Audio Player
Starting at:
Preparing for AGI with National Defense Researcher | Thomas Pike
May 21, 2024
•
53:24
•
undefined
Audio:
Download MP3
ℹ️ Timestamps visible: Timestamps may be inaccurate if the MP3 has dynamically injected ads. Hide timestamps.
Transcript
Enhanced with Timestamps
131 sentences
8,811 words
Method: api-polled
Transcription time: 52m 44s
The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze.
Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates.
Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a toe listener, you get a special discount. Head over to economist.com slash TOE to subscribe. That's economist.com slash TOE for your discount.
Hola, Miami! When's the last time you've been to Burlington? We've updated, organized, and added fresh fashion. See for yourself Friday, November 14th to Sunday, November 16th at our Big Deal event. You can enter for a chance to win free wawa gas for a year, plus more surprises in your Burlington. Miami, that means so many ways and days to save. Burlington. Deals. Brands. Wow! No purchase necessary. Visit BigDealEvent.com for more details.
As a democracy in crisis, how's this going to tip into the future? So your signals have changed, the butterfly has kind of flapped its wings. Will we tip into another strange attractor? Because we could take a student and, you know, a 10-week class, take it from almost no coding, you know, to building a facial recognition with a support vector machine. Like, that's amazing.
Thomas Pike is a faculty at National Intelligence University. This talk was given at MindFest, put on by the Center for the Future Mind, which is spearheaded by Professor of Philosophy Susan Schneider. It's a conference that's annually held where they merge artificial intelligence and consciousness studies and held at Florida Atlantic University. The links to all of these will be in the description. There's also a playlist here for MindFest. Again, that's that conference, Merging AI and Consciousness. There are previous talks from people like Scott Aronson, David Chalmers,
Stuart Hameroff, Sarah Walker, Stephen Wolfram, and Ben Gortzel.
From that perspective, an analytical one, but as well as a philosophical one, discerning, well, what's consciousness's relationship to fundamental reality? What is reality? Are the laws as they exist even the laws and should they be mathematical? But instead, I was invited down to film these talks and bring them to you courtesy of the Center for the Future Mind. Enjoy this talk from MindFest.
This morning, it's my distinct pleasure to introduce Dr. Tom Pike. He is the Dean of the Ottinger School of Science and Technology Intelligence at the National Intelligence University. Today, he's got a talk planned for talking about how complex systems can alter the course of evolution of artificial intelligence. Really looking forward to this talk, Tom. So when you're ready, please take it away. Thank you very much, Eric. All right. So hello, Tom Pike. So first,
Take care of some mandatory stuff, so my views do not reflect those of the U.S. Army, National Intelligence University, or U.S. government. And then the second bit to kind of deal with the elephant in the room is intelligence is often associated with bluntly jackbooted thugs like the Gestapo, KGB, things of that nature.
Right. And so when people hear intelligence, they usually think, you know, like, like in the US where they're putting LSD in the water in San Francisco, and we have the pike and church committees. Right. And so although there will always be a need for secrets, and that's always going to be paramount to even your relationship with your neighbors, right, it's important to understand there's a whole nother aspect, which is just trying to understand how the world works. Right. And if you look at Afghanistan, in particular, as the most recent example,
That wasn't a failure because we didn't have enough secrets on how the Taliban were operating. That was because we don't understand really how to grow a democracy. We don't understand how to get those critical dynamics into place in order to have a functioning, productive society that respects humans' rights. So there's intelligence, the secrets part, really my focus or all the work I do with Dr. Bailey is on how do we just understand the world? And that's all in the open.
Right. And so we do it in the open. I do open source development. I do things of that nature. So we say intelligence in this part really say it's more like we're just trying to understand how the world works. Right. And this is a look at Director Haynes. She did an op-ed in the Wall Street Journal about two years ago where she was saying like, hey, you know, we as a democracy, we need less secrets.
Hi, all right.
So, I'm going to use this quote from Scott Page, the Santa Fe Institute, the University of Michigan, right? Which is like, the right perspective makes hard problems easy, while the wrong perspective makes easy problems hard, right? So, what I'm going to do is like yesterday, we had Origin of Life from Sarah Walker, we had lots of talks about consciousness and quantum mechanics, right? So, what I'm going to look at is more like societies, it's kind of like a meta-consciousness, right? And so, as you look at this, as we've gone through life,
All right, as humanities, you know, kind of evolved over the last couple thousand years, right, is that we found superior perspectives that then makes our life better, right? So look at the Copernican view of the universe, you know, theory of, you know, theory of bacteria, gravity, all those things, you know, Newton and whatnot made our lives better as we found those perspectives, right? And we were able to share those perspectives. Now humanity could say, oh, this actually really works. I can take that work and continue to build on it.
I'm doing a specific story on it, but this comes from a book by Hilton Root out of George Mason University called Network Origins of the Global Economy.
What's interesting about this is the Chinese developed the printing press 400 years before Gutenberg did in Europe. Not only did they invent it 400 years before Gutenberg, but they stopped using it 80 years before Gutenberg. They said, this knowledge replication technology, if you will, nothing to see here.
We don't need it. Then you go to Gutenberg in Europe, which is this massive small world network of competing fiefdoms that are forming alliances and getting in wars with each other and stuff like that. It becomes this massive catalyst for the Reformation, the scientific enlightenment, and really Western Europe just blowing up as kind of like a global powerhouse.
All right. And so in Hilton Root's book, right, the way the conclusion they make is like was how that network was functioning. It's the differences between those.
Right. So you look at the Chinese network under the Chinese emperor at the time. Right. It was a hub and spoke network. And so the bureaucratic elite were incentivized, you know, to maintain their status. And if somebody didn't want that technology to get out, they could cut it off. Right. Well, in Europe, because it was a small world network, people like Galileo Copernicus could sneak their their writings that, you know, the earth is actually not the center of the universe. Like only paths, the Catholic Church into like places like the Netherlands.
Right. All that sun, that knowledge was getting out there and spreading. Right. And so it became this, you know, and it just blew up. Right. For good and bad. Right. And so if we look at this, you know, much like a large language model or a deep learning network. Right. Is I look at humanity's collective consciousness is like a fractal problem solving network. You have all these folks all over the world, particularly like the open source community. Right. That this is this global problem solving network.
And it's in a very real way functioning just like a neural network. So they're just searching this hyperdimensional landscape of possibilities, as Sarah Walker talked about yesterday, and finding new perspectives that help us understand the world and engage with it better.
And so our knowledge storing and sharing technology are artifacts. Herbert Simon, if you're familiar with the sciences of the artificial, are ways to help us optimize our Explore, Exploit search mechanism. Just because it's complexity,
Big fan of complexity theory, like Sarah Walker said yesterday, right? She's effectively came up with assembly theory, right? We know from science like rigor really matters, right? And how do we test this and make it, you know, how do we have rigor to this new theory? But she's effectively proposing a new way to look at the world that defies our traditional physics types measurements.
Alright and so you know potentially that will blow up and give us a whole new ways to interpret and understand the world right give us a new superior perspective solve these problems are finding really hard right now will become easy.
After we look at the printing presses, things have got a lot better in numerous areas. Increased lifespan, child mortality has plummeted, improved health care, for many improved quality of life, increased education, probably the most literate society that we've ever had, but it's also come with some absolutely horrible things.
You have slavery and child labor, indigenous people's genocide, pollution. Right now, we're looking at the current rise of autocrats, based off the dynamics of real-time, in my opinion, the dynamics of real-time bidding and the need to optimize wealth functions. It also comes with a lot of bad. Those are the things we want to try and prevent if we're going through another great revolution.
The idea from this is that we really have a new type of printing press, if you will, right? Knowledge storing and sharing technology, right? So you have that if you're not familiar with this, right? If you don't code, just give you a quick rundown, right? So you have the internet, right? Where you can share knowledge globally, right? So I work on a shameless plug, but I work on Mesa, agent-based modeling in Python, right? And we got people from all over the world, particularly right now, the Technical University of Delft that are contributing to it and making it vastly better. Like I can't even keep up.
Right. And so we're able to store that in GitHub on the left. There's also GitLab, Bitbucket. There's all these like knowledge coding repositories that make it much easier to say like, well, this was your idea. I just made this change to it. Now it's much better. Right. And you can instantly merge it and then push it out to the entire world. Right. And then you get these like naturally selected libraries where somebody has a really good library.
or somebody has a really good idea, then somebody else can go in there and use it and take that knowledge. So I think about like a year and a half ago before I got stuck in academic administration, we built a crop simulator, look at famines in West Africa, and I know Jack about crop growth, but I was able to find a reputable library for a person in England, and we could reliably show, based off the current weather patterns in West Africa,
This is the dangers of a famine based off these kind of staple crops in Niger. So you're able to find those. You got things like Scikit-learn, TensorFlow, PyTorch. Now we're getting like hugging face and they'll have different large language model attributes. So we can now take that knowledge. So I don't have to spend years doing it. I can rapidly apply it, which even comes with its own problems. If you don't understand what you're doing, you start applying tools that can give you answers that are appealing, but somewhat wrong.
So now the real point is we're in this next great revolution of knowledge storing and sharing. I personally contend this is a real revolution that's driving AI. It's not that if it was really stuck at these boutique firms or universities, it would not have blown up like it has. But because we could take a student and a 10-week class, take it from almost no coding to building a facial recognition with a support vector machine,
like that's that's amazing right so um always uh the other bit that we kind of contend is that local optimization uh will always mess this is a little bit of an aside but it becomes important later on some of the choices we're making right um is they always want to be a locally optimized right so you can kind of see this now with chat gpt i'll let scott contradict me because he knows much more about it than i do but where you're starting to get this uh family
I like your chat GPT now you get these families of GPT you can use that are optimized for image generation or code tutoring or things of that nature other from the no free lunch theorem the ideas like weather. Always like i'm gonna have a local problem that i need to solve and i could get like the ninety percent solution that's out there right but i'm gonna have to tune that at the end to make it as optimized as possible for my particular problem set.
I contend this also is very fundamental to complex systems. Darwin's finches is a classic example from biology, where you have the proto-finch, and then they customized its beak based off which island they lived on, so it could maximize its chance for survival. Then it goes all the way down to the Lorenza tractor.
which is where the butterfly effect came from, which is that details will always matter. There will always be a measurement error. There will always be some other rounding variable out there that you didn't account for. So you want to always be able to locally tune any type of model that you have. And so from this, the thought is, from my view, from democracy, is that a great idea can come from anywhere.
Right. And so what really is you go back to, you know, like kind of feudal times or whatnot is that's where they said, hey, free speech matters because just because you're a royal writer, you're a blue blood or you live in a castle doesn't mean that that person, you know, that's out there living in squalor in the hut doesn't have some great idea that can make us all better. Right. And so if we empower people to share their great ideas, right, then we as a society will get better. Right. Now, the challenge is we've got to prevent the bad.
