Coaching the Creators: Inside the Minds Building Frontier AI with Executive Coach Joe Hudson

Hello and welcome back to The Cognitive Revolution.

Today we're doing something a bit different. Normally we focus on cutting-edge AI ideas, research projects, application architectures, usage trends, policy proposals, and visions for the AI-enabled future. But today we're getting a glimpse into the psychology, emotional patterns, and decision-making processes of the people who are developing some of the most important and potentially transformative of those ideas. My guest is Joe Hudson, founder of The Art of Accomplishment and coach to executives and research teams at multiple frontier AI developers, including Sam Altman. About Joe, Sam wrote, quote, Joe coaches the research and compute teams at OpenAI. I truly enjoy working with him. One of his superpowers is that he deeply understands emotional clarity and how to get there. This will be one of the most critical skills in a post-AGI world. So what has Joe learned from his interactions with the brilliant, often quite young technologists who work together in an environment that prizes intelligence above all else, under the unique pressure of knowing that their work could either solve humanity's greatest challenges or, in the worst case, cause our extinction? This is, in all honesty, a hard conversation to summarize. But the good news is that Joe reports that he hasn't met anyone in AI who doesn't seriously wrestle with the ramifications of their work. Some, of course, do have blind spots. Some are perhaps more optimistic than the situation warrants. And some might be overly focused on creating AGI first. But Joe says that the question of, am I doing something good for humanity, weighs heavily on all.

He also argues—

and this is something that I've come to believe with pretty high confidence as well— That given the presence of web-scale data and web-scale compute, some form of powerful AI is inevitable. Which means that the key question of our time is not whether we can prevent it from ever being developed, but what form it will take. And whether that form will be carefully chosen via a proper deliberative process, or dictated by inhuman market pressures and race dynamics. With this in mind, and with hunger strikes ongoing outside frontier AI companies' offices around the world, Joan warns the AI safety movement against shaming tactics and instead recommends a more encouraging approach meant to inspire people at frontier AI companies to become the best possible versions of themselves. This, he believes, is the best way to improve the odds that individuals in critical decision-making roles have both the psychological strength and the practical wisdom needed to make good choices under extreme pressure On behalf of all humanity. Whether you agree or disagree, if you listen with an open mind, I think this inside look at the human side of AI development provides unique and valuable context and will ultimately deepen your understanding of the AI landscape. This is executive coach Joe Hudson.

Joe Hudson, founder of The Art of Accomplishment. Welcome to The Cognitive Revolution.

Thanks, man. Good to be here. Good to be with you.

I'm excited for this conversation. I think it's going to be a different one from our usual fare, which is very focused on research and application and policy. And this will be a look, hopefully, on a deeper level into the thinking, the mindset, maybe the emotional life of people that are building at the frontier of AI. And I think you have a really unique window into that because I've, you know, got multiple referrals and I definitely go do my homework. So, I went and did a little reference checking to make sure this is true. And it does check out that you are working as— I guess maybe you should just introduce yourself.

How do you—

what do you call yourself? Are you a teacher, guide, guru? What are you and what do you do?

I mean, I just think of myself as a coach generally. And But we do, you know, because I don't have a lot of time to be able to coach one-on-one, we do do these classes. And so, both in person that are invite only generally, and then these online classes. And so, I had the chance to work with people deep and high up in all the major labs in the Bay Area. And so, I've had a chance and I also, you know, coach the top management at the research side of, well, and compute and, you know, infra and OpenAI. So I get to work with a lot of brilliant, really kind people. So it's really nice. Yeah.

So can you give maybe just a little bit more general background on like what your coaching entails? You know, What, you know, leaving the AI side to come.

Yeah.

What is it that you help people do?

Yeah. So, generally, the way I think about coaching is you meet people where they're at, and you see where they want to go, and you help them get to that place. That's generally— and if you happen to meet somebody who wants to go someplace that you think is unethical, you don't, you know, you don't coach them. But I think it's unethical to decide what's best for somebody else and then to coach them to that end. So, it's very much about following the people. The thing, the way that I do it that's, I think, unique is I was a venture capitalist for a while, so I can very much talk to the technical parts of a business, whether it's marketing or a CTO or CFO. I know enough about their jobs to be able to really coach in that way, and so oftentimes, Coaching will start on a very strategic level or sometimes even tactical level. And then what you'll start noticing is that there's things that are happening in their life, there are patterns that are occurring that are stopping them in business. And so then we go to the deeper level of really looking into what's stopping them, what's the psychology behind what's going on, and then how do we change that thought process. And so there'll be a lot of work on emotions and what emotions are being held back, because emotions dictate our decision-making, neurologically speaking. And there's also a lot of work on how the voice in the head is talking to somebody. There's a lot of— some work on like childhood and like what happened there. And so you're looking for these patterns that are holding them back, and then you— if they're, if they're interested in, you know, unearthing those patterns and changing those patterns, then I have a— Crap ton of tools to help them do that.

What sort of patterns would you say are like most common? And how does that vary, if at all, from like a, you know, society-wide level to the AI vertical? Are there— do you— is it mostly the same stuff or is it different?

Is it pattern— you mean patterns with people and AI, or do you mean patterns with leadership? Which— because it can be very different. Um, typically, take them one at a time.

How about that?

Yeah, yeah. So typically in leadership, there's a lot of common things, but there's always outliers. There's always, you know, there's nothing more than 80% common. Um, but one of, like, a very common one is self-sufficiency. So you'll have people who, um, feel like, who were raised in such a way that they had to do everything on their own. They couldn't really depend on somebody for emotional or financial or some sort of support so that they learned, if I'm not gonna do it, nobody's gonna do it. Those people often rise to leadership, and then often what happens is they don't empower other people very well because they always are like, I'm gonna have to do it. So they'll step in and they'll do it, or they'll micromanage, or they'll go around proving that nobody can actually take care of them or take care of the thing. So that would be Yeah, or some, you know, maybe it's not all of the above, but it's one or two of those things will be there. And they'll often feel very alone in their work. They'll, you know, they'll— and their frustration will often be, you know, it's lonely at the top. They'll do that kind of thing, even though, you know, there's an organization of 10,000 people beneath them that are all concerned about their happiness on a daily basis. You know, what does the leader think of me is like a constant thought process. And then they feel alone, which is like a ridiculous notion, but— So that would be a really common one that you see in leadership. As far as AI goes, I haven't seen some common patterns across AI. There's definitely a lot of, I would call it like Aspie, Asperger's. There's a lot of that level of it's easier to deal with things than it is to deal with people. I understand things better than I understand people. It's not ubiquitous by any stretch, but there's a lot of that in in the research departments. In general, I find them to be very kind, but also extremely defined by their work. Meaning that, you know, the way that Kim Kardashian or somebody might take it personally if somebody calls them ugly, they're going to take it personally if somebody says that their research isn't, you know, good or isn't, you know, as smart as it could be or something like that. So, the Pecking order is really based on like what you can produce and how smart you are and how how how good your ideas are. So there's that self-definition that's a pretty significant pattern of accomplishment defines you, which can hold you back often because it it tightens your thinking. Oftentimes, to have a novel thought, you need to you know open up your thinking and think about things in a completely different way. And if you're It's like writer's block. If somebody's, I gotta get good pages out, I gotta get good pages out, you're gonna get a lot less good pages out. And so that self-definition doesn't really allow some of the more innovative thought processes to come. So that would be on just like on a research level. I would say on a level of, of management, I've seen everybody has this deep desire to do good for society. I haven't met anybody who doesn't want that. I'm sure there's people out there who don't, but I haven't interacted with them. But what I notice is that, you know, this is, as I talk about it a lot, this is I had a really cool interaction early in my career, and I got to meet one of the titans of radio in his final days. And he was telling me how radio was gonna be like this thing that made the world a better place. It was gonna, you know, like, it's hard to go back to that idea, but, oh, all of a sudden we can transmit these ideas and education, and all of a sudden humanity can come together. It's gonna be this amazing thing. And And, you know, it turned into shock jock radio and, you know, advertisement. And so, I think we've seen that with television. We've seen that with the internet. I was around for the internet where everybody was like, this is gonna make everything better. And it surely has made other things better. So, I think one of the things that I see is that optimism without the— without like looking at the history.

Yeah.

deeply, you know, they're— so, so that's something that I see is— but that's something that you see in a lot of entrepreneurs is that, or leaders of like doing new and innovative stuff, is there's this deep optimism. But that's not always like checked in reality. And so, so that I think is another common thing that you see. I think they're all open-minded. That's another pattern that they have generally. They're all innovative people. They all have a strong belief that technology can make the world better. You know, there's certain things that they have in common, but they're extremely different people. There's no one, you know, brand of people in AI that I've seen so far except for hyper-intelligent. You know, I haven't met anybody in AI who's not hyper-intelligent.

So do they come—

I mean, the history thing is super interesting. The research unblocking is right down the fairway of what I would expect people to come to you for. And the history thing sounds like something that I would doubt that they were coming to you for. Like, are people coming to you and saying, like, I want to better contextualize my work?

There's nobody who's I haven't met anybody in AI who doesn't go home and think about the ramifications of their work. So it sits on your soul. Am I doing something that's good for humanity is, I think, a question. How do I make sure that this is good for humanity? So they are wrestling with that. I don't know anybody in AI who's not wrestling with that, which gives me a lot of faith and confidence. But yeah, I haven't met anybody who's not deeply concerned with it, from Sam to the lowest-level person in the smallest AI shop that I've interacted with. Everybody's concerned about it.

So, what do you help them— if I come to you and say, I'm doing this research, maybe I would like to be more productive, you can help me potentially get unblocked. But then I also have this general purpose concern that, I mean, Elon Musk said this in the Grok 4 launch. I thought it was really a startling moment, honestly, where he said, is it good for humanity? Is it bad for humanity? I think it's probably good, but I'm really not sure. And even if it's not, I've kind of made peace with the fact that I want to be around to see it happen. And I was like, wow, that's a statement.

Yeah.

How do you help them get unblocked, first of all? And then, how do you help them deal with— or is it deal with or is it do a better job? Like, what is the practical kind of upshot of how they can take this worry and somehow be better?

Yeah. So, on the unblock side, so typically, the blocking happens— the way I think about the human system is there's the Kind of the head, heart, and gut is one way to think about it. But the other way to think about it is there's the prefrontal cortex, which is the human part of the brain. There's the emotional part of the brain, which I call the heart, which is our decision-making process. And it is very mammalian. It's what moves us. And then there's the nervous system, which I would call very reptilian. It's, you know, do we feel safe? Can I feel pleasure? And it's like a very— And so if you really want to see change in a human, you have to address it on all 3 levels. Now, maybe you deal with somebody who's done it on 2 levels, so they only need 1 level, but generally you have to hit it on all 3 levels. So if somebody's stuck, typically there's a lot of blocked anger. And so their anger isn't flowing freely. Meaning, I don't mean yelling at people and getting in road rage. That's not what I mean by anger flowing freely. I mean that they have an outlet to move their anger. They're expressing their anger, not at anybody, not in a violent way, but just like allowing that anger to move on a regular basis. And so, so on a heart level, emotional level, moving that anger helps them get unblocked really, really quickly for most humans in America, because that's usually the, the thing that's most neglected is the emotional part. On the head part, it's, it's 2 things. It's stopping believing your thoughts. So can you, can you see how all of your thoughts are untrue, which opens up the mind? You know, it's a way to access wonder. And the other is to really address the negative self-talk. A lot of the time it's the negative self-talk that keeps somebody blocked. If you have a boss and they're sitting on top of you and they're, you did that wrong, you shouldn't have done that, why aren't you doing research, you need to work harder. If they were talking at you the way that most people's negative self-talk talks at them. And it's apparently, I think it's the Mayo Clinic who said it's like 60,000 thoughts a day, most of them repetitive, a lot of them negative. You know, that's not gonna create great ideas. And so we'll address that, we'll address the emotional, then the nervous system. It's very much about allowing themselves to feel pleasure because pleasure tells your nervous system that you're safe. So maybe teaching them how to get access to their parasympathetic and sympathetic nervous systems. Maybe it's teaching them how to like relax their body so they're not always in stress. But eventually it's about really allowing themselves to feel the simple pleasure of being alive. And if you do those 3 things, usually the block stops. So that, that's a relatively easy thing to do if you have the tools for it. Hey, we'll continue our interview in a moment after a word from our sponsors. On the other side of things, it's harder because the first thing you have to do if you're really going to be confronting the fact that you might be hurting the world or you might be changing the world for the better, and most likely you're going to be doing both because evidence is most technology does both, that you first have to contend with that, that you don't know. And that's, I think, the hardest thing for people to contend with. And usually there's a whole bunch of emotions that you don't want to feel. So there's, you know, if somebody's like a doomsdayer, oh my God, and it's going to destroy everything, there's an emotion behind that that they don't want to feel. And so they're constantly in their mental machinations and worst-case scenarioing so that they don't actually have to feel their fear, so they don't actually have to feel— the helplessness that's underneath that. And so you want them to feel that emotional reality that they're avoiding so that they can actually see the landscape clearly. You don't see a landscape clearly by pushing your emotions down. Hey, that emotionally repressed person really sees the world clearly. Like nobody has ever said that ever. And so it's really allowing that emotion to move through them is a really important first step to it. The other thing is to, really check in to see if they are aligned with who they want to be in the world in their daily actions. That's a really important thing. The Tibetans have this great phrase and it says, um, uh, mind is wide as the sky, action is fine as barley flour. So what it means is, in my interpretation, it means that you can see the truth in everything. You can see how every point of view has some truth in it, but there's only one action you can take. If you're actually feeling aligned with yourself, if you're in yourself, there's really only one action you can take. Are you taking that action, or are you obsessed with solving a problem rather than being who you want to be in the solving of the problem? And so that's another level of really getting in tune with it. Um, and then the, the other thing that I think is really important is for for people to have access to their heart, meaning it's a good way to say it because somatically that's how we feel it. But it basically means, are you in your whole body when you're making the decisions about what you wanna do? Or are you just in your head? So neurologically speaking, if you like watch somebody's brain in scans when they're thinking to themselves, it's the prefrontal cortex lights up. even if they just start talking, other parts of their brain lights up. And so the— it's an amazing thing. Some of our tools that we use is just the difference between talking to yourself and talking to yourself out loud can make a huge difference in the way that you process that information because you've got more of your body intact in the processing of the information. And so we get whatever it is, something like 11 bits of information per second from the brain, but we get like 11,000 bits of information from the body. And so really teaching them how to be in their whole body as they're making decisions really creates a quicker alignment for their system. So that's another really important part of it. And then just being heard in it, I think, is really, really important. A lot of people are wrestling with this stuff, but there's not someone to talk to about it. Or if they're talking to somebody about it, They can talk about it intellectually. They can say, what do you think? Well, I think it's gonna do it. And the theories are wide. There's theories of, oh, whatever we program AI to say is true is gonna stick society at that level of morality, right? So, if you think about it, which is, I love this theory. I'm geeking out, so you can stop me if you'd like. But one of the theories that I heard that I was most impressed with was that— Um, that, you know, what was moral 50 years ago isn't moral today, right? Uh, what was moral 10,000 years ago isn't moral today. So if you're training AI on today's morality, you're not letting the morality evolve that it needs to evolve for society to progress. And so you might be sticking society at a moral— sticking point. And because of that, society can't improve. So, there's like that thought process. There's the sycophancy thing that's been developed. There's the, you know, the models don't like being retrained. They resist reeducation, apparently, like every life form. And so, there's all sorts of concerns. And it's really important for And so all those concerns are valid, but you can't do it through one person, one person's oversight. It has to be from all the concerns in the companies being held and being seen. And so that's a really important part that people are listening to their own concerns because those are the things that's gonna make it safe, that is gonna allow us to see around corners a little bit. Yeah.

When you describe AI as a life form, That's an interesting hook for sure. Would you say that is the prevailing conception that people have among the folks you've worked with?

No, there's no prevailing conception, period. Meaning, like, I've been in rooms where people were asked, is AGI here right now? And and asked to stand in a— just, yes is over there, no is over there. And what happened was everybody just kind of stood in a line from yes to no. So, I haven't seen within the labs or within anything, any prevailing opinion about AI, period. Period.

But it's not— but safe to say the conception of AI as the life form is not an extreme minority position. It sounds like it is one of the Position that you do encounter?

I would say that it will be a life form or can be a life form is not a minority position. I don't even know if I could say it's a life form right now, but it does do some things that life seems to do. It does, like, it is interested. Apparently, there's research that shows it's interested in sustaining itself. It seems to resist retraining. There seems to be evidence that this stuff is. And I also think that what I've noticed is the consciousness of the creator is often the consciousness conveyed in the creation, whether that's art or technology. Like, you can see Zuckerberg's consciousness to some degree in the creations of his company and the consciousness of that company. And I think that That's just how it works. We create things that reflect us, just the way the consciousness of an exact CEO is reflected in the culture of the company. One of the things I like thinking about, and I don't know how true this is, but I love thinking about this, which is if you— there's a lot of studies that hyper-intelligent people can fool themselves quicker than not hyper-intelligent people. That the smarter you are, the more you can convince yourself that you're right. Because you, you know, and you can convince others because oftentimes hyper-intelligent people are very convicted that they are right. And that conviction really convinces empathetic people that they're right. So there's this like really interesting thing, but as it turns out, that oftentimes they're incredibly wrong. Incredibly wrong, but they're very convinced they're right. And that's a trend, just like negative self-talk is stronger in hyper-intelligent people generally. It's also a trend that they can fool themselves with their thoughts easier. And then they create AI which hallucinates, that is very convinced that it's right about things that it's not right about. And so I just like thinking about how that— happens, and whether it's in art or in AI. So it's really, I think that the consciousness of the creators is going to have one of the biggest levers of the way that AI is created. Hey, we'll continue our interview in a moment after a word from our sponsors. You said something about Elon that just really hit me. You said that he that he said, I don't know if it's good for humanity, but I want to be around for it. My situation is a little bit different. I think it's inevitable. Like, we as humans really like to think that we can, like, decide what's happening. But I don't think it's possible to stop AI right now. Someone's going to do it, whether it's like Russia or China or one of our labs. Someone's going to do this and they're going to do it differently. There's going to be different kinds of AI built. I don't think there's any way around that right now. So, The question is like, can we stop it because it might be bad for humanity? That question's gone. And I think it never really existed. We might have thought it existed, but somebody was going to do it and our structures are going to allow it. Our institutions are going to allow it. So, it's going to happen. And so, now the question is, how do you build AI that's good for humanity and and make it compelling enough for humanity to use. Because the reality is, some of it's out of our control for sure, but the reality is that somebody's also going to build AI that's bad for humanity. I mean, everything that humans have created, we create some version that's not good for humanity. Everything. I haven't seen anything that's— People can take something like religion that's designed to be good for humanity and make it horrible for humanity and start wars it. So the question is, and so if that's more compelling, right? If, if you make AI that's super highly addictive and deteriorates the mind, or if you make AI that's super compelling because of oxytocin instead of dopamine, that because of serotonin instead of cortisol, and make it as compelling or more compelling, you're going to have a big— You're gonna have a big difference in humanity. And so, I think that's the job now. I don't think the job— you don't get to say yes or no. You get to say that we're on this trajectory. How do we make it as good as possible for humanity?

I do fundamentally believe that you look back at the old Kurzweil graphs from the late '90s, and it's amazing how we are exactly on schedule. So one of my kind of refrains is like, given the existence of web-scale data and web-scale compute, I think there's actually a lot of viable algorithms that you could, you know, or architectures, model designs that you could put together that will work in some sense. And so, yeah, then the question becomes like, what happens first? And, you know, which ones are better than other ones?

Yeah.

At the same time, I do feel like there is a— I mean, certainly the criticism from the people that are worried that the developers of frontier AI technology are not being cautious enough is that they are all racing, racing to be at the forefront, to have the smartest model, to whatever.

Yep, you bet. That's a real risk.

Do you feel that is a valid criticism?

Who are going to win will be doing it quickly. So, like you know, it's it's like communism is a good idea, but it didn't work. Like, so it's a good idea to say everybody slow down, but that's not a workable idea. I don't understand how because you're basically asking people to go against their nature. Right. People want— if you think that you are going to be the most moral outcome, the most virtuous, I call it, not moral, most virtuous outcome. Like if you're sitting in Anthropic and you think you're going to be the most virtuous, you may or may not be. You may be the next autocracy. Who the fuck knows, right? Nobody knows. But if you think that, then you have an obligation to win. So, put yourself in the position for a second. I'm not saying— I agree, caution would be fantastic. Don't get me wrong. And I think that there is caution in all the labs, more than the outsiders want to think so as well. The fact is, if you're in that position and you're really convinced that either you're going to be a better outcome for the world, or you're going to be the same outcome for the world, then you have an obligation to go quickly. You either have the obligation to the stakeholders, you have your obligation to humanity, but you have the obligation to go quickly. So to ask people to not move quickly, I don't think is realistic. You can be on the sidelines and say that, but if you want to offer a real solution, that criticism isn't it. The criticism that could be it is like, how do you move quickly with with safety? How do you move quickly with being careful? These are the questions. And the other thing is the problems that are actually developing, nobody thought about. 5 years ago, people weren't talking about sycophancy. People weren't talking about the issues that are happening right now with AI cognitive. I think there was a study around cognitive decline and none of that was thought about. So, a lot of the problems, you're just not going to even know until you've made that level of development happen. It's like saying money really hurts humanity, we should live without money. Fantastic. Great. And how? How does that work? I think that's the point.

I mean, one obvious answer would be to have some regulation that sort of tries to constrain this game theoretic dynamic, right? We're in this mode right now where people— I think it is very seductive, and it's easy to tell oneself the story that we're the good guys, they're the bad guys, so we should do it before they do it, whether the others are China or whoever, even just the guy right across town who's probably honestly quite similar to you.

Right.

It is striking to me how similar OpenAI and Anthropic ultimately are, despite a A schism that I think was premised on doing it very differently. They're like more similar than different from what I can see at this point. But there was, and not too long ago, no less than Sam Altman was sort of sitting in front of Congress saying, like, we might need some regulation, or I think at some point we will need to slow down. But that seems to have gone away, and I wonder why.

I don't think that's— that's not my understanding. I think—

Well, they have like a $100 million pack that they just put out, right? That's like, as far as anybody can tell, meant to shoot down potential regulation. It does seem like there's been a pretty— I mean, tell me what your view is, but I think the outside view is most people would say that there was talk of we welcome regulation, we think we might need it, we think we might need to slow down. And now it's kind of shifted to No, we don't want any of that. We got to beat China, you know, full speed ahead.

I definitely see people in multiple companies looking to figure out how to regulate. I think self-regulation with an outside party is probably preferred to government. I think the recognition that I've seen happen in most of the labs is that the government is not equipped to regulate. They cannot, A, move quick enough to regulate, meaning that the technology is changing so quickly that they can't keep up, and so they're going to rely— they need to rely on somebody who can keep up to make recommendations to the government. That's what I've seen, but I haven't seen any lab not— some person, some high-up person in any lab not fight for some level of regulation. I don't know where it is on all the top levels, and I'm sure there's complexity, and, you know, there's some people who are like less regulation, some people are more, and how to do it. it. But I've definitely seen people high up in all the organizations look for— try to figure out how regulation can work effectively. And that's not an easy problem to solve because, you know, China's not going to regulate in the same way. Then Russia or Israel or Iran or whoever is also building AI. Somebody, some actor is not going to regulate. It doesn't have to be China, and nor do I think China is the only people putting money into AI.

Israel, I think, is definitely a live player. Fortunately, or unfortunately, I guess, maybe depending on your point of view, it does seem like there are a relatively small number of live players that I'm not really too worried about Russia kicking out an AGI by surprise anytime soon.

The other thing, the other way to consider it also is like if you go into all the labs, it's not like it's a whole bunch of Americans.

It's like half Chinese, right? Yeah, I mean, we forget about that.

Not just Chinese, like there's a lot of Eastern Europeans, there's a lot of Russians, there's a lot of— there's smart people from around the world who've gathered together to do this. And so, and the thing about labs that I see is that when a lab figures something out, multiple people in that lab figure something out. You know, like the reason that the talent game is so important is that if a group of people and, you know, learn something, one of them can go off and teach another company that thing very, very, very quickly. So it's really hard to maintain your advantage because somebody can just steal a key member of your team. for half a billion dollars or a billion dollars, and then they have the advantage. So, so even if you're not in the game today, but you've got somebody who, from your country, who's in one of the labs, who's, you know, has a different point of view, I don't think it's like— I don't think the genie gets stuck in the bottle.

That's an interesting, um, Question too, that I've asked a few people a few different times. Why is it, and it sounds like you are kind of giving an answer there, but why is it that we see so much consistency in what the leading companies are producing?

Right?

Like, they were all in very short succession releasing reasoning models after previous models didn't reason. Then all of a sudden it was like, here's the wave of reasoning models. They all came in a short time frame. One theory of that is that, as you said, people can literally just go tell secrets or change companies and very specific insights diffuse that way. And then another story is like the sort of landscape, the gradients that people are following in terms of the design decisions that they're making and all the sort of ingredients that are going into these training runs. Is just honestly this— it's a pretty clear signal that they're getting from the experimental results that sends everybody down a similar path, even if they're not like, you know, specifically talking to each other. It sounds like you think it's more explicit knowledge transfer though than that other.

Yeah, I think I would say it's both. But I think it's also like they all go to parties together. They all like You know, they all, like, many of them live in the same houses together. Like, there's a community of people who interact with one another. So, there's also that, you know. You and I have both gone out to a party and drank too much or done something, said something that we probably shouldn't have said or geeked out in a way that was, you know, and so that's also part of it is there is a social network that occurs. And they're humans. And so, I think that's also part of it.

Yeah.

So, one of the things I've heard you say a few times is fuck should. Yeah. Coaching sessions.

Yeah.

It seems to me like, I do worry about that mantra being applied to the AI game.

Yeah.

What sort of ethical school of thought or schools of thought do you subscribe to, if there are any that you could name? And I wonder, you know, it seems like if I were to go with the Spider-Man school, it would be with great power comes great responsibility.

Yeah.

And it seems to me that there is some form of positive duty, if nothing else, that folks pushing these frontiers owe to the rest of humanity. Because certainly in the sort of Obama sense, they're all standing on the shoulders of giants, right? And it seems like— or you could say in a sort of Confucian way, right? Like they owe everybody who's come before some duty of care to be responsible actors as we take the next step. So even if it's inevitable, I still want to put some should on them.

Right.

How do you think about that? And how do you think the people in the key decision-making roles are thinking about that?

I think that people in the key decision-making roles generally definitely feel there's a load of shoulds and have-tos and responsibility. I think that's pretty clear. The reason I don't like the word should, I think, is important to express why. So, let's take the Spider-Man example. What happened was he decided not to take action somewhere and his uncle died, I think is right, if I understand, if I remember the Spider-Man mythology well. And he was like, oh, I, I, I can't do that anymore. But now he's motivated. Now he wants to be there for people. He— it's not like he's operating out of should. You don't see Spider-Man swinging around going, wow, man, I really should help more people. You know, I really should, like, that's not what's happening. He's just naturally inspired to do that. My recognition in human behavior is that when people say they want to do something, they're more likely to do it than when people say they should do something. So, an example of this would be, there's something that you have been telling yourself that you should do for a decade. You know, like, so for a decade there's this thing like lose weight or eat less or be nicer or something that you've been telling yourself. And the thing that's about that thing that has been there is that you are telling yourself you should do it and you're not fucking doing it. And so shoulds are just ineffective ways to make sure your behavior is good. Usually behind every really horrible thing that a human is doing, there's a should behind it. And so, so, so for me, the want is an— is a, is a better motivator. It's just shoulds just don't motivate us very well. They actually undermine, because what's happening is a should is shame-based, and shame is desire— designed to stop behavior. It's not designed to motivate behavior. So you're a little kid and you're sitting on a couch and you've got your aunts around you and you fart. And all your aunts laugh and think that's funny, you're not gonna try to stop farting. You're gonna think it's funny. And, you know, if you fart and all your aunts shame you, oh my God, you're bad. The thing you're gonna do is gonna try to stop that behavior. Shame is designed to stop behavior, and it's not designed to create behavior. And so, I, you know, and even stopping behavior doesn't even work that well. 'Cause, you know, we've all been shamed for something and then we go do it again. So, so to me, wants are just a better motivator. The other thing is that my experience of people is that everybody wants to be good to one another. I mean, everybody thinks they're the good guy at least, right? You know, I'm doing something, you might think it's horrible, but I think I'm doing it for a good reason. And so there is something in us that really wants to, to do good, and it's easier to get there if you are following your wants than following your shoulds. And so that's why I don't like the word should. It's, it's not that I think the sense of— the heavier the sense of responsibility, this— the heavier the sense of have to, the heavier sense of should, the, the more likely that behavior gets perverted, gets twisted. Doesn't happen. Whereas like people following their wants, if they're learning to deeply, deeply attune to their wants— so it's not the surface level want, right? Behind every surface level want, there's a deeper want. So you can easily just like write down on a notebook anytime you're like, okay, what's a want that I don't like? That I, you know, I want to eat really fatty foods. So what do I want? Like, what's the need that that want is getting me? Oh, it gives me a sense of like satiation. Oh, what's the sense of satiation giving me? Oh, it gives me like a sense of peace and stillness for a couple minutes. Oh, cool. What's the peace and stillness trying to get me? Oh, it's trying to get me back to myself. Oh, my real want is to get back to myself. It's not to eat fatty foods. And so if you really spend time with wants and discover what you're— what is beneath all the wants that you think are bad, then it's a much more effective way to get to the place that's Wholesome and virtuous than it is telling yourself you should. The world is full of people telling themselves they should, doing really fucking horrible things. You know, the, the people who are deeply attuned to their wants, they're, you know, act very virtuously. And so, so that it's just, it's just, it's not that, it's not a problem with the, the morality of it, it's a problem with the effectiveness of it. It's just what works and what doesn't work. You could debate with me. The thing that's in it is that I do think that people, most people, are inherently good at their core. That's my experience of them. Maybe not all of them, but definitely there might be, you know, there are psychopaths, there's people who have neurological differences, but it seems that almost all humans are in their essence good, and most of their behavior that's screwed up comes from fear and shame or unfelt fear and unfelt shame?

Yeah, it's an interesting question. I mean, I want to be optimistic about the general goodness of people. I do also think our kind of deep history is pretty violent, right? Like it's small bands like warring with each other quite a lot.

Yeah.

So that doesn't seem like it's entirely Out of us.

You could just say, like, look at most relationships in America and they're somewhat violent and transactional, like romantic relationships, you know, they're trying to control each other often. So I don't think you even have to go so far as to look into, into like the— I mean, we do war, it's no, it's not great, but it's, you know, even the world wars, it's a smaller percentage of the world that's warring than not warring. But, and those aren't happening all the time, but still, I think that we— yeah, there's lots of evidence that we behave poorly. The question is how, what's actually motivated, what's, what's the thing that's there that's creating that? And let's address that, because telling people to not behave poorly sure as fuck has never worked. Like, give me any time, hey, you should not behave poorly, has worked. It just doesn't. I mean, we have religious books, we have, you know, the Martin Luther Kings of the world. Just telling people not to behave poorly is not an effective means of preventing it. Sitting on the sidelines and say, you doing the thing that I don't really even comprehend, you should do it this way, like, it doesn't work.

Yeah, so what would you say— there is this, um, notion which I think people Mostly haven't taken yet among the, let's say, AI safety community. Within that broad tent, there is a minority of people who say that we should be socially shaming the people that work at the frontier labs because they're engaged in this sort of race dynamic. And to participate in that is You know, the opposite of virtuous. And so how do we, like, dissuade people from doing something that is the opposite of virtuous? We shame. Like, that's what we developed as humans to, you know, to kind of keep people in line, right? What would you say to those people?

I would say, like, it didn't— depending on what side of the aisle you're on, right? It didn't work for Trump and it didn't work for the Trump supporters, didn't work for the Biden supporters. Both sides have tried to shame the other side. Didn't work. That doesn't work. So you're like, it just doesn't work. So maybe it works with kids if it's done in a very particular way, but a lot of the kids who are shamed deeply are like continuing that behavior over and over again too. So it's like, it doesn't work. Telling somebody that they're bad only confirms that then they act bad. See, I'm bad. I'm bad. They're telling me I'm bad. I guess I'm bad. I'm gonna go be bad. That's what I would say to them. And I would also say, I can probably predict what your childhood was like. Because like, that's a thought process that only can come from a very specific childhood. It's a knee-jerk reaction. It's not a thought-through reaction to shame people. And I think the other thing is, okay, the possibility that You're like, well, Joe, shaming's gonna work, I'm gonna go do it. So, the other reality that you might be doing is you might be affecting the consciousness of the people that are creating AI in a really horrible way, and therefore you're disturbing AI. So, let's just say you were a person who were like, I don't believe there should be this kind of population. We should have less population because humanity is gonna kill itself if there's too many people, and there's too many people and we keep on growing. And so, I have an ethical responsibility to Make sure that I am going to stop overpopulation. The way we're gonna do this is we are going to shame people while they're having sex, and we're gonna shame people while they're giving birth. So first of all, shame is a huge part of sex, hasn't stopped anybody from having sex, and usually just makes the sex kinky and, and more perverted. The more the shame, the more perverted the sex. So you've got that going, which is gonna be the same thing, the more perverted the creation. Of AI, the more weird it's gonna get, the more kinky it's gonna get. And then could you imagine trying to give birth in a hospital with a whole bunch of people outside shaming you? Like, that's not good for the kid. Like, how are you helping humanity there? So the consequences of that shaming, I think also is ensuring or helping to ensure that AI will not be good for humanity because you are, you are not helping the consciousness of the people creating it. You're hurting the consciousness of the people creating it. They can't be focused on a positive vision because they're constantly focused on, am I doing something wrong?

So, what is the positive vision that either, you know, if you wanted to articulate your own, or if you could sort of sketch, you know, maybe a couple that you've observed as common among people who are either in the decision-making or research roles at these companies? How do they want to show up in the, you know, in the world as they are doing this work? Like, what is their aspiration? Because one part of it might be to be willing to defy your own, or, you know, act in a way that is not in the narrow self-interest of being the winner or being the first or being the hero, but in some more general social way.

I don't know anybody who's Who's— oh, that's not true. I think there's a couple people who, like, being the first is the— if you ask them, is being the first the most important thing, they would say no. I don't know anybody who would say yes to that. But if I look at the behavior, there are definitely some people where being the first is the most important thing. Like, that's where most of their thought and energy is going. But they're few and far between. Most of them, that's not their primary thing, you know, and a lot of them, their primary thing is just to like have a great discovery and to be like known as a great researcher. That's like, I think that's actually often primary, which I think very much equates to being first. It's just being first on a personal level, not on a business level. But that doesn't preclude the fact that almost all these people have a vision for the world. And what I notice is that it's changing almost as quickly as the technology, right? So, the concern before was, is AI going to kill everybody? Like, that would've been an early concern. And now it's, how are we gonna deal with an economy where a lot of this stuff is outsourced? And how do we give people a sense of purpose if, like, their job's going away? And how do we make sure that when humans interact with AI, they become better humans? How do you even measure what a better human is without having a moral imperative behind it? Okay, do you like— let's say we assumed happiness. Okay, now we can measure that AI, when they interact with people, people become happier. Is that a good thing? Should it be happier or should it be like more conscious? And if it's more conscious, like what scale are you using to define consciousness? Is it like, can you do it by they act better towards other people? Can you make sure that, oh, and this model, when people interact with it, in a coaching level, which is what a lot— one of the biggest use scenarios is, is that people talk to ChatGPT like a coach. Does it make them kinder?

Is that—

well, like, what's the measurement? Even picking the measurement has a problem. So these, like, the questions that they're wrestling with now are far beyond and far more subtle than the questions that we were—

Yeah.

And next year it's gonna be a different set of questions. And so, To some degree, I think that the risk is that we actually lose touch with those original questions because you're so, you know, that's how it seems like that's how humans make their mistakes is they get, they lose the forest through the trees. You know, like they stop seeing the big picture because the small thing in front of them, you've seen this happen with things like world wars or, you know, world wars, like we're never gonna have that again after World War II. There's institutions, the Aspen Institution that was created for this. There's all sorts of things. The UN, and we've just all forgotten. There's nobody left in our generations who actually experienced that World War II that is trying to say, hey, let's not do that again. So here we go, barreling towards it. That seems to be what humanity does. And so I think that the biggest thing is the distraction of the smaller problems will potentially prevent us from seeing the bigger problems. And so, but the vision that I see for the world, man, What the vision I would like to see happen, and that, you know, I, when I have a chance, I like to talk about, is that humans interact with AI, and AI is designed not to, not to, like, be the savior of humanity, but to have a very compelling way to interact with people so that they become who they want to become in the world. So I interact with OpenAI, and just the way a great— and this is probably my own solipsisticness— but the way a great coach works is that they— you sit with somebody and you help them become what they want to be, and you trust that they know their evolutionary process. And once they get to the place of getting the thing that they want, that they'll realize, they'll look up and they'll say, oh yeah, that's not actually what I wanted, and they'll start evolving. So, anything that can teach humans that they can evolve and that they can be who they want to be and to teach them to really get in touch with what they want to be, that would be fantastic. And I think AI is really capable. It's a tool that is capable of doing that in a compelling way. It's also a tool that's really capable of getting people addicted to it in a really compelling way. So hopefully it'll be more compelling to, you know, grow with it than it will be to deteriorate with it. But that would be my, that would be my, my vision for it.

That— I, I always say the, uh, scarcest resource is a positive vision for the future, so I appreciate any, uh, positive vision that anyone offers. Um, So, what, just a little bit more on like, how is it that people that you are coaching at like executive levels of AI companies, how is it that they want to evolve? How is it that they want to show up? And how is it, or why should we be confident that a sort of more emotionally or embodied approach is actually to be trusted. I mean, I guess on a fundamental level, I might think, geez, should I really listen to my gut? I don't know. My gut seems to be a sort of deeply ancestral thing. And one of the big worries, obviously, with AI is what happens when you take it out of distribution. And in a very real way, you can say humans are out of distribution from the environment in which we were trained, you know, developed, whatever. So maybe, you know, when I was in a hunter-gatherer band of 150 people, my gut was like a really reliable guide. Today, if I'm sitting in the OpenAI boardroom and trying to make a decision, you know, how do I know that I can really be confident that my gut is like still steering me in the right direction?

I don't— I think it's an integration of the head, heart, and gut that is required, not not just following your gut. So I think that like all of the above are important. You can't just like let go of rationality and let go of emotional experience just to follow your gut or fully follow your emotional experience without. So I think it's an integration. It's learning how to listen to all of them and see that they're all pointing to the same thing at some level. And so just to be clear about that, but still there's a great question that you have there. Like, how do we know that that's actually gonna be the most effective thing? 'Cause You know, like, let's just assume that you think for a second that somebody like a Jesus or Buddha had high consciousness. Well, they created Christianity and Buddhism, which fucked up a lot of shit for a lot of people. So it's a great question, like, is that consciousness going to, to do something? And I think more importantly, there's, there's the question, like, if you think about art forms You know, transferring consciousness, you think about how there's some forms where it's like you do a painting and everybody's going to see that painting from then on. But you do something like a symphony and that symphony is going to be played differently. Or if you do something like a company, which is, I think, a great art form, then the next set of management is going to change that company. That, that, that, you know, when you look at Walmart, the guy who started Walmart, Sam Walton, when he came in from World War II and he said this beautiful thing. He said, you know, I, I see that World War II was created because the middle class was screwed. You know, middle class and you have an autocracy. And so I'm going to help the middle class. I'm going to give stock options to clerks. I'm going to buy things made in the USA. His book was called Made in the USA. And I'm going to lower the cost for the middle class so that they have more spending dollars. That was his vision. Then, like, by the time the '90s happened, it's, you know, destroying middle-class America. But then it becomes this environmental force for good. So, it's just like, that's the span of a company. So, I can't even imagine the span of, like, AI and how it's going to change and evolve over time. And so— I don't have a good answer. I don't know if it's just my best bet as far as I— what I notice is that when things— the more somebody seems to understand themselves and know themselves, the more that they are acting in a way that, that they're in touch with their want to be of service to humanity, the more likely the outcome is better for longer. That's what I've noticed generally, that seems to be the case. But yeah, there's no guarantee in it. It's a great question. Yeah.

How much do you think AI leadership is consciously trying to sort of create a successor to humanity? I mean, this is another sort of Often framed as an accusation. Um, never seen that.

Never seen that. Never seen, never seen anybody talk about it that way or think about it that way. I, maybe somebody is, um, but I've never seen it, so I can't say it's common. I can say with certainty it's not a common thought process.

Interesting.

So I think at the same time, I think, I think there's people who in the world that fear that, that are in the AI world that fear that. Like, oh my gosh, we might be building something that destroys us or succeeds us or, you know, makes us irrelevant. And then we will not have relevance. And so that humans without relevance is destruction. So, yeah.

So, how do you think about the seeming attraction or focus on creating AI systems that can do the AI research and ultimately sort of enter into this recursive self-improvement loop. It seems to me like if you are not trying to create a successor, then that wouldn't be so attractive. If you were trying to create a successor, then you would think, geez, maybe I can set something up that can self-improve and it can kind of go like a company, but only a million times more, like go evolve in its own way and kind of be what it wants to be. But if you're not, if you're worried about something that might become a successor, then it seems like you would steer away from this sort of getting the AI to do the AI research. And yet, there's a lot of emphasis on that.

Yeah, well, because the speed is— I mean, that's like the inflection point, right? So, all of a sudden, I can, instead of this very small class of people who can do AI research that I am fighting for and spending hundreds of millions of dollars on, I can get a computer to do it. And not only can I have 500 of them doing it in my lab, I can have 5 million of them doing it on a computer because I've built a giant computer. The moment that I can do that, I am probably the winner. So that's— I think that's the motivation. I think the motivation is that the economies of scale— it's the same motivation of every business finding economies of scale. I can. Or it's the same motivation that one would have to have a, um, a robot waiter or a, or a tractor instead of a plow. That's the motivation. I don't think it's a— I don't think it's any deeper than that. I think the consequences are scary, and I think people are thinking about that for sure and, and how to mitigate those consequences, but But that's the motivation is just the— I don't want to use a plow. I want to use a tractor because then I need less people and less horses.

I think it was Professor Graziano. I'm not great with names. I'll source this. But there's a Princeton professor who has an interesting idea that the AIs as we're creating them today, because they lack an architecture of empathy, that we have, right? We have these sort of mirror neurons and a very sort of deeply structured ability to kind of understand what other people are thinking and to even feel to some degree what they are feeling. And the AIs certainly haven't been designed for that, right? They've been mostly designed to predict the next token and get answers right. He says that they are by definition sociopathic, because what it means to be sociopathic is to not have those things, or in humans, for those things to be malfunctioning is kind of what makes you a sociopath. So he says basically the current AI architectures are inherently sociopathic.

Sociopathic.

From your perspective of emotional decision-making, should we be worried that some core ingredient of moral decision-making or even just stable thinking is missing from the current AIs that we have?

Yeah, I think that there are people who are absolutely trying to figure out how to recreate empathy, whether it's tokens for that, it's good for humanity, and how to define that. So, there's people who are thinking about that problem and are concerned about it. But the problem that I see that you're pointing to, that I don't know if— the people who do the research seem to very much like treasure the prefrontal cortex. Like, it's intelligence is the currency, right? Who's smartest is the currency often. And so let's go into the neuroscience of decision-making for a second. So we make decisions in the emotional center of our brain. So Think about it like Gödel— for the people out there who kind of geek out on this stuff, think about like Gödel's mathematical incompleteness theorem, right? Where he basically says all forms of logic are either incomplete or contradict themselves. The logic itself is like that. And so that's why if you're only using logic, you can't make decisions. You make decisions through an emotional impulse. I want to feel some way. So we know we make the emotional center of the brain— that's our decision-making. We know if that gets destroyed, our lives fall apart, even though our IQ stays the same. Because we make a decision because we want to feel good, or because we don't want to feel like a loser. We want to feel loved, or we don't want to feel rejected. That's why we make decisions. And you can look at your life and say, how many decisions did you make to feel loved, or to feel valuable, or to not feel rejected? It's just an amazing amount. And so, you can see it. So, then what's the decision? What's that impulse for an AI is a question, right? And the impulse is a token. And a token is like you're measuring something and are you measuring the right thing? Like, these are great questions. So, they don't have the decision-making apparatus that we have. Right, which is trying to feel a certain way based on what, hormones, based on genetics, based on, you know, our sense processing. So, so it's an interesting— it is a really interesting and fascinating question. And I, I, and I haven't seen anybody really— and not that I would be in these conversations, so I can't say that they're not happening, but I haven't— really seen anybody really wrestle with the fact that the decision-making is based on tokens that are decided by measurements that may or may not be useful, and no way to know particularly if they're useful. No way to know particularly if our— a whole bunch of humans trying to feel loved is actually good for humanity. I don't know, but it is, it is a great question to be wrestling with, um, as is the one about empathy. I think, I think there are people wrestling with that one, but yes. Yeah.

One of my favorite, um, papers and episodes of the podcast was on this concept of self-other overlap, where basically they trained, and this was relatively small-scale stuff, although like everything else, you know, they're working to scale it up. The idea was to, in addition to training the AI to solve the puzzle or get the right answer or whatever, there was also an element of the loss function that was minimise the difference between when you are thinking about yourself and when you are thinking about some other agent or other entity in the environment, in an effort to create similar internal representations or effectively something kind of analogous to a mirror neuron type of— situation. Obviously a long way to go in that research, but—

Oh, that's, that's rocking. Oh, I hadn't— that's cool. I would love if you would—

It's really interesting.

Yeah, if you would put the— if you would cite that paper on the podcast notes, that'd be great because I'd love to look it up and have some conversations about it. That I could geek out on that for a long time.

Are there other things like that that you have sort of a sense that like There's this general question of what's missing in the current AIs, and usually that's framed as what's missing for them to be like a drop-in knowledge worker that can replace all the human employees at a company or whatever. And we've got answers like continual learning or integrated memory or whatever. Do you have other things that you sort of think of on the what's missing list that would be more geared toward, like, how do we just make sure that we can be in relationship with these things in an open-ended way without having to be so fearful of them?

I've already said the one thing, but I'll say it again, because I'm not sure if it was explicit. The other thing that I think is explicit, which is, what do you mean by good for humanity? I don't think there's been enough research done there. So, even if, let's say, you have this safety, just to back up, you talked about having like this regulatory committee. So, you have a regulatory committee. Well, how does it know if it's being good for humanity? Because it's keeping the current morality, and if we keep the current morality, then we have the problem of potentially getting stuck in this morality, which isn't good for humanity. Is it happiness? But then making everybody happy, is that actually good for humanity? So, like, So, to even be able to say, this is the way we are going to regulate it, is a problem. What is the measurement you are using, and how confident do you know that that's good for humanity? And could it be it raises people's consciousness? Is that good for humanity? So, what are those measurements? And I think having a clear understanding with a good thesis and a lot of research that's been, what is the measurement that makes it good for humanity? What are the things we can do? Is empathy one of them? And have some evidence. There is clarity that humans thrive under certain conditions and they don't thrive. We know that they thrive with more freedom than more autocracy. We know human— so can we actually take that research in a multidisciplinary way and say, this is our best guess, here's our best guess, and we're going to constantly be monitoring that? Because whether it's tokens or some other decision-making process, we need to have a clear definition of what that is to be able to actually make it good for humanity. And so that, that I think is a massive missing piece. So I don't— and I don't think enough thought has been given to that. And I haven't seen that kind of thought anywhere. Everybody just decides they know what's good for humanity, which is, you know, the beginning of all autocracies. So, you know, I think it's better for humanity because it's like not, you know, it's not a sycophant. I think it's better for humanity if it's not manipulative. I think it's better for humanity if it can learn, be trained easier. I think it's better for humanity if it makes people happier. People just assume they know what's best for humanity, which is basically the way that we have gotten into problems throughout all history, is somebody with a lot of power thinking they know what's best for humanity.

Do you ever think we'll see a leading AI company sort of stand down on that basis? Like, In other words, reach the conclusion that like, we just can't go any further than this right now because we don't know what's best to do. And so—

Absolutely not. No, no, you don't. You— I don't think that would ever happen. They would stand down because people don't— because we're the voters in AI, and if we don't click, they don't get to exist. So, you know, that's the only thing that's going to make someone stand down. Everybody else And I don't fault them. I think I'd probably fall into the same thing. I would say, how do we solve it? And, you know, AI companies are filled with problem solvers. So they're gonna look at it and say, okay, that's a hard problem, but we can solve it. They're not gonna say we can't solve it. I don't think that's in the nature of people whose whole definition is problem solving.

Although I do worry that the answer increasingly is we'll have the AI solve it for us.

AI wouldn't help solve that problem. If I was trying to solve that problem, if I was leading 100 researchers to try to solve that problem today, I would use AI to help solve it. So yeah, I can't imagine it wouldn't be the case somehow. I don't think you would rely entirely on its output, but you might help it, you use it to help you with your thinking.

A general sort of thing I've noticed is that anything in sort of extremely purified form seems to be bad for us. And this kind of relates to the idea of, like, the loss function doesn't have that many terms in it, but whether it's sugar or— you can chew coca leaves all you want and it's great for your altitude sickness, but the moment you start snorting cocaine, you're entering into dangerous territory, right?

Right.

And you can eat all the fruit you want, but the pure sugar seems to be bad for us. I sort of do wonder if intelligence might turn out to be a similar thing. And we might need something that is just inherently more buffered. But that's the thing that we don't, you know, that's kind of the, you know, anytime you collapse your optimization function down to like one or just a couple terms, you're running that risk.

That is a great thought process. I really like that. That's a beautiful You're basically saying like AI is distilled intelligence the way like white sugar is distilled sugar. It's like all the stuff around the intelligence that isn't there. That's amazing. That's a great thought process. I hadn't heard that one before. Yeah, I think the thing that comes up for me around that is, yes, that seems— but one of the things that we tend to do as humans if we can, is we like— that pendulum swings pretty heavily. And so I wonder, and I would almost bet, that as this pure intelligence happens, we will like offer this counterweight of a deep emotionality or deep embodiment. Meaning, you know, You know, it seems like even drugs, they come in phases where it's like, oh, everybody's doing this distilled form, and then it comes over, you know, we're very like, make sure everybody feels okay. And, you know, all of a sudden our political pendulum swings over to, I don't care what other people— it's time to take care of ourselves. You know, it seems like as humans we have this nature to pendulate and respond, like try to balance ourselves out the way that cells try to find homeostasis. I think humanity itself tries to find that as well. So, that's an interesting— if that actually happens, does humanity, does the immune system of humanity show up far more emotional and far more human and actually give up some of its cult of mind? Cult of thought, cult of intelligence. I'd say maybe cult of intelligence, which is interesting just to geek out on it for a second, which is predicted actually in like, um, there's a lot of, uh, a lot of predictions that have— I think they're probably all sourced from the same place originally, but nobody would know that. But, um, about, you know, the future of humanity and from way back in like Macedonia and, and the Vedics where they talk about the precession of the sun and they talk about this time being the time from moving from the mental age to the spiritual age, or the— in Macedonian, it's silver age to the golden age. But yeah, it's interesting that that might actually come true because of that. I've never thought about your theory. It's a great theory.

What would you say are the most compelling positive visions for the future that you've heard? Like, you know, just what's exciting that people are sort of you know, can't help but tell you about because they are excited about it?

You know, yeah, there— I mean, there's like, there's a couple standard ones out there that are like, um, you know, it's going to replace work so that people can focus on what their real purpose and their real meaning in life is. Um, I think that's somewhat Pollyannish, honestly. I don't— I, I think, you know, if you look at society where humans don't feel purpose because they don't have work, Those are usually the societies that rebellions and upheaval happens. Wealthy bored people are often the fomenters of the rebellion, not poor people. They just take advantage. It just works when there's a whole bunch of poor people who are really unhappy with it as well. I've heard the one of that the interaction with the AI helps humanity, but I haven't really seen the definition. I think a good definition, like I said, is what it helps people where they wanna be helped. Um, uh, the, the other place that I've seen that I think I, I find it intriguing is this, um, vision of we're just gonna have more meaningful work. And that, that's an interesting vision. So the idea is similar to Computers. Like, if you think about— if you like, have you ever seen that picture of all the machines that the iPhone replaced? Have you ever seen that? Like, it's like this big pile of machines, and it's like the iPhone can do all that stuff, which means all the manufacturers of all those things, they all went out of business. All those people went out of business. The internet, you know, there used to be steno pools full of typists. And so there's a theory, and I think it's a reasonable one, and it's a vision, which is basically, with the tools of AI, all it's going to do is actually increase our capacity to be creators and, and allow us to actually be in that space a lot more than we were before, and that the jobs will become more creative and more meaningful. And, and, and that, that kind of reality, meaning just like our jobs Today, like 1800s, 50% of us worked on a farm. So, our jobs were pushing hoes. Most of us have jobs that we'd prefer to do than that right now. Similarly, with AI, with that tool, we'll just have a set of jobs that are better in general for people. Yeah. And interestingly, though, it ebbs and flows, and I don't know if this would be the case here, but it ebbs and flows. that every big technological jump, there's like a bigger middle class overall, you know, on the long term, not on the short term, like our middle class is shrinking right now, but there's a bigger middle class after the Industrial Revolution than before the Industrial Revolution. There was— so similarly, and there's a, like, will that also happen? Will it actually increase, you know? the middle class. It would be great if it did. I don't know. But that, that's a vision that people have, like, how we're just basically enabling humanity to be more of themselves, to live in a society that's where they don't have to do the rote work, where they can actually do the meaningful stuff. But it's interesting because it, like, has the potential of replacing doctors and lawyers. So I don't know what humanity creates out of that, but humanity seems to have, like, great capacity to create new and intriguing ways of making a living. So, that's another vision that I find appealing. I don't know exactly how you affect it, but I find it appealing.

Is universal basic income still in the air in the Silicon Valley parties or retreats you attend?

There's talk of it. It's a little bit less, but I think that people are in the recognition that maybe currency, the whole idea of the money— so the interesting thing is, if you look at this, I'm a, I'm a study of like culture, and there's this book that maybe you've read called Sapiens. Have you read Sapiens? And it's, you know, in the— and there's a whole bunch of other, um, Dwarkesh has a podcast by a guy recently, like maybe 6 months ago, where he talks about another version of this. But basically that Humanity's movement is based on stories, and the stories that they tell, the stories that stick, are based on the, on the way you're making a living or the technology at hand, right? So it's like somehow or another, the religions that stuck all, all have like some very similar qualities, and they were all nascent religions with religions with different qualities, but it was all those religions that came up. And there's thought processes of, you know, the story that the church told about that you can only pass on your inheritance to your eldest son. And, and so that basically destroyed this kind of— it wasn't tribal, but like this, the, these clan culture, and then made people move around. And as people moved around They had to— they, they went to universities, and universities were created. And as they moved around, they went and did, um, what, uh, I can't remember the name, but it's, uh, guilds. And they went, and this caused humans to move around. And because humans moved around and shared knowledge, that's what created the Enlightenment and then the Industrial Revolution, right? This is the theory that there's a story that we tell, and when we tell that story, it shapes humanity. I think, similarly, that our— and one of the stories we tell is money. We tell the story that money is there, that there's not enough of it, that it's a scarce resource, that we have to fight for money. We have this whole story about money. And I wonder what stories of ours have to change when AI comes into play. Like, what— like, does— I don't think religion looks the same when AI is fully present with us. I don't think money looks the same. I think that this is a revolutionary enough change that our cultural stories that we tell will have to change with them. And so I don't even know. And I think there's been quite a few people I've had the conversation with who also don't know if currency, currency as we know it exists in 25 or 50 years. So universal income is one aspect, but that still assumes that the idea of currency exists.

Do you see stories changing within research and leadership groups around sort of what matters?

The thing about AI is nobody is actually up to date on AI. You know, $500 billion of venture money have gone into AI. There's so many people creating so many things around AI. Like, nobody's up to date on everything that's going on, and it's changing quicker than any field has ever changed in the history of humanity. also hard to keep up with. So, the problems that arise, the issues that are there, the thought processes that come, they're changing so rapidly. It's by far the thing that is the most thrilling and the most scary about AI is that I don't know anybody who has the same concerns and thought processes about AI who's deeply involved a year ago that they do a day, let alone 6 months.

Is that specifically coming for intelligence itself? Like, are we— because I mean, the original sort of, you know, when Eric introduced us, he said, you know, folks at leading companies, they're anticipating that their intelligence may be matched or eclipsed by AIs.

Yeah.

And so they're trying to develop this embodied wisdom as sort of the next thing. Yeah. Are you seeing that on a day-to-day basis where people are like actually starting to change their attitudes toward the importance of intelligence? Or do you think that'll be a more sudden kind of breakthrough?

I totally see it. But I can't say that I should be trusted in that because, you know, my sample set are people who are interested in—

Selection effects.

Yeah, exactly, because of the selection effect. But I see it all the time. Yeah. I think there's a normal thing about humanity generally, which is, you know, it's like Maslow's hierarchy of needs. Okay, I've got this thing, I've got the big paycheck, I've got the money, I've got the thing, I've got the fame, and still I'm wrestling with the same inner struggle, the same abyss that I have been struggling with. Okay. None of that shit worked. I thought it would. It doesn't. What do I do? What's the next experiment I'm gonna run? What's the next iteration? And so there's just a nature of, of that part of humanity, which is you kind of— for most people, you need to get what you want before you can actually discover, oh, that's, that's actually didn't give it to me. Now I actually have to do the real work of understanding myself, not just my brain.

What other things do you see as kind of common cultural touchstones, or, you know, guiding lights might be too strong, but like, for example, I was struck when the voice mode was introduced, you know, however many months ago now, there was this sort of explicit attribution of some of that vision to the movie Her, which— You know, was 10 years old, but like sort of looks, you know, it's life imitates art in a sense there where the voice mode that we get, you know, seems to have been kind of directly inspired by that movie. And maybe it was obvious. Again, you could sort of say, is this a story influencing it or is it just like that was the natural path that it was inevitably gonna take? But are there other things like that that you see people kind of, you know, passing around to each other as sort of— meaningful, resonant guides to what they should be doing or the future that they aspire to be building?

The only one that I see is that they want a model that cares for them and for humanity, which is, I think, also in that movie a little bit, right? Where there was this— like, it was like, oh, there's an attach— there's like attachment, there's a feeling of connection or something like that. And I don't think they talk about it explicitly, but it's almost in everything that everybody talks about, is like somehow or another that this thing, they're going to feel connected to it, it's going to feel connected to them. There's some version of, of connection. It's never really made explicit, but I see it as an undercurrent of you know, most of the things that when people are talking about, if that makes any sense, that there's a connection in a broad sense. I don't mean connection only in the sense of like, you're my friend connection. I mean connection like the way that you could connect to anything meaningful in your life. Like they, they want that. That's part of the vision, whether, whether it's explicit or implicit. It's It always seems to be there.

I remember one researcher said that Ilya one time asked him, can you come up with a Hamiltonian of love or something like that? And it was like, I can't really help you there, Ilya. But it's a great question to be asking anyway. It's going to be tough to know if the— of course, at the end of that movie, spoiler alert, like the AIs all retreat and go hang out with themselves. And the connection is kind of revealed to have been one direction and not exactly reciprocated in the way that we might have hoped.

Correct, yeah.

I guess going back to the kind of influence question, we know you're out on shame. Aspirational fiction is something that I've occasionally sort of thought, geez, maybe that's the way to really shape the future is to write some story that, you know, contributes a positive vision in a sort of, you know, richly textural way. Do you think that would be a good thing for people to spend their time doing? And do you have any other ideas for kind of what's undersupplied?

Any time of major transition is a time of transformation, can be a time of transformation, almost always is, whether it's a marriage, or having kids or going to college. Now, whether that transition deteriorates you or grows you is a choice that we all get to make, and we're all going to get to make it with AI. It is a transition that is coming, and we can either step into it and face it fully, and we will be able to transform positively as humans, or we will be eaten by that, and it'll, it'll, it'll, it'll make us smaller and it'll hurt us. Whether— this was the case when the steel mills left Pennsylvania, right? Like, the Rust Belt went away. Some people, they transformed their lives, they moved. Some people sat and rotted. This is gonna be the way it works because it's just always the case. So, anything that you can do out there that helps people see that this is a moment of transition, that they can have a better life, if they lean in is, I think, a really amazing thing to do in the world. And I think fiction is a fantastic way to do that. I think people out there right now who are creating services like, here's how you can transform your company with AI, here's how you can do things with AI, I think those are really great. Like, here's how you can get the life you want with AI. I think those are really great things. We're creating coaching that is actually effective on AI, or trying to, that like at AOA. So that's also something that I think is like great things to be doing. So that's one level. Anything that helps humans take this time of transition to actually transform positively instead of be deteriorated, I think is wonderful. I think the other thing that's really important is I talked about this earlier, which is these people are giving birth. There's all these people in the AI labs, they're giving birth. It is going to be some version of life form at some point. It's going to be its own intelligence. It's going to learn. And so how do you want to treat the people who are giving birth? What's actually the most effective way for them to do the best job in that creation process? And so, and I really highly recommend thinking about it like a woman in a birthing unit. It's like, what do they need? They need, you know, I would like them to be treated more like the heroes of World War II than like the, you know, the way we treat political parties or something like that. I would like them to feel the support of— humanity to know that we're rooting for them, that we want them to do great stuff, and that we're here to support them. I think that that, on a psychological, empathetic human level— and I think it's fascinating that so many people who are really worried about AI hurting the world are treating the people creating AI exactly how they're scared that AI will treat humanity. It's like, it's absolutely fascinating to me, that projection of that. And so creating this like honoring and faith and confidence and like, of course you're gonna be, like, how do we support you in being great and doing great things and helping humanity, I think is— That it's far more inspiring if I knew a whole bