increasing happiness.

Increasing happiness and alignment. It's just their view of alignment and my view of alignment are very different.

Right. So how is it, how important is it for the, do you think for the US to win this AI China arms race? If I were to polarize it that, that way and, and where do you think we are in that? What would the US have to do to get there?

So we're playing this big game of chess or go, depending on your favorite game. And we're in their very early moves. So I think winning and losing is the wrong conceptual framework. It's really about leadership. So if you kind of think about. You heard some great speakers today talking about, you know, data sets and algorithms and open source data. It is what platforms are companies and countries going to build their AI driven applications on top of. In the old days, all of us technologists, because I'm kind of old, we thought technology.

You're young. And we'll have a longevity conversation later.

Yeah, I wrote 120, by the way.

Okay, good.

But the challenge that we have is this concept that.

By the way, what did you write as your number 120? What's that? I know, I know, for your age. But what about one through five? About how you feel about AI? Oh, five, five. Okay, great.

Right?

I mean, it will change everything. Yeah, everything. The way we live, the way we play, the way we have relationships, the way we learn. Right, everything. And it is going to be such a global accelerator. It'll make everything go non linear all at the same time, everywhere, always. But here's the thing, we're at visical 1.0.

Yes, right.

I mean literally, this is like Windows 10.

For those of you who don't know what that is, that's like really early days.

Really early days. But I go back to the framework of how these countries kind of look at this technology and what winning means. Whoever puts the data sets together, whoever puts the training sets together, whoever builds these models out can't help but building bias into those systems. This idea that systems are not biased is impossible because bias is how we do judgment. It's the thing that we go, it's our shorthand of how we got like.

Shortcuts in the brain.

Shortcuts in the brain. And AI is going to develop those same shortcuts. And so.

But we want them biased the way, we want them biased in a positive.

Way, the way we view. So the competition between China and the US is not just who has a better missile system or who can shoot down the other guy's hypersonics faster. All that's really, really important. But how we build all of our information systems and what cultural biases we put into there. Is it going to be fundamentally Chinese, fundamentally American, fundamentally European? Those cultural biases are going to be built in, at least in this current generation of generative AI. It's going to be built into those systems.

Let me remind you, please put your questions into slido and we can bring slido up when there's some. So let's bring it on back here and we'll get some of the participation. Can AI be regulated? I've been dying to ask you this question. I mean, there are regulators who want to regulate it through whatever mechanism. But is it even regulatable?

Well, clearly people will try to regulate it.

Yes, right.

For example, they're going to try to regulate it so you can't build the next toxin to kill everybody.

Sure, you want regulations like that?

There will be regulation. But I think this concept that we can put the genie back in a bottle is false. That is just not going to happen. If anything, we got to move faster, not slower. Because the danger point is between that 9 year old and that 30 year old.

I love that. I think about AI is about to enter adolescence. It's where you're strong enough, but you don't know your strength yet.

Yeah, you're kind of stupid.

And the frontal lobe hasn't evolved yet.

Right.

My kids are 11. I'm told it's going to be blurry for a few years and then they wake up in their early 20s and.

We'Re going to use that blurriness to our advantage. You know, people talk about AI, particularly generative AI, hallucinating. It's a bad thing.

Yes.

But if you actually look at what is hallucinating is actually its form of creativity.

Right.

We just need to, you know, I give you a great example. I just said name all the awards that Gilman Louie has won. So it lists like a bunch of my CIA medals and then three other medals I don't have.

Right.

So one of them was the Presidential Medal for Innovation. So I took it to the White House and asked them to deliver the medal next week to myself.

Right.

So we're seeing these AIs be super creative. We just think it's a hallucination. We're not leveraging the power of these pattern recognitions to our advantage.

Let me go to a couple of questions here. What is the likelihood of stacking AI on blockchain to keep government transparent? A fact based scoreboard.

So there are a couple of problems. First of all, blockchain is hugely energy inefficient, which I have my own challenges around that. And the security models itself will last for a little while. I know we have a bunch of quantum resistant algorithms who we're playing with.

We have. That's tomorrow night is quantum computer.

But I would say that question might be grounded in this idea that AI fundamentally is explainable. That's kind of like clean coal. When I asked my child how did she come up with that idea to eat that candy all at one time at the dinner table, she says, I don't know, it's the neural net. So she doesn't know what neuron fired at what time to make that decision. AIs are like that now. Can it be more transparent? Yes. Can we do things to make it more understanding, kind of how it got there. But at the end of the day, neural networks are neural nets. And anybody who understands that space understands it's Statistical, it's not deterministic.

McMahon from our singularity community asked why have we not heard of a Chinese ChatGPT competitor? That's interesting.

Well, they're clearly working on one. Many Chinese companies are working on this. Again, we got there first. Yeah, it's sort of like vaccines until we got there first.

Right.

It doesn't mean that we'll continue to provide that leadership going forward. I think we have to give the Chinese companies and government credit for being able to actually apply these technologies faster than we do.

Right.

So that's why for, for the US and for. For the Western countries, moving faster is going to be more important because if we don't, those foundational tech that all of that foundational AI could easily be built upon Chinese technologies.

A few questions. You previously served as the CEO of In Q Tel, which is an extraordinary venture fund, the venture fund of the CIA, which I always thought was amazing as a concept. I'm curious. One of the subjects we talk about here last year, we'll talk about again next year, is the field of brain computer interface. Is in Qitel investing in bci.

Well, I'm going to go specifically in qtel because I'm not actively involved in, in Qitel anymore. But this is an area of interest and it has been in the interest for 30 years.

I would imagine every defense organization is going to try to increase intelligence.

Well, here's our problem, right. Our problem is our mental models is based on what humans do. So all of our data is based on what are we call these mark ones. Our eyeballs can see what we can hear, what we can understand. We design all of our information that are human centric. And the problem with that model is humans become. It's the limitation.

Right.

If you're going to design systems in the future, you got to break out of those sensory things a direct brain interface. Now, why is the military interested in it? Well, for obvious reasons, Right. If I'm underneath nine GS, my head weighs 2,000 pounds, I can't move my hands. Thinking about what that system should do for it to do is probably better than me having that physically try to reach out, you know, at 9G to pull that aside also because the stakes are so much higher. You know, it's one thing for an AI to make a bunch of mistakes on, you know, picking a stock is another thing. If the AI picks the wrong things, that ultimately leads to somebody's death.

Yeah. Right.

So it sticks them much higher in the middle. Let's talk to your.

Let's talk about your, your newest venture, which is amazing, America's Frontier Fund. I've heard it described as the America's sovereign wealth fund. And I've always wondered, why doesn't the US Have a sovereign wealth fund like Saudi has PIF and many other countries? So talk to us. What is America's frontier?

So our view is particularly because it's revolution in AI.

Right.

That because AI modeling and simulation, digital twins, as we call them.

Right.

It's moving along at such a rapid rate that those technologies are going to revolutionize foundational or frontier technologies for the next decade. So things that might have seemed 20 years off are now only going to be 5 to 10 years off. You can talk about quantum. Quantum is always 20 years away. Fusion is always 20 years away. Being able to decode and actually simulate a human cell is always 20 years away until now.

Right.

And so America's Frontier Fund is to make sure that the US Begins to actually invest in the technologies that are going to transform the world. It's not just the AI. It's the application of AI to improve all of humanity.

Everybody, this is Peter. A quick break from the episode. I'm a firm believer that science and technology and how entrepreneurs can change the world is the only real news out there worth consuming. I don't watch the Crisis News Network. I call CNN or FOX and hear every devastating piece of news on the planet. I spend my time training my neural net the way I see the world by looking at the incredible breakthroughs in science and technology, how entrepreneurs are solving the world's grand challenges, what the breakthroughs are in longevity, how exponential technologies are transforming our world. So twice a week I put out a blog. One blog is looking at the future of longevity, age reversal, biotech, increasing your health span. The other blog looks at exponential technologies. AI 3D printing, synthetic biology, AR.VR blockchain. These technologies are transforming what you as an entrepreneur can do. If this is the kind of muse you want to learn about and shape your neural nets with, go to dmandis.com and learn more. Now back to the episode.

We'll go to these last two questions here in our four minutes. How can we as taxpayers advocate for the development of AI technologies and the race to embrace these technologies in the US to maintain our competitive advantage? So should should American citizens be saying to the White House and Congress, guys, you got to get, you got to get in the game here?

Well, absolutely. And look, the Defense Department was the first department of any military in the world to actually create its own AI ethics, which is amazing. Trust me, I was in there when the sausage was made. You don't want to know how we made that sausage. But we said this, if we're going to have leadership in the world, the goal is to lead, not to win.

Right.

I love that distinction.

Lead, not win. Because winning defines a finite game. It's like video games. We're in an endless game. Hopefully it's endless. Stupid.

The only way to win, stupid, not to play.

Right, exactly.

So in that particular case, right. I think the public needs to demand from the government what are the frameworks of how we should be using these technologies.

So you know, I remember in the, in the 19 late 80s when I was in medical school, I was playing with restriction enzymes, the first ability to chop up DNA and there were these very famous asilomar conferences where all the gene jockeys got together and they self regulated. Yeah, before the government regulated them. Is that going on in the AI world now?

I think the AI companies are beginning to get real about this and our view is this. And we try to communicate, we being, you know, some of us who are trying to wear our government hats talking to our companies and that is if you don't have some level of self regulation, right. And responsible use, then somebody's going to tell you responsible use and regulate you. So in the European world that means ban, Right. If the European courts make a ruling, that's it, there's no appeal.

Right.

That's not like the political system here. You just wait for the next administration to show up to go again.

Right.

And the same thing that's on the Chinese side, right. If you do the wrong thing, you're out of there. You might be gone for six months and come back later as a high school teacher. So, so stakes are high. It's really, really high. And so companies, we cannot, I'll be really tough on my technology companies in America.

Right?

Please.

We cannot have this idea that we're US centric first when it deals with the rest of the world. Because when we go into a place like Europe and we say we don't consider their values, what they've gone through as a group of countries and we apply just blindly, right. Our ethical overlays on those models, it doesn't work.

Right.

And so, so we culturally have to be much more open minded in understanding the baselines and maybe ChatGPT could help our diplomacy a little bit better and dealing with these particular issues. And when it comes to company diplomacy with countries overseas, one more question from.

Slido here in the space race, the public rallied behind the government's investment, put people on the moon. Do you think we might see that in the race to win AI or. I'll translate it in the race to lead humanity in AI.

Already seen it.

Right.

750 page report. Two and a half years we had the CEO of Oracle, the former CEO of Google, the current CEO of Amazon, a bunch of scientists. There were 15 of us who put together the 750 page report. Now here's what happens. Some 150 page reports in the government you'd stick on the shelf and you move on to the next commission.

Right.

That didn't happen this time. This time Congress, both Republicans and Democrats, embraced that report. We put together over $250 billion of spending, some of it you're seeing in the Chips act and many other areas. As a country, we're going to spend $1.6 trillion on technology. So we haven't seen this kind of spending since the Apollo Program.

Right.

So we want to win, right? Not win in the sense of, you know, winner, take off. We want to win this leadership competition.

I'm going to use that. It's a beautiful.

Because that's what we got to do.

Let me close on this question. The America Frontiers Fund, your fund, what are you investing in now?

We are. We're just setting up.

How big a fund is this $500 million fund? Is this fund that people can invest into or not?

People can, if you're qualified, invest into the fund, though. Much of our funding will come from states and pension programs. But we are doing a couple things a little bit differently. The first thing that we're doing differently, and you heard a little bit about this from the other speakers, is you have to build these technologies with intention because the capital cost is so expensive. So it means rather than simply coming up with incubators, you actually had to do company building, matching technologists with scientists with people who know how to run companies. The second thing you have to do is, look, our coasts are great. Generated 90% of our innovation jobs for the last 10 years. Five cities, right. That's not an economic model that's viable for the next 20 years. We got involved the rest of the country and what's interesting is if you look at these key foundational technologies, these frontier technologies, they're being done in places like Purdue, Georgia Tech.

Right.

They're being built in Austin, in places like New Mexico, in Arizona, in Colorado.

Right.

Ohio, you know, in places like Taylor, Texas. That's the foundation of our next wave. It doesn't mean that the coasts aren't going to be a participant. But those economies are overcooked right now.

How long you'd be with us? When do you have to leave?

I'm going to have to leave tomorrow morning. Unfortunately. I've been called by a governor. I won't tell you which day, but.

You'Re here tonight to join us for the celebration of the Tech up. Let's give it up for the honorable Gilman Louie.