Right now from this, this is kind of all a Jeffrey West out of the Santa Fe Institute, right? This will, you know, the concern I suppose is that technology is accelerating faster and faster, right? And then if we get to a perfectly up and down, so you're divided by zero, then we'll have a singularity, right? Or it could just be sigmoidal where a bounce off or it will kind of flatten out and we'll reach some kind of upper limit. All right. But regardless,
This episode is brought to you by State Farm. Listening to this podcast? Smart move. Being financially savvy? Smart move. Another smart move? Having State Farm help you create a competitive price when you choose to bundle home and auto. Bundling. Just another way to save with a personal price plan. Like a good neighbor, State Farm is there. Prices are based on rating plans that vary by state. Coverage options are selected by the customer. Availability, amount of discounts and savings, and eligibility vary by state.
Embracing this idea will create exponential growth for good and bad. The question is, how does humanity mitigate the bad? We don't want to replicate some of the horrors that we've seen over the last couple hundred years. I don't know. It's a very hard problem because how do you anticipate something you can't necessarily predict from emerging effects?
Never give up. So this is where I start from, like complex adaptive systems. So first we acknowledge our limitations, which is like, well, we don't know. Fundamentally, and I look at this, so I say, we don't know. I'm saying from a policy perspective, we have no good way to say, hey, we need to stop, let's say the violence that's happening in Haiti right now, or the kind of turmoil that's happening in Central America with like a massive refugee crisis.
We don't know what policies we can put in a place that will require minimal resources for maximum benefit to create a stable society that allows people's life to flourish.
Life, liberty, and the pursuit of happiness so much. We don't really have tools to deal with it. The thought is, at this point, it's all conjecture on my point. This is what we're trying to get to. I would say take everything with several massive grains of salt, but what that thing down there on the right is supposed to be is a bunch of different strange attractors orbiting around a larger strange attractor.
What we do know is that you get into a kind of stable state, and this can be scale-free, so it could be your business, it could be your neighborhood, it could be your entire country, it could be the globe. And so the globe kind of makes it the most reasonable, where we'll talk about climate change.
right is that the the earth is in this kind of like stable homeostasis where it's a dynamical system that is constantly changing but it's orbiting around these kind of like strange attractors right that we we can't really envision all right and then as we've uh in this case you know put pollutants into the world water and created um uh created you know global warming and climate change right we've changed what uh the technical terms your fitness landscape right and then we're going to go to another uh type of uh of
Right. And so we've changed it and that could be different. Right. And you saw this in the Great Depression, right, where they had this massive economic collapse. Right. And a lot of people say, well, it will right itself. Right. You know, kind of liaison fair. The Austrian School of Economics, it'll right itself. Don't touch it. Right. And it didn't because it was stuck in one of these like low points here. Right. And it couldn't get back out. And so then you had like the New Deal and stuff like that, that changed the fitness landscape.
Right. And then we came to another stable state and kind of got pulled out of the great depression. Right. Um, and so that's the idea of how we think it's working. Right. The question is, well, how do we understand that? Okay. Um, and so this is by far not like a, uh, an exhaustive list. Uh, there's a great, um, article by David Cracker and several other people from the Santa Fe Institute that talk about some of the computational tools. Uh, but one of the things I've been looking at, right, is a work from Martin Scheffer, uh, out of, out of Germany.
I'm sorry give the bumper sticker for complex adaptive system is just an adaptive network right as you try to identify those signals that are going between that network.
This seems to work at the brain level, at the ecosystem level, and at the global level. Although we haven't seen the research on this yet, I will contend this is intuitively how I understand societies. We want to understand if we're having an impact on a society, it will follow a similar pattern. Then you can see as it changes, this work can identify when a system is about to tip,
Right. But at this point, I haven't seen anything that could say what it's going to tip into. So if you think of the bifurcation diagram, right, is that it's like, hey, we know it's about to go one of two directions. Right. But we can't tell you which one it's going to go to. Right. But you can. So I would say you could see that we look at even the U.S. society right now. Right. As a democracy is in crisis. Right. Is like, OK, what like how's this going to tip into the future?
Signals have changed, the butterfly has kind of flapped its wings and so will we tip into another strange attractor? How will our society evolve? So that kind of shows on this picture here, we have a resilient status quo and then you get to a fragile status quo and you would theoretically tip into another type of status quo. The reality is we don't understand it.
We're getting better, but this would be from an understanding perspective. How do we institute policies that create stable, thriving societies? Through your kind of meta-consciousness, we don't get hands down, but we're trying.
All right. And so the other bit too is that, you know, so now it's okay, what tools can we do? And again, I've no idea. I'm just saying this is the little path that I'm on to try and help figure this out. Right. But the one thing I do know is that better understanding is not effective policy, right? Because you have these emergent systems, like it's like, well, I know that this is having this impact. It's like, well, that's great, but what can replace it? All right. Like what new policy can I put in place to replace what's happening in order to produce this new emergence behavior of the system?
All right, so we need to develop new tools. This is primarily what my PhD was in computational social science, right? It's saying like, hey, we got to use these artifacts. We got to generate essentially virtual laboratories of the world so we can test things out in silico, as they say, prior to actually implementing it.
Right. And so you have generative science, which is fundamentally different from reductionist science, right? Where it's like, uh, I'm not trying to, uh, grow or I'm trying to grow the system, right? Cause I can't just disprove things. Right. Um, and then, uh, so create simulations, you know, essentially these virtual laboratories, we can try stuff out. And so that is happening. Uh, so, uh, George Mason, Oxford and
Princeton built a simulation back in 2014 to try and replicate the 2008 housing market crisis. In this kind of science, because you can't just prove stuff, you do what's called pattern-oriented modeling. They had nine statistics from that time period on the
The Central Bank of England uses it for London and the Central Bank of Norway has one for Oslo where essentially they try out policies in Silico and then they try to implement those policies
What am I doing personally for this? First off, you're trying to increase tech literacy.
There's a lot of myths and stuff there about AI and what it can do and fear and too much people watching cable news. Right. And so it's, you know, the luckily our faculty, we have a PhD in curriculum and instruction who's very good. And so we're trying to feel, well, how do we, instead of like giving people the classic data science thing, it was like, okay, we're going to start you out coding. You're going to learn, you know, all these statistics and these data science tools be like, because we deal with them.
Hey you really know a lot about let's say you know political science or
Then we'll give you an analytic that meets you in your space where you're an expert. Then as you can realize that, hey, this analytic is not exactly right, you could just flip it to the code. So if you're familiar with Jupyter Notebooks and stuff like that and not see the code right away, but see the choices that that developer made and then start to explore that computational tool at deeper and deeper levels based off where you feel most comfortable.
Right, and that kind of takes best practices of adult learning. It's a put it up there, but one of my thesis students built cell toolbar, which is work. It's not exactly where it should be yet, so you might want to give it a couple of months, right? But it's the idea is that you can hide code better and store knowledge so it's easier for people that aren't technically literate to interact with the code, right? And that's on the pie PI library says extension for Jupiter notebooks, right? And then and then I still continue work on Mesa, agent based modeling and Python.
Right. Which is like, how do we get, how do we develop a simulation ecosystem? So this is very much my own heavily biased opinion. But if you look at like, if you're familiar, like scikit-learn, which is great, you can, you know, build all sorts of machine learning models and like three lines of code more or less, right. And you've got like TensorFlow and PyTorch for TensorFlow is from Google, PyTorch is from Facebook. And so you can very easily build out these deep learning models and things of that nature. Right.
Um, but what does that look like for the age of base modeling ecosystem? So again, bias is, which is I would say is order of magnitude harder to have an ecosystem where I can rapidly bring together, uh, the pieces for and build a model, you know, quickly that's relevant, right? So, uh, the example I use is like an elementary school principal during like a COVID outbreak, right? Like how can that person try out, uh, you know, um, policies
that are valid for their particular school based off their situation, right? Like how can you say, okay, I got the mindset of a kindergartner and I got the mindset of a sixth grader, right? And now this is exactly in how my school's laid out, whether it's in remote Kansas or, you know, New York City and what policies can I put in place, right, to try this virtual laboratory that will dramatically mitigate the outbreak in my institution, right? It might be as simple as make sure a teacher watches the kindergartners wash their hands.
How do we create an environment we can have simulations that you rapidly put together?
All right, so questions.
So you mentioned the housing crisis model and you said that they used eight of nine indicators and that the Bank of England is now using this model, but given the butterfly effect, you're skipping an indicator. How do you resolve that kind of quandary? I think you ever resolve it, right? Because it's always going to be in there.
Right, so you can you can you got like an event horizon if you will we could kind of see out this far but it's always kind of you're always having to constantly tweak the knobs and update your understanding of the situation right so you can you can never solve it it's i would say unequivocally an impossible problem to solve i think the technical term from chaos theory is it's deterministically unpredictable right which means the past the past creates the future but because of all these rounding errors and variables it's you can't actually predict it
Thank you so much. It's very enlightening and I don't really know how to ask this but there's studies about the morphic field because the thing is like the United States when he started the country they had really good people worrying about the people for the people and everything and they also had a saying of in God we trust because they believe in the supernatural
So I feel like a lot of the science sometimes they're forgetting about this supernatural thing that has been being studied from, you know, ancient Egypt or mysticism occult sciences. So the morphogenetic field is the thing that is not indivisible and is still like is right now is strong like for therapy and is actually being implemented in Brazil in the judiciary and the health system.
because it has shown effectiveness because this field actually is sky. We have the Akashi field that is our cloud in a base and the morphogenetic field is our programming. So the way that things are programmed, we can go back and reprogram, but it's more on the
Occult scientists, I guess, I mean the supernatural. Nikola Tesla said, like, when we start studying the non-physical phenomena, we will have more advances in science in a decade than in centuries. And I do believe we need to go back to the sacred, like, ancient scriptures. And not only that, like, the, like, the secret
I'm sorry, I'm nervous. I'm not really good with talking, but are you familiar with Manly P Hall? With who? Manly P Hall. No, he is the writer of the secret teachers of all ages, right? I'm not, uh, I can't remember the book now, but there's a lot of wisdom in that. And right now what we need in consciousness is also the technology, understanding wisdom, because like quantum computing is like you getting information from all sources in order to build this intelligence.
We have all this information. We're only missing this part that is like doing the things that is right. Because even everything that the government has done, that you're saying that now we need to be more clear about it, is still there. It still shows on a morphic field. Whenever we do this right through, you know, the supernatural, like using love, compassion,
We can rewrite what was done wrong and I think we can have a better society and better leaders because that's what we're needing right now. We need someone that is going to guide us without interest of you know like businesses or things like that. People that really care about people and we have that we care about each other but in a way we've been programmed to think differently. Hear that sound?
That's the sweet sound of success with Shopify. Shopify is the all-encompassing commerce platform that's with you from the first flicker of an idea to the moment you realize you're running a global enterprise. Whether it's handcrafted jewelry or high-tech gadgets, Shopify supports you at every point of sale, both online and in person. They streamline the process with the Internet's best converting checkout, making it 36% more effective than other leading platforms.
There's also something called Shopify Magic, your AI-powered assistant that's like an all-star team member working tirelessly behind the scenes. What I find fascinating about Shopify is how it scales with your ambition. No matter how big you want to grow, Shopify gives you everything you need to take control and take your business to the next level. Join the ranks of businesses in 175 countries that have made Shopify the backbone.
of their commerce. Shopify, by the way, powers 10% of all e-commerce in the United States, including huge names like Allbirds, Rothy's, and Brooklyn. If you ever need help, their award-winning support is like having a mentor that's just a click away. Now, are you ready to start your own success story? Sign up for a $1 per month trial period at shopify.com slash theories, all lowercase.
Go to Shopify.com slash theories now to grow your business no matter what stage you're in Shopify.com slash theories.
I'm going to take what you just said and I'm not going to disagree. I'm going to put it in these terms. What I would say is you're arguing for a different perspective that's superior. They say, hey, I think this is superior. If we implement these types of policies, we as a humanity, society of different levels will do better.
Right. And so I'd say is like using the kind of idea of a explore, explore, exploit algorithm is like you can say that. And then as people implement that, the trick is, are we optimizing our environment, right? To allow the ideas that are working, right? To actually bubble to the surface. Because the trick is there's a lot, but a lot of bad ideas that have bubbled to the surface that didn't work, right? That, uh,
that people encourage, you know, for like, like Nazis and fascism and stuff like that, right? So that I would say, I don't disagree there. It's just like, okay, so you say, hey, these ideas, we're seeing these in Brazil, they're working really good, right? How do we optimize the policies, you know, technological environments of our society so that those ideas can be validated?
Right. And bubble up to the surface. Right. And I mean, just politically, as you can all see in our very kind of caustic environment in the U.S. right now, right, is that people will say something like, hey, this is right. Right. Like, hey, you know, it's been shown as like a controversial idea that you could take somebody that's like, say, homeless and alcoholic. And it's better just to put them in like a home where they have access to alcohol because then you're actually spending less money on ER visits and stuff like that. But that's a very controversial idea.
Right. And so it's, you know, I don't know the answer, but that's, I would say the bit is how do you take the ideas that even might not be intuitive, right. And say like, Hey, if we do this, it actually works really well. Right. And you know, the trick is we start going against culture, right. Then people like, no, I'm not going to, I will literally not believe that no matter what happens. Right. Like the earth is flat. Uh, yeah. Thank you for the talk. Uh, Tom, uh, I enjoyed it. I want to press you a little bit on this notion of stability.
I guess in this sense, right? So part of what the project is to maintain stability as a way to maybe effectively govern or create a more just society. I guess the way stability is in this context, it makes it sound like it's just stability is always good, right? But in some ways, I wonder if like, actually, that should be our kind of maximizing principle, right? Because you might think, well, say something unjust is going on, the complete opposite of stability seems to be what actually would be the normative, like how you ought not to
So I wonder, like, you know, say we put certain practices in the place, right, which always maximize stability, but you have unethical or unjust society or something like that. Well, then it seems like you're just helping to maintain that particular thing. So I'm just wondering in some ways, like, how do you, because the question I get from the talk, right, is this is to have a positive impact, right? In some ways.
And I'm wondering, well, actually, it seems like maybe stability could have a very negative impact in some ways. So I'm wondering, like, how does that get counterbalanced in the work? So I mean, that's an awesome question, Gary. That's phenomenal. I'd say it's you're absolutely right. Like, it can never stop moving. Right. I think the trick in economics is if you're going to have change, you just want it to not be completely abrupt.
Right, so much like a democracy, in my opinion, after spending my adult life fighting counter-insurgency and stuff, is that democracy is nobody wins. Once somebody wins, they just try and steal all the power. So I figure that stability is more like,
It's not like some kind of utopian stability that can't exist. It's more like how do we, as we're constantly evolving and maybe going through these spots of punctuated equilibrium, do we make sure you don't have like a massive famine or like this have a couple of times in China where literally millions of people are dying, right? Like how do we essentially shift from one kind of strange attractor, go through a punctuated equilibrium to a new kind of homeostasis stable state, right? How do we
help manage that so it's not as traumatic. Great talk, Tom. So there's been a resurgence around agent-based thinking recently with the LLM models. What are your thoughts about taking this agent-based modeling, which has been around for some time with all this great work in complex systems, how do we tie in the LLMs? Do you envision putting an agent on to each little agent, rather put an LLM inside each little
Each little spot. I think so. And actually Mark, Mark and I've kind of talked about this for a while. Um, but we, I mean, it's pretty straight. If I had a student do this, uh, last quarter directed study is that you can link up LLMs to talk to each other as essentially like little agents. Right. Uh, I think in the end, like it's, you know, you start talking abstractions, uh, you know, if complex systems are an adaptive network, what's effectively on LLM, it's an adaptive network. Right. So I think it's some of the problems that open AI particular is trying to solve when you get to AGI.
is in my own mind and this is like very much conjecture right is that at some point the kind of tools will probably merge together right and then just how they're presented by the computer scientists is just different abstractions so they're more understandable but i think it's both adaptive networks and now it's like well how these adaptive network functions to solve problems because you can imagine like the housing simulator but each one actually wants a house or it thinks it does yeah yeah so there you go
So you have a, you had a chart there where we're going towards a singularity. One of the interesting things about getting close to a singularity is the laws start breaking down, whether you're leaving from a singularity or going towards a singularity. And I would suggest that a lot of the assumptions are beginning to break down.
For example, the assumption that policy matters because used to there was a policy people would follow the policy. Now we have so much not following the policy. There's ways around the system. Um, there are other effects like facts don't matter as much. And so a lot of the assumptions we have, those are beginning to break down as we're getting closer to the singularity. So we almost need a meta.
Consideration about what we would do. Do you have any thoughts about that? I I I would say so I'm not sure we're getting to the singularity I'll go to this slide Right is I think we're at that back that back hill up there where it's a more fragile status quo as the fitness Landscape has changed like our chances of tipping into a new dynamic are just increasing right? We're seeing that all the time with numerous unprecedented historical events, right? so I think that's it and I think I
My only other thought would be that it is like kind of fractal in nature, right? And kind of like yesterday I started talking the universe and the universe is the universe is all the way up.
Can you use tools like this to essentially view your world like you're in the universe up? To the point of things breaking down, I think we're definitely seeing that doing some stuff on the open source ecosystem. Like, hey, this defies all traditional economics. Why are people volunteering their time to do these massively valuable inputs when they should be getting paid for? What does that mean for an economy?
All right, so I think you're absolutely right. I'm not sure we're going to the singularity though. We are going faster, but I think we are definitely in a spot of more, like a more fragile state where we might tip into some new type of reality, right? Not like, like just another historical reality, right? Like, you know, 18th century and 19th century, not like a, like a, I guess more like tipping into a singularity. So I wanted to ask you a question that's related to those last two questions.
You started your talk with the basically diversity of thought being a good thing and leading to expansion of knowledge and cultural revolution. I can't help but think to recent history in the 1850s, it seems like there was a lot of diversity of thought in America. And I would argue that that could be perceived as a bad thing because it led to the Civil War, 600,000 Americans dying. And so that's like a saddle point basically. And the way out of that saddle point was
You know, arguably Abraham Lincoln was implementing what would be considered very fascistic policies like imprisoning dissenters, political dissenters. So two part question is, um, you know, how do you distinguish good diversity of thought from bad diversity of thought that's destabilizing and, um,
You know, to what extent or what value can you or how can you implement a network that is going to be able to feed forward knowledge from, you know, centuries of lessons learned that can actually impact the future evolution? I can't help but think that any future civil war that might be avoided by a recollection of the not too distant past, the horrors. That's it. So the short answer is I have no idea because you're absolutely right, right? Like the right
because we just we don't understand it right like we can't uh we don't if you i mean you go back to the like the civil war or any world war two there was there was moments that that were very brief and very important you know comparatively like eddiesburg um the battle of midway and stuff like that where you could have had a vastly different reality right uh that that came out of it and and so i guess what i would say you just said like that to me epitomizes the problem
We have no scientific or rigorous way to say, well, in this situation, you should do these things and that will help push the system in this direction. Right? Like, like everything I've seen is that's like we're making small progress, but we functionally don't have tools that will allow us to do that. Right. And, and so that's what, you know, now, now spend 15 years trying to figure that out. Yeah. It's a tough one, right? Yeah. So.
It seems like in those last few slides, you're talking about human tractable and interpretable algorithms with human tractable and interpretable variables so that someone can take their local situation and sort of tailor the algorithm to them, like your example of the school with COVID. In situations like that where the person's answer from the algorithm actually matters in them making real local policy decisions,
How do you reconcile this idea of building human tractable and interpretable algorithms with the fact that it seems like the non tractable and non interpretable algorithms where you just throw in 2000 data points and ask it to predict something seem to have much more predictive power and would be much better for local policy decisions? Well, so, so this is what I'd say and this is what I've given this conference because that way people can call BS and I know there's a lot of folks very proficient at LLMs and stuff like that. All right. Um, is that,
The challenges with like these current tools, right, is I can break those tools if the world changes, right? So my understanding was when COVID hit and people's behavior changed drastically overnight and did something that never seen before, right, that that broke a bunch of like the tech giants tools, right? Because the trick is we're not trying to predict the future. We're trying to shape the future.
I like surface choice between genocide war world war three and nuclear weapons we don't want that to happen how do we implement things that help us that put us on a different path so i add so far and i'm not like me deep into so the large language model stuff
is I haven't seen, from my perspective, I haven't seen models that can account for that, right? That can say, hey, here's a policy you should do to create this new future. In fact, the only thing that I've seen that even gets close, and this kind of echoes Harmut's talk yesterday, was the BOM experiment from quantum mechanics, where it's like, hey, it's actually not small, it's all the way up. In the BOM experiment, they're able to show that quantum could show what didn't happen.
I just wanted to go back to in the beginning you talked
A little bit about meta consciousness. Could you give us a definition of what do you mean by meta consciousness and what does that entail as far as looking at these complex systems?
So kinda classic when i've been exposed to on this is jane jacobs like the wisdom of cities right like it's amazing you can look at cities you like tammany hall in the eighteen hundreds in the u.s. and lagos nigeria now and it's like they they form a kind of collective consciousness on how they function writer even london during like dickinsonian times right so you have these you know see have these.
I know the Santa Fe Institute is really looking into that and they started a new journal, I think it's Jessica Flack and Scott Page, called Collective Intelligence. I think the book is Solaris. There's a book called Solaris, then there's a Russian movie, then there's a George Clooney movie that sucks.
But in that, it's the idea that this one planet is a singular consciousness, and it's all just functioning together. I will give you a definition, I'll give you the field of study, and I would look at the Collective Intelligence Journal that's only a year old or so now. But just to clarify, it's
collective intelligence, not like sensational consciousness of a society. Yes. Yeah. Yeah. Right. But that's, well, I mean, cause that goes to the other, but we could go down, uh, downward, right? Where are, are we as a conscious human being, just the byproduct of a microbiome that needed to survive.
Extra value meals are back. That means 10 tender juicy McNuggets and medium fries and a drink are just $8.
I just wanted to make a quick comment on the collective intelligence, consciousness kind of thing. I always like to think of our vegetables, fruits, certain plants that we like to use to fuel in different altered states are kind of using us to propagate their species in a sense. Like if you think about a lot of the animals that we eat or the plants that we eat,
What better way of ensuring their survival than being tasty or nutritious to us and now they're just everywhere. So it is kind of a very interesting dynamic to think about. That's a good defer to an evolutionary biologist. Hi, so I lived in New York for 30 years where the conversation was a lot about learning organizations and just-in-time knowledge that sort of thing. So now I live in a rural area in Florida where the biggest problem locally is whether or not people are allowed to
park big trucks in their backyard and they resent any kind of opinions coming from anywhere else about, you know, the aesthetics of it, the utility of it, all of that sort of thing. So how do you get past like the local culture that totally rejects any kind of input from a larger outside organization? Well, so I don't think you do. This kind of goes, I think, to your question a bit, Gare, which is
I don't know if you know who David Kilcullen is. He's an Australian, one of the world's foremost counter-insurgents. He was advising the US at the most senior levels through Iraq and Afghanistan. He wrote a book called Out of the Mountains, where he proposed the theory of competitive control. Out of the mountains? In that, the idea is, we're in these homeostasis stable states, but you're never going to have
homogeneity in a population. I wouldn't necessarily say that diversity of opinions isn't necessarily bad, but they do select over time. Because rural Florida is vastly different than urban New York City, you're going to select different things based off whatever your optimal dynamic is. When you start to get to problems, it's like, well, now your optimal dynamic is actually, say,
the fertilizers in Iowa, you know, creating red tides and killing the fish down in Texas. So how'd you balance that? Right? And that's, so it's, I guess I just say, you always want heterogeneity in any type of, you know, population. And then it's, what are the, how'd you set up dynamics to optimize that so you can, you know, let a lot of people just go about their life and not be interrupted, but also optimize how that society continues to function and doesn't, you know, get a war with itself.
Firstly, thank you for watching. Thank you for listening. There's now a website, kurtjymungle.org, and that has a mailing list. The reason being that large platforms like YouTube, like Patreon, they can disable you for whatever reason, whenever they like.
That's just part of the terms of service. Now, a direct mailing list ensures that I have an untrammeled communication with you. Plus, soon I'll be releasing a one-page PDF of my top 10 toes. It's not as Quentin Tarantino as it sounds like. Secondly, if you haven't subscribed or clicked that like button, now is the time to do so. Why? Because each subscribe, each like helps YouTube push this content to more people like yourself
Plus, it helps out Kurt directly, aka me. I also found out last year that external links count plenty toward the algorithm, which means that whenever you share on Twitter, say on Facebook or even on Reddit, etc., it shows YouTube, hey, people are talking about this content outside of YouTube, which in turn
Greatly aids the distribution on YouTube. Thirdly, there's a remarkably active Discord and subreddit for theories of everything where people explicate toes. They disagree respectfully about theories and build as a community our own toe. Links to both are in the description. Fourthly, you should know this podcast is on iTunes. It's on Spotify. It's on all of the audio platforms. All you have to do is type in theories of everything and you'll find it. Personally, I gained from rewatching lectures and podcasts.
I also read in the comments
and donating with whatever you like there's also paypal there's also crypto there's also just joining on youtube again keep in mind it's support from the sponsors and you that allow me to work on toe full time you also get early access to ad free episodes whether it's audio or video it's audio in the case of patreon video in the case of youtube for instance this episode that you're listening to right now was released a few days earlier
Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon
. . . . . . . . .
▶ View Full JSON Data (Word-Level Timestamps)
{
"source": "transcribe.metaboat.io",
"workspace_id": "AXs1igz",
"job_seq": 5566,
"audio_duration_seconds": 3163.93,
"completed_at": "2025-11-30T23:46:30Z",
"segments": [
{
"end_time": 20.896,
"index": 0,
"start_time": 0.009,
"text": " The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze."
},
{
"end_time": 36.067,
"index": 1,
"start_time": 20.896,
"text": " Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates."
},
{
"end_time": 64.514,
"index": 2,
"start_time": 36.34,
"text": " Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a toe listener, you get a special discount. Head over to economist.com slash TOE to subscribe. That's economist.com slash TOE for your discount."
},
{
"end_time": 93.439,
"index": 3,
"start_time": 66.852,
"text": " Hola, Miami! When's the last time you've been to Burlington? We've updated, organized, and added fresh fashion. See for yourself Friday, November 14th to Sunday, November 16th at our Big Deal event. You can enter for a chance to win free wawa gas for a year, plus more surprises in your Burlington. Miami, that means so many ways and days to save. Burlington. Deals. Brands. Wow! No purchase necessary. Visit BigDealEvent.com for more details."
},
{
"end_time": 113.183,
"index": 4,
"start_time": 94.326,
"text": " As a democracy in crisis, how's this going to tip into the future? So your signals have changed, the butterfly has kind of flapped its wings. Will we tip into another strange attractor? Because we could take a student and, you know, a 10-week class, take it from almost no coding, you know, to building a facial recognition with a support vector machine. Like, that's amazing."
},
{
"end_time": 145.145,
"index": 5,
"start_time": 115.162,
"text": " Thomas Pike is a faculty at National Intelligence University. This talk was given at MindFest, put on by the Center for the Future Mind, which is spearheaded by Professor of Philosophy Susan Schneider. It's a conference that's annually held where they merge artificial intelligence and consciousness studies and held at Florida Atlantic University. The links to all of these will be in the description. There's also a playlist here for MindFest. Again, that's that conference, Merging AI and Consciousness. There are previous talks from people like Scott Aronson, David Chalmers,"
},
{
"end_time": 160.964,
"index": 6,
"start_time": 145.145,
"text": " Stuart Hameroff, Sarah Walker, Stephen Wolfram, and Ben Gortzel."
},
{
"end_time": 181.681,
"index": 7,
"start_time": 160.964,
"text": " From that perspective, an analytical one, but as well as a philosophical one, discerning, well, what's consciousness's relationship to fundamental reality? What is reality? Are the laws as they exist even the laws and should they be mathematical? But instead, I was invited down to film these talks and bring them to you courtesy of the Center for the Future Mind. Enjoy this talk from MindFest."
},
{
"end_time": 209.548,
"index": 8,
"start_time": 183.183,
"text": " This morning, it's my distinct pleasure to introduce Dr. Tom Pike. He is the Dean of the Ottinger School of Science and Technology Intelligence at the National Intelligence University. Today, he's got a talk planned for talking about how complex systems can alter the course of evolution of artificial intelligence. Really looking forward to this talk, Tom. So when you're ready, please take it away. Thank you very much, Eric. All right. So hello, Tom Pike. So first,"
},
{
"end_time": 228.695,
"index": 9,
"start_time": 210.077,
"text": " Take care of some mandatory stuff, so my views do not reflect those of the U.S. Army, National Intelligence University, or U.S. government. And then the second bit to kind of deal with the elephant in the room is intelligence is often associated with bluntly jackbooted thugs like the Gestapo, KGB, things of that nature."
},
{
"end_time": 258.148,
"index": 10,
"start_time": 228.695,
"text": " Right. And so when people hear intelligence, they usually think, you know, like, like in the US where they're putting LSD in the water in San Francisco, and we have the pike and church committees. Right. And so although there will always be a need for secrets, and that's always going to be paramount to even your relationship with your neighbors, right, it's important to understand there's a whole nother aspect, which is just trying to understand how the world works. Right. And if you look at Afghanistan, in particular, as the most recent example,"
},
{
"end_time": 287.944,
"index": 11,
"start_time": 258.78,
"text": " That wasn't a failure because we didn't have enough secrets on how the Taliban were operating. That was because we don't understand really how to grow a democracy. We don't understand how to get those critical dynamics into place in order to have a functioning, productive society that respects humans' rights. So there's intelligence, the secrets part, really my focus or all the work I do with Dr. Bailey is on how do we just understand the world? And that's all in the open."
},
{
"end_time": 307.995,
"index": 12,
"start_time": 288.592,
"text": " Right. And so we do it in the open. I do open source development. I do things of that nature. So we say intelligence in this part really say it's more like we're just trying to understand how the world works. Right. And this is a look at Director Haynes. She did an op-ed in the Wall Street Journal about two years ago where she was saying like, hey, you know, we as a democracy, we need less secrets."
},
{
"end_time": 327.005,
"index": 13,
"start_time": 307.995,
"text": " Hi, all right."
},
{
"end_time": 356.51,
"index": 14,
"start_time": 327.312,
"text": " So, I'm going to use this quote from Scott Page, the Santa Fe Institute, the University of Michigan, right? Which is like, the right perspective makes hard problems easy, while the wrong perspective makes easy problems hard, right? So, what I'm going to do is like yesterday, we had Origin of Life from Sarah Walker, we had lots of talks about consciousness and quantum mechanics, right? So, what I'm going to look at is more like societies, it's kind of like a meta-consciousness, right? And so, as you look at this, as we've gone through life,"
},
{
"end_time": 386.015,
"index": 15,
"start_time": 357.039,
"text": " All right, as humanities, you know, kind of evolved over the last couple thousand years, right, is that we found superior perspectives that then makes our life better, right? So look at the Copernican view of the universe, you know, theory of, you know, theory of bacteria, gravity, all those things, you know, Newton and whatnot made our lives better as we found those perspectives, right? And we were able to share those perspectives. Now humanity could say, oh, this actually really works. I can take that work and continue to build on it."
},
{
"end_time": 402.142,
"index": 16,
"start_time": 386.288,
"text": " I'm doing a specific story on it, but this comes from a book by Hilton Root out of George Mason University called Network Origins of the Global Economy."
},
{
"end_time": 425.077,
"index": 17,
"start_time": 402.568,
"text": " What's interesting about this is the Chinese developed the printing press 400 years before Gutenberg did in Europe. Not only did they invent it 400 years before Gutenberg, but they stopped using it 80 years before Gutenberg. They said, this knowledge replication technology, if you will, nothing to see here."
},
{
"end_time": 449.77,
"index": 18,
"start_time": 425.401,
"text": " We don't need it. Then you go to Gutenberg in Europe, which is this massive small world network of competing fiefdoms that are forming alliances and getting in wars with each other and stuff like that. It becomes this massive catalyst for the Reformation, the scientific enlightenment, and really Western Europe just blowing up as kind of like a global powerhouse."
},
{
"end_time": 458.558,
"index": 19,
"start_time": 450.247,
"text": " All right. And so in Hilton Root's book, right, the way the conclusion they make is like was how that network was functioning. It's the differences between those."
},
{
"end_time": 487.142,
"index": 20,
"start_time": 458.643,
"text": " Right. So you look at the Chinese network under the Chinese emperor at the time. Right. It was a hub and spoke network. And so the bureaucratic elite were incentivized, you know, to maintain their status. And if somebody didn't want that technology to get out, they could cut it off. Right. Well, in Europe, because it was a small world network, people like Galileo Copernicus could sneak their their writings that, you know, the earth is actually not the center of the universe. Like only paths, the Catholic Church into like places like the Netherlands."
},
{
"end_time": 514.411,
"index": 21,
"start_time": 487.517,
"text": " Right. All that sun, that knowledge was getting out there and spreading. Right. And so it became this, you know, and it just blew up. Right. For good and bad. Right. And so if we look at this, you know, much like a large language model or a deep learning network. Right. Is I look at humanity's collective consciousness is like a fractal problem solving network. You have all these folks all over the world, particularly like the open source community. Right. That this is this global problem solving network."
},
{
"end_time": 534.138,
"index": 22,
"start_time": 514.77,
"text": " And it's in a very real way functioning just like a neural network. So they're just searching this hyperdimensional landscape of possibilities, as Sarah Walker talked about yesterday, and finding new perspectives that help us understand the world and engage with it better."
},
{
"end_time": 553.882,
"index": 23,
"start_time": 534.684,
"text": " And so our knowledge storing and sharing technology are artifacts. Herbert Simon, if you're familiar with the sciences of the artificial, are ways to help us optimize our Explore, Exploit search mechanism. Just because it's complexity,"
},
{
"end_time": 576.613,
"index": 24,
"start_time": 553.882,
"text": " Big fan of complexity theory, like Sarah Walker said yesterday, right? She's effectively came up with assembly theory, right? We know from science like rigor really matters, right? And how do we test this and make it, you know, how do we have rigor to this new theory? But she's effectively proposing a new way to look at the world that defies our traditional physics types measurements."
},
{
"end_time": 588.063,
"index": 25,
"start_time": 576.852,
"text": " Alright and so you know potentially that will blow up and give us a whole new ways to interpret and understand the world right give us a new superior perspective solve these problems are finding really hard right now will become easy."
},
{
"end_time": 613.695,
"index": 26,
"start_time": 589.497,
"text": " After we look at the printing presses, things have got a lot better in numerous areas. Increased lifespan, child mortality has plummeted, improved health care, for many improved quality of life, increased education, probably the most literate society that we've ever had, but it's also come with some absolutely horrible things."
},
{
"end_time": 641.8,
"index": 27,
"start_time": 614.002,
"text": " You have slavery and child labor, indigenous people's genocide, pollution. Right now, we're looking at the current rise of autocrats, based off the dynamics of real-time, in my opinion, the dynamics of real-time bidding and the need to optimize wealth functions. It also comes with a lot of bad. Those are the things we want to try and prevent if we're going through another great revolution."
},
{
"end_time": 670.043,
"index": 28,
"start_time": 642.398,
"text": " The idea from this is that we really have a new type of printing press, if you will, right? Knowledge storing and sharing technology, right? So you have that if you're not familiar with this, right? If you don't code, just give you a quick rundown, right? So you have the internet, right? Where you can share knowledge globally, right? So I work on a shameless plug, but I work on Mesa, agent-based modeling in Python, right? And we got people from all over the world, particularly right now, the Technical University of Delft that are contributing to it and making it vastly better. Like I can't even keep up."
},
{
"end_time": 693.933,
"index": 29,
"start_time": 670.367,
"text": " Right. And so we're able to store that in GitHub on the left. There's also GitLab, Bitbucket. There's all these like knowledge coding repositories that make it much easier to say like, well, this was your idea. I just made this change to it. Now it's much better. Right. And you can instantly merge it and then push it out to the entire world. Right. And then you get these like naturally selected libraries where somebody has a really good library."
},
{
"end_time": 723.439,
"index": 30,
"start_time": 694.462,
"text": " or somebody has a really good idea, then somebody else can go in there and use it and take that knowledge. So I think about like a year and a half ago before I got stuck in academic administration, we built a crop simulator, look at famines in West Africa, and I know Jack about crop growth, but I was able to find a reputable library for a person in England, and we could reliably show, based off the current weather patterns in West Africa,"
},
{
"end_time": 753.541,
"index": 31,
"start_time": 723.695,
"text": " This is the dangers of a famine based off these kind of staple crops in Niger. So you're able to find those. You got things like Scikit-learn, TensorFlow, PyTorch. Now we're getting like hugging face and they'll have different large language model attributes. So we can now take that knowledge. So I don't have to spend years doing it. I can rapidly apply it, which even comes with its own problems. If you don't understand what you're doing, you start applying tools that can give you answers that are appealing, but somewhat wrong."
},
{
"end_time": 782.773,
"index": 32,
"start_time": 754.104,
"text": " So now the real point is we're in this next great revolution of knowledge storing and sharing. I personally contend this is a real revolution that's driving AI. It's not that if it was really stuck at these boutique firms or universities, it would not have blown up like it has. But because we could take a student and a 10-week class, take it from almost no coding to building a facial recognition with a support vector machine,"
},
{
"end_time": 807.568,
"index": 33,
"start_time": 783.029,
"text": " like that's that's amazing right so um always uh the other bit that we kind of contend is that local optimization uh will always mess this is a little bit of an aside but it becomes important later on some of the choices we're making right um is they always want to be a locally optimized right so you can kind of see this now with chat gpt i'll let scott contradict me because he knows much more about it than i do but where you're starting to get this uh family"
},
{
"end_time": 832.039,
"index": 34,
"start_time": 807.568,
"text": " I like your chat GPT now you get these families of GPT you can use that are optimized for image generation or code tutoring or things of that nature other from the no free lunch theorem the ideas like weather. Always like i'm gonna have a local problem that i need to solve and i could get like the ninety percent solution that's out there right but i'm gonna have to tune that at the end to make it as optimized as possible for my particular problem set."
},
{
"end_time": 851.391,
"index": 35,
"start_time": 832.295,
"text": " I contend this also is very fundamental to complex systems. Darwin's finches is a classic example from biology, where you have the proto-finch, and then they customized its beak based off which island they lived on, so it could maximize its chance for survival. Then it goes all the way down to the Lorenza tractor."
},
{
"end_time": 875.213,
"index": 36,
"start_time": 851.852,
"text": " which is where the butterfly effect came from, which is that details will always matter. There will always be a measurement error. There will always be some other rounding variable out there that you didn't account for. So you want to always be able to locally tune any type of model that you have. And so from this, the thought is, from my view, from democracy, is that a great idea can come from anywhere."
},
{
"end_time": 905.179,
"index": 37,
"start_time": 875.879,
"text": " Right. And so what really is you go back to, you know, like kind of feudal times or whatnot is that's where they said, hey, free speech matters because just because you're a royal writer, you're a blue blood or you live in a castle doesn't mean that that person, you know, that's out there living in squalor in the hut doesn't have some great idea that can make us all better. Right. And so if we empower people to share their great ideas, right, then we as a society will get better. Right. Now, the challenge is we've got to prevent the bad."
},
{
"end_time": 929.343,
"index": 38,
"start_time": 905.418,
"text": " Right now from this, this is kind of all a Jeffrey West out of the Santa Fe Institute, right? This will, you know, the concern I suppose is that technology is accelerating faster and faster, right? And then if we get to a perfectly up and down, so you're divided by zero, then we'll have a singularity, right? Or it could just be sigmoidal where a bounce off or it will kind of flatten out and we'll reach some kind of upper limit. All right. But regardless,"
},
{
"end_time": 958.865,
"index": 39,
"start_time": 930.128,
"text": " This episode is brought to you by State Farm. Listening to this podcast? Smart move. Being financially savvy? Smart move. Another smart move? Having State Farm help you create a competitive price when you choose to bundle home and auto. Bundling. Just another way to save with a personal price plan. Like a good neighbor, State Farm is there. Prices are based on rating plans that vary by state. Coverage options are selected by the customer. Availability, amount of discounts and savings, and eligibility vary by state."
},
{
"end_time": 982.927,
"index": 40,
"start_time": 960.401,
"text": " Embracing this idea will create exponential growth for good and bad. The question is, how does humanity mitigate the bad? We don't want to replicate some of the horrors that we've seen over the last couple hundred years. I don't know. It's a very hard problem because how do you anticipate something you can't necessarily predict from emerging effects?"
},
{
"end_time": 1009.514,
"index": 41,
"start_time": 983.319,
"text": " Never give up. So this is where I start from, like complex adaptive systems. So first we acknowledge our limitations, which is like, well, we don't know. Fundamentally, and I look at this, so I say, we don't know. I'm saying from a policy perspective, we have no good way to say, hey, we need to stop, let's say the violence that's happening in Haiti right now, or the kind of turmoil that's happening in Central America with like a massive refugee crisis."
},
{
"end_time": 1021.067,
"index": 42,
"start_time": 1010.282,
"text": " We don't know what policies we can put in a place that will require minimal resources for maximum benefit to create a stable society that allows people's life to flourish."
},
{
"end_time": 1042.142,
"index": 43,
"start_time": 1021.323,
"text": " Life, liberty, and the pursuit of happiness so much. We don't really have tools to deal with it. The thought is, at this point, it's all conjecture on my point. This is what we're trying to get to. I would say take everything with several massive grains of salt, but what that thing down there on the right is supposed to be is a bunch of different strange attractors orbiting around a larger strange attractor."
},
{
"end_time": 1061.288,
"index": 44,
"start_time": 1042.739,
"text": " What we do know is that you get into a kind of stable state, and this can be scale-free, so it could be your business, it could be your neighborhood, it could be your entire country, it could be the globe. And so the globe kind of makes it the most reasonable, where we'll talk about climate change."
},
{
"end_time": 1091.698,
"index": 45,
"start_time": 1062.005,
"text": " right is that the the earth is in this kind of like stable homeostasis where it's a dynamical system that is constantly changing but it's orbiting around these kind of like strange attractors right that we we can't really envision all right and then as we've uh in this case you know put pollutants into the world water and created um uh created you know global warming and climate change right we've changed what uh the technical terms your fitness landscape right and then we're going to go to another uh type of uh of"
},
{
"end_time": 1120.111,
"index": 46,
"start_time": 1091.886,
"text": " Right. And so we've changed it and that could be different. Right. And you saw this in the Great Depression, right, where they had this massive economic collapse. Right. And a lot of people say, well, it will right itself. Right. You know, kind of liaison fair. The Austrian School of Economics, it'll right itself. Don't touch it. Right. And it didn't because it was stuck in one of these like low points here. Right. And it couldn't get back out. And so then you had like the New Deal and stuff like that, that changed the fitness landscape."
},
{
"end_time": 1150.026,
"index": 47,
"start_time": 1120.435,
"text": " Right. And then we came to another stable state and kind of got pulled out of the great depression. Right. Um, and so that's the idea of how we think it's working. Right. The question is, well, how do we understand that? Okay. Um, and so this is by far not like a, uh, an exhaustive list. Uh, there's a great, um, article by David Cracker and several other people from the Santa Fe Institute that talk about some of the computational tools. Uh, but one of the things I've been looking at, right, is a work from Martin Scheffer, uh, out of, out of Germany."
},
{
"end_time": 1168.643,
"index": 48,
"start_time": 1150.503,
"text": " I'm sorry give the bumper sticker for complex adaptive system is just an adaptive network right as you try to identify those signals that are going between that network."
},
{
"end_time": 1193.66,
"index": 49,
"start_time": 1168.643,
"text": " This seems to work at the brain level, at the ecosystem level, and at the global level. Although we haven't seen the research on this yet, I will contend this is intuitively how I understand societies. We want to understand if we're having an impact on a society, it will follow a similar pattern. Then you can see as it changes, this work can identify when a system is about to tip,"
},
{
"end_time": 1218.234,
"index": 50,
"start_time": 1194.019,
"text": " Right. But at this point, I haven't seen anything that could say what it's going to tip into. So if you think of the bifurcation diagram, right, is that it's like, hey, we know it's about to go one of two directions. Right. But we can't tell you which one it's going to go to. Right. But you can. So I would say you could see that we look at even the U.S. society right now. Right. As a democracy is in crisis. Right. Is like, OK, what like how's this going to tip into the future?"
},
{
"end_time": 1246.869,
"index": 51,
"start_time": 1219.036,
"text": " Signals have changed, the butterfly has kind of flapped its wings and so will we tip into another strange attractor? How will our society evolve? So that kind of shows on this picture here, we have a resilient status quo and then you get to a fragile status quo and you would theoretically tip into another type of status quo. The reality is we don't understand it."
},
{
"end_time": 1266.869,
"index": 52,
"start_time": 1247.5,
"text": " We're getting better, but this would be from an understanding perspective. How do we institute policies that create stable, thriving societies? Through your kind of meta-consciousness, we don't get hands down, but we're trying."
},
{
"end_time": 1299.855,
"index": 53,
"start_time": 1270.401,
"text": " All right. And so the other bit too is that, you know, so now it's okay, what tools can we do? And again, I've no idea. I'm just saying this is the little path that I'm on to try and help figure this out. Right. But the one thing I do know is that better understanding is not effective policy, right? Because you have these emergent systems, like it's like, well, I know that this is having this impact. It's like, well, that's great, but what can replace it? All right. Like what new policy can I put in place to replace what's happening in order to produce this new emergence behavior of the system?"
},
{
"end_time": 1319.48,
"index": 54,
"start_time": 1300.486,
"text": " All right, so we need to develop new tools. This is primarily what my PhD was in computational social science, right? It's saying like, hey, we got to use these artifacts. We got to generate essentially virtual laboratories of the world so we can test things out in silico, as they say, prior to actually implementing it."
},
{
"end_time": 1341.664,
"index": 55,
"start_time": 1319.77,
"text": " Right. And so you have generative science, which is fundamentally different from reductionist science, right? Where it's like, uh, I'm not trying to, uh, grow or I'm trying to grow the system, right? Cause I can't just disprove things. Right. Um, and then, uh, so create simulations, you know, essentially these virtual laboratories, we can try stuff out. And so that is happening. Uh, so, uh, George Mason, Oxford and"
},
{
"end_time": 1362.244,
"index": 56,
"start_time": 1342.056,
"text": " Princeton built a simulation back in 2014 to try and replicate the 2008 housing market crisis. In this kind of science, because you can't just prove stuff, you do what's called pattern-oriented modeling. They had nine statistics from that time period on the"
},
{
"end_time": 1383.831,
"index": 57,
"start_time": 1362.551,
"text": " The Central Bank of England uses it for London and the Central Bank of Norway has one for Oslo where essentially they try out policies in Silico and then they try to implement those policies"
},
{
"end_time": 1404.923,
"index": 58,
"start_time": 1384.087,
"text": " What am I doing personally for this? First off, you're trying to increase tech literacy."
},
{
"end_time": 1431.544,
"index": 59,
"start_time": 1405.179,
"text": " There's a lot of myths and stuff there about AI and what it can do and fear and too much people watching cable news. Right. And so it's, you know, the luckily our faculty, we have a PhD in curriculum and instruction who's very good. And so we're trying to feel, well, how do we, instead of like giving people the classic data science thing, it was like, okay, we're going to start you out coding. You're going to learn, you know, all these statistics and these data science tools be like, because we deal with them."
},
{
"end_time": 1440.845,
"index": 60,
"start_time": 1431.869,
"text": " Hey you really know a lot about let's say you know political science or"
},
{
"end_time": 1469.753,
"index": 61,
"start_time": 1441.22,
"text": " Then we'll give you an analytic that meets you in your space where you're an expert. Then as you can realize that, hey, this analytic is not exactly right, you could just flip it to the code. So if you're familiar with Jupyter Notebooks and stuff like that and not see the code right away, but see the choices that that developer made and then start to explore that computational tool at deeper and deeper levels based off where you feel most comfortable."
},
{
"end_time": 1498.166,
"index": 62,
"start_time": 1470.265,
"text": " Right, and that kind of takes best practices of adult learning. It's a put it up there, but one of my thesis students built cell toolbar, which is work. It's not exactly where it should be yet, so you might want to give it a couple of months, right? But it's the idea is that you can hide code better and store knowledge so it's easier for people that aren't technically literate to interact with the code, right? And that's on the pie PI library says extension for Jupiter notebooks, right? And then and then I still continue work on Mesa, agent based modeling and Python."
},
{
"end_time": 1525.691,
"index": 63,
"start_time": 1498.677,
"text": " Right. Which is like, how do we get, how do we develop a simulation ecosystem? So this is very much my own heavily biased opinion. But if you look at like, if you're familiar, like scikit-learn, which is great, you can, you know, build all sorts of machine learning models and like three lines of code more or less, right. And you've got like TensorFlow and PyTorch for TensorFlow is from Google, PyTorch is from Facebook. And so you can very easily build out these deep learning models and things of that nature. Right."
},
{
"end_time": 1551.032,
"index": 64,
"start_time": 1526.084,
"text": " Um, but what does that look like for the age of base modeling ecosystem? So again, bias is, which is I would say is order of magnitude harder to have an ecosystem where I can rapidly bring together, uh, the pieces for and build a model, you know, quickly that's relevant, right? So, uh, the example I use is like an elementary school principal during like a COVID outbreak, right? Like how can that person try out, uh, you know, um, policies"
},
{
"end_time": 1580.179,
"index": 65,
"start_time": 1551.032,
"text": " that are valid for their particular school based off their situation, right? Like how can you say, okay, I got the mindset of a kindergartner and I got the mindset of a sixth grader, right? And now this is exactly in how my school's laid out, whether it's in remote Kansas or, you know, New York City and what policies can I put in place, right, to try this virtual laboratory that will dramatically mitigate the outbreak in my institution, right? It might be as simple as make sure a teacher watches the kindergartners wash their hands."
},
{
"end_time": 1598.575,
"index": 66,
"start_time": 1580.776,
"text": " How do we create an environment we can have simulations that you rapidly put together?"
},
{
"end_time": 1619.684,
"index": 67,
"start_time": 1598.831,
"text": " All right, so questions."
},
{
"end_time": 1642.944,
"index": 68,
"start_time": 1621.067,
"text": " So you mentioned the housing crisis model and you said that they used eight of nine indicators and that the Bank of England is now using this model, but given the butterfly effect, you're skipping an indicator. How do you resolve that kind of quandary? I think you ever resolve it, right? Because it's always going to be in there."
},
{
"end_time": 1670.879,
"index": 69,
"start_time": 1642.944,
"text": " Right, so you can you can you got like an event horizon if you will we could kind of see out this far but it's always kind of you're always having to constantly tweak the knobs and update your understanding of the situation right so you can you can never solve it it's i would say unequivocally an impossible problem to solve i think the technical term from chaos theory is it's deterministically unpredictable right which means the past the past creates the future but because of all these rounding errors and variables it's you can't actually predict it"
},
{
"end_time": 1697.671,
"index": 70,
"start_time": 1673.2,
"text": " Thank you so much. It's very enlightening and I don't really know how to ask this but there's studies about the morphic field because the thing is like the United States when he started the country they had really good people worrying about the people for the people and everything and they also had a saying of in God we trust because they believe in the supernatural"
},
{
"end_time": 1727.927,
"index": 71,
"start_time": 1698.2,
"text": " So I feel like a lot of the science sometimes they're forgetting about this supernatural thing that has been being studied from, you know, ancient Egypt or mysticism occult sciences. So the morphogenetic field is the thing that is not indivisible and is still like is right now is strong like for therapy and is actually being implemented in Brazil in the judiciary and the health system."
},
{
"end_time": 1748.49,
"index": 72,
"start_time": 1728.49,
"text": " because it has shown effectiveness because this field actually is sky. We have the Akashi field that is our cloud in a base and the morphogenetic field is our programming. So the way that things are programmed, we can go back and reprogram, but it's more on the"
},
{
"end_time": 1773.626,
"index": 73,
"start_time": 1748.78,
"text": " Occult scientists, I guess, I mean the supernatural. Nikola Tesla said, like, when we start studying the non-physical phenomena, we will have more advances in science in a decade than in centuries. And I do believe we need to go back to the sacred, like, ancient scriptures. And not only that, like, the, like, the secret"
},
{
"end_time": 1804.821,
"index": 74,
"start_time": 1775.162,
"text": " I'm sorry, I'm nervous. I'm not really good with talking, but are you familiar with Manly P Hall? With who? Manly P Hall. No, he is the writer of the secret teachers of all ages, right? I'm not, uh, I can't remember the book now, but there's a lot of wisdom in that. And right now what we need in consciousness is also the technology, understanding wisdom, because like quantum computing is like you getting information from all sources in order to build this intelligence."
},
{
"end_time": 1828.541,
"index": 75,
"start_time": 1805.23,
"text": " We have all this information. We're only missing this part that is like doing the things that is right. Because even everything that the government has done, that you're saying that now we need to be more clear about it, is still there. It still shows on a morphic field. Whenever we do this right through, you know, the supernatural, like using love, compassion,"
},
{
"end_time": 1853.899,
"index": 76,
"start_time": 1829.019,
"text": " We can rewrite what was done wrong and I think we can have a better society and better leaders because that's what we're needing right now. We need someone that is going to guide us without interest of you know like businesses or things like that. People that really care about people and we have that we care about each other but in a way we've been programmed to think differently. Hear that sound?"
},
{
"end_time": 1880.896,
"index": 77,
"start_time": 1854.804,
"text": " That's the sweet sound of success with Shopify. Shopify is the all-encompassing commerce platform that's with you from the first flicker of an idea to the moment you realize you're running a global enterprise. Whether it's handcrafted jewelry or high-tech gadgets, Shopify supports you at every point of sale, both online and in person. They streamline the process with the Internet's best converting checkout, making it 36% more effective than other leading platforms."
},
{
"end_time": 1907.022,
"index": 78,
"start_time": 1880.896,
"text": " There's also something called Shopify Magic, your AI-powered assistant that's like an all-star team member working tirelessly behind the scenes. What I find fascinating about Shopify is how it scales with your ambition. No matter how big you want to grow, Shopify gives you everything you need to take control and take your business to the next level. Join the ranks of businesses in 175 countries that have made Shopify the backbone."
},
{
"end_time": 1932.79,
"index": 79,
"start_time": 1907.022,
"text": " of their commerce. Shopify, by the way, powers 10% of all e-commerce in the United States, including huge names like Allbirds, Rothy's, and Brooklyn. If you ever need help, their award-winning support is like having a mentor that's just a click away. Now, are you ready to start your own success story? Sign up for a $1 per month trial period at shopify.com slash theories, all lowercase."
},
{
"end_time": 1944.633,
"index": 80,
"start_time": 1932.79,
"text": " Go to Shopify.com slash theories now to grow your business no matter what stage you're in Shopify.com slash theories."
},
{
"end_time": 1963.251,
"index": 81,
"start_time": 1944.957,
"text": " I'm going to take what you just said and I'm not going to disagree. I'm going to put it in these terms. What I would say is you're arguing for a different perspective that's superior. They say, hey, I think this is superior. If we implement these types of policies, we as a humanity, society of different levels will do better."
},
{
"end_time": 1986.237,
"index": 82,
"start_time": 1963.251,
"text": " Right. And so I'd say is like using the kind of idea of a explore, explore, exploit algorithm is like you can say that. And then as people implement that, the trick is, are we optimizing our environment, right? To allow the ideas that are working, right? To actually bubble to the surface. Because the trick is there's a lot, but a lot of bad ideas that have bubbled to the surface that didn't work, right? That, uh,"
},
{
"end_time": 2009.633,
"index": 83,
"start_time": 1986.578,
"text": " that people encourage, you know, for like, like Nazis and fascism and stuff like that, right? So that I would say, I don't disagree there. It's just like, okay, so you say, hey, these ideas, we're seeing these in Brazil, they're working really good, right? How do we optimize the policies, you know, technological environments of our society so that those ideas can be validated?"
},
{
"end_time": 2039.445,
"index": 84,
"start_time": 2009.633,
"text": " Right. And bubble up to the surface. Right. And I mean, just politically, as you can all see in our very kind of caustic environment in the U.S. right now, right, is that people will say something like, hey, this is right. Right. Like, hey, you know, it's been shown as like a controversial idea that you could take somebody that's like, say, homeless and alcoholic. And it's better just to put them in like a home where they have access to alcohol because then you're actually spending less money on ER visits and stuff like that. But that's a very controversial idea."
},
{
"end_time": 2067.995,
"index": 85,
"start_time": 2039.974,
"text": " Right. And so it's, you know, I don't know the answer, but that's, I would say the bit is how do you take the ideas that even might not be intuitive, right. And say like, Hey, if we do this, it actually works really well. Right. And you know, the trick is we start going against culture, right. Then people like, no, I'm not going to, I will literally not believe that no matter what happens. Right. Like the earth is flat. Uh, yeah. Thank you for the talk. Uh, Tom, uh, I enjoyed it. I want to press you a little bit on this notion of stability."
},
{
"end_time": 2097.705,
"index": 86,
"start_time": 2068.285,
"text": " I guess in this sense, right? So part of what the project is to maintain stability as a way to maybe effectively govern or create a more just society. I guess the way stability is in this context, it makes it sound like it's just stability is always good, right? But in some ways, I wonder if like, actually, that should be our kind of maximizing principle, right? Because you might think, well, say something unjust is going on, the complete opposite of stability seems to be what actually would be the normative, like how you ought not to"
},
{
"end_time": 2118.49,
"index": 87,
"start_time": 2098.439,
"text": " So I wonder, like, you know, say we put certain practices in the place, right, which always maximize stability, but you have unethical or unjust society or something like that. Well, then it seems like you're just helping to maintain that particular thing. So I'm just wondering in some ways, like, how do you, because the question I get from the talk, right, is this is to have a positive impact, right? In some ways."
},
{
"end_time": 2136.954,
"index": 88,
"start_time": 2118.985,
"text": " And I'm wondering, well, actually, it seems like maybe stability could have a very negative impact in some ways. So I'm wondering, like, how does that get counterbalanced in the work? So I mean, that's an awesome question, Gary. That's phenomenal. I'd say it's you're absolutely right. Like, it can never stop moving. Right. I think the trick in economics is if you're going to have change, you just want it to not be completely abrupt."
},
{
"end_time": 2154.77,
"index": 89,
"start_time": 2137.295,
"text": " Right, so much like a democracy, in my opinion, after spending my adult life fighting counter-insurgency and stuff, is that democracy is nobody wins. Once somebody wins, they just try and steal all the power. So I figure that stability is more like,"
},
{
"end_time": 2182.705,
"index": 90,
"start_time": 2155.299,
"text": " It's not like some kind of utopian stability that can't exist. It's more like how do we, as we're constantly evolving and maybe going through these spots of punctuated equilibrium, do we make sure you don't have like a massive famine or like this have a couple of times in China where literally millions of people are dying, right? Like how do we essentially shift from one kind of strange attractor, go through a punctuated equilibrium to a new kind of homeostasis stable state, right? How do we"
},
{
"end_time": 2210.026,
"index": 91,
"start_time": 2183.251,
"text": " help manage that so it's not as traumatic. Great talk, Tom. So there's been a resurgence around agent-based thinking recently with the LLM models. What are your thoughts about taking this agent-based modeling, which has been around for some time with all this great work in complex systems, how do we tie in the LLMs? Do you envision putting an agent on to each little agent, rather put an LLM inside each little"
},
{
"end_time": 2239.377,
"index": 92,
"start_time": 2210.811,
"text": " Each little spot. I think so. And actually Mark, Mark and I've kind of talked about this for a while. Um, but we, I mean, it's pretty straight. If I had a student do this, uh, last quarter directed study is that you can link up LLMs to talk to each other as essentially like little agents. Right. Uh, I think in the end, like it's, you know, you start talking abstractions, uh, you know, if complex systems are an adaptive network, what's effectively on LLM, it's an adaptive network. Right. So I think it's some of the problems that open AI particular is trying to solve when you get to AGI."
},
{
"end_time": 2265.964,
"index": 93,
"start_time": 2239.582,
"text": " is in my own mind and this is like very much conjecture right is that at some point the kind of tools will probably merge together right and then just how they're presented by the computer scientists is just different abstractions so they're more understandable but i think it's both adaptive networks and now it's like well how these adaptive network functions to solve problems because you can imagine like the housing simulator but each one actually wants a house or it thinks it does yeah yeah so there you go"
},
{
"end_time": 2288.183,
"index": 94,
"start_time": 2266.34,
"text": " So you have a, you had a chart there where we're going towards a singularity. One of the interesting things about getting close to a singularity is the laws start breaking down, whether you're leaving from a singularity or going towards a singularity. And I would suggest that a lot of the assumptions are beginning to break down."
},
{
"end_time": 2316.971,
"index": 95,
"start_time": 2288.677,
"text": " For example, the assumption that policy matters because used to there was a policy people would follow the policy. Now we have so much not following the policy. There's ways around the system. Um, there are other effects like facts don't matter as much. And so a lot of the assumptions we have, those are beginning to break down as we're getting closer to the singularity. So we almost need a meta."
},
{
"end_time": 2344.94,
"index": 96,
"start_time": 2317.398,
"text": " Consideration about what we would do. Do you have any thoughts about that? I I I would say so I'm not sure we're getting to the singularity I'll go to this slide Right is I think we're at that back that back hill up there where it's a more fragile status quo as the fitness Landscape has changed like our chances of tipping into a new dynamic are just increasing right? We're seeing that all the time with numerous unprecedented historical events, right? so I think that's it and I think I"
},
{
"end_time": 2353.37,
"index": 97,
"start_time": 2345.452,
"text": " My only other thought would be that it is like kind of fractal in nature, right? And kind of like yesterday I started talking the universe and the universe is the universe is all the way up."
},
{
"end_time": 2380.691,
"index": 98,
"start_time": 2353.592,
"text": " Can you use tools like this to essentially view your world like you're in the universe up? To the point of things breaking down, I think we're definitely seeing that doing some stuff on the open source ecosystem. Like, hey, this defies all traditional economics. Why are people volunteering their time to do these massively valuable inputs when they should be getting paid for? What does that mean for an economy?"
},
{
"end_time": 2407.875,
"index": 99,
"start_time": 2380.964,
"text": " All right, so I think you're absolutely right. I'm not sure we're going to the singularity though. We are going faster, but I think we are definitely in a spot of more, like a more fragile state where we might tip into some new type of reality, right? Not like, like just another historical reality, right? Like, you know, 18th century and 19th century, not like a, like a, I guess more like tipping into a singularity. So I wanted to ask you a question that's related to those last two questions."
},
{
"end_time": 2434.172,
"index": 100,
"start_time": 2408.439,
"text": " You started your talk with the basically diversity of thought being a good thing and leading to expansion of knowledge and cultural revolution. I can't help but think to recent history in the 1850s, it seems like there was a lot of diversity of thought in America. And I would argue that that could be perceived as a bad thing because it led to the Civil War, 600,000 Americans dying. And so that's like a saddle point basically. And the way out of that saddle point was"
},
{
"end_time": 2453.08,
"index": 101,
"start_time": 2434.65,
"text": " You know, arguably Abraham Lincoln was implementing what would be considered very fascistic policies like imprisoning dissenters, political dissenters. So two part question is, um, you know, how do you distinguish good diversity of thought from bad diversity of thought that's destabilizing and, um,"
},
{
"end_time": 2480.845,
"index": 102,
"start_time": 2453.541,
"text": " You know, to what extent or what value can you or how can you implement a network that is going to be able to feed forward knowledge from, you know, centuries of lessons learned that can actually impact the future evolution? I can't help but think that any future civil war that might be avoided by a recollection of the not too distant past, the horrors. That's it. So the short answer is I have no idea because you're absolutely right, right? Like the right"
},
{
"end_time": 2508.404,
"index": 103,
"start_time": 2481.288,
"text": " because we just we don't understand it right like we can't uh we don't if you i mean you go back to the like the civil war or any world war two there was there was moments that that were very brief and very important you know comparatively like eddiesburg um the battle of midway and stuff like that where you could have had a vastly different reality right uh that that came out of it and and so i guess what i would say you just said like that to me epitomizes the problem"
},
{
"end_time": 2534.258,
"index": 104,
"start_time": 2508.643,
"text": " We have no scientific or rigorous way to say, well, in this situation, you should do these things and that will help push the system in this direction. Right? Like, like everything I've seen is that's like we're making small progress, but we functionally don't have tools that will allow us to do that. Right. And, and so that's what, you know, now, now spend 15 years trying to figure that out. Yeah. It's a tough one, right? Yeah. So."
},
{
"end_time": 2558.422,
"index": 105,
"start_time": 2534.923,
"text": " It seems like in those last few slides, you're talking about human tractable and interpretable algorithms with human tractable and interpretable variables so that someone can take their local situation and sort of tailor the algorithm to them, like your example of the school with COVID. In situations like that where the person's answer from the algorithm actually matters in them making real local policy decisions,"
},
{
"end_time": 2588.268,
"index": 106,
"start_time": 2559.087,
"text": " How do you reconcile this idea of building human tractable and interpretable algorithms with the fact that it seems like the non tractable and non interpretable algorithms where you just throw in 2000 data points and ask it to predict something seem to have much more predictive power and would be much better for local policy decisions? Well, so, so this is what I'd say and this is what I've given this conference because that way people can call BS and I know there's a lot of folks very proficient at LLMs and stuff like that. All right. Um, is that,"
},
{
"end_time": 2609.957,
"index": 107,
"start_time": 2588.473,
"text": " The challenges with like these current tools, right, is I can break those tools if the world changes, right? So my understanding was when COVID hit and people's behavior changed drastically overnight and did something that never seen before, right, that that broke a bunch of like the tech giants tools, right? Because the trick is we're not trying to predict the future. We're trying to shape the future."
},
{
"end_time": 2629.991,
"index": 108,
"start_time": 2610.469,
"text": " I like surface choice between genocide war world war three and nuclear weapons we don't want that to happen how do we implement things that help us that put us on a different path so i add so far and i'm not like me deep into so the large language model stuff"
},
{
"end_time": 2656.015,
"index": 109,
"start_time": 2630.418,
"text": " is I haven't seen, from my perspective, I haven't seen models that can account for that, right? That can say, hey, here's a policy you should do to create this new future. In fact, the only thing that I've seen that even gets close, and this kind of echoes Harmut's talk yesterday, was the BOM experiment from quantum mechanics, where it's like, hey, it's actually not small, it's all the way up. In the BOM experiment, they're able to show that quantum could show what didn't happen."
},
{
"end_time": 2672.722,
"index": 110,
"start_time": 2656.817,
"text": " I just wanted to go back to in the beginning you talked"
},
{
"end_time": 2689.462,
"index": 111,
"start_time": 2673.422,
"text": " A little bit about meta consciousness. Could you give us a definition of what do you mean by meta consciousness and what does that entail as far as looking at these complex systems?"
},
{
"end_time": 2713.439,
"index": 112,
"start_time": 2689.462,
"text": " So kinda classic when i've been exposed to on this is jane jacobs like the wisdom of cities right like it's amazing you can look at cities you like tammany hall in the eighteen hundreds in the u.s. and lagos nigeria now and it's like they they form a kind of collective consciousness on how they function writer even london during like dickinsonian times right so you have these you know see have these."
},
{
"end_time": 2740.06,
"index": 113,
"start_time": 2713.439,
"text": " I know the Santa Fe Institute is really looking into that and they started a new journal, I think it's Jessica Flack and Scott Page, called Collective Intelligence. I think the book is Solaris. There's a book called Solaris, then there's a Russian movie, then there's a George Clooney movie that sucks."
},
{
"end_time": 2763.695,
"index": 114,
"start_time": 2740.282,
"text": " But in that, it's the idea that this one planet is a singular consciousness, and it's all just functioning together. I will give you a definition, I'll give you the field of study, and I would look at the Collective Intelligence Journal that's only a year old or so now. But just to clarify, it's"
},
{
"end_time": 2780.077,
"index": 115,
"start_time": 2764.002,
"text": " collective intelligence, not like sensational consciousness of a society. Yes. Yeah. Yeah. Right. But that's, well, I mean, cause that goes to the other, but we could go down, uh, downward, right? Where are, are we as a conscious human being, just the byproduct of a microbiome that needed to survive."
},
{
"end_time": 2804.633,
"index": 116,
"start_time": 2780.469,
"text": " Extra value meals are back. That means 10 tender juicy McNuggets and medium fries and a drink are just $8."
},
{
"end_time": 2832.654,
"index": 117,
"start_time": 2804.633,
"text": " I just wanted to make a quick comment on the collective intelligence, consciousness kind of thing. I always like to think of our vegetables, fruits, certain plants that we like to use to fuel in different altered states are kind of using us to propagate their species in a sense. Like if you think about a lot of the animals that we eat or the plants that we eat,"
},
{
"end_time": 2862.449,
"index": 118,
"start_time": 2833.08,
"text": " What better way of ensuring their survival than being tasty or nutritious to us and now they're just everywhere. So it is kind of a very interesting dynamic to think about. That's a good defer to an evolutionary biologist. Hi, so I lived in New York for 30 years where the conversation was a lot about learning organizations and just-in-time knowledge that sort of thing. So now I live in a rural area in Florida where the biggest problem locally is whether or not people are allowed to"
},
{
"end_time": 2891.015,
"index": 119,
"start_time": 2862.705,
"text": " park big trucks in their backyard and they resent any kind of opinions coming from anywhere else about, you know, the aesthetics of it, the utility of it, all of that sort of thing. So how do you get past like the local culture that totally rejects any kind of input from a larger outside organization? Well, so I don't think you do. This kind of goes, I think, to your question a bit, Gare, which is"
},
{
"end_time": 2916.8,
"index": 120,
"start_time": 2891.271,
"text": " I don't know if you know who David Kilcullen is. He's an Australian, one of the world's foremost counter-insurgents. He was advising the US at the most senior levels through Iraq and Afghanistan. He wrote a book called Out of the Mountains, where he proposed the theory of competitive control. Out of the mountains? In that, the idea is, we're in these homeostasis stable states, but you're never going to have"
},
{
"end_time": 2945.316,
"index": 121,
"start_time": 2917.312,
"text": " homogeneity in a population. I wouldn't necessarily say that diversity of opinions isn't necessarily bad, but they do select over time. Because rural Florida is vastly different than urban New York City, you're going to select different things based off whatever your optimal dynamic is. When you start to get to problems, it's like, well, now your optimal dynamic is actually, say,"
},
{
"end_time": 2973.217,
"index": 122,
"start_time": 2945.316,
"text": " the fertilizers in Iowa, you know, creating red tides and killing the fish down in Texas. So how'd you balance that? Right? And that's, so it's, I guess I just say, you always want heterogeneity in any type of, you know, population. And then it's, what are the, how'd you set up dynamics to optimize that so you can, you know, let a lot of people just go about their life and not be interrupted, but also optimize how that society continues to function and doesn't, you know, get a war with itself."
},
{
"end_time": 2997.995,
"index": 123,
"start_time": 2974.036,
"text": " Firstly, thank you for watching. Thank you for listening. There's now a website, kurtjymungle.org, and that has a mailing list. The reason being that large platforms like YouTube, like Patreon, they can disable you for whatever reason, whenever they like."
},
{
"end_time": 3024.462,
"index": 124,
"start_time": 2998.251,
"text": " That's just part of the terms of service. Now, a direct mailing list ensures that I have an untrammeled communication with you. Plus, soon I'll be releasing a one-page PDF of my top 10 toes. It's not as Quentin Tarantino as it sounds like. Secondly, if you haven't subscribed or clicked that like button, now is the time to do so. Why? Because each subscribe, each like helps YouTube push this content to more people like yourself"
},
{
"end_time": 3042.961,
"index": 125,
"start_time": 3024.462,
"text": " Plus, it helps out Kurt directly, aka me. I also found out last year that external links count plenty toward the algorithm, which means that whenever you share on Twitter, say on Facebook or even on Reddit, etc., it shows YouTube, hey, people are talking about this content outside of YouTube, which in turn"
},
{
"end_time": 3071.203,
"index": 126,
"start_time": 3043.166,
"text": " Greatly aids the distribution on YouTube. Thirdly, there's a remarkably active Discord and subreddit for theories of everything where people explicate toes. They disagree respectfully about theories and build as a community our own toe. Links to both are in the description. Fourthly, you should know this podcast is on iTunes. It's on Spotify. It's on all of the audio platforms. All you have to do is type in theories of everything and you'll find it. Personally, I gained from rewatching lectures and podcasts."
},
{
"end_time": 3091.135,
"index": 127,
"start_time": 3071.203,
"text": " I also read in the comments"
},
{
"end_time": 3114.616,
"index": 128,
"start_time": 3091.135,
"text": " and donating with whatever you like there's also paypal there's also crypto there's also just joining on youtube again keep in mind it's support from the sponsors and you that allow me to work on toe full time you also get early access to ad free episodes whether it's audio or video it's audio in the case of patreon video in the case of youtube for instance this episode that you're listening to right now was released a few days earlier"
},
{
"end_time": 3139.531,
"index": 129,
"start_time": 3114.616,
"text": " Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon"
},
{
"end_time": 3163.933,
"index": 130,
"start_time": 3144.121,
"text": " . . . . . . . . ."
}
]
}No transcript available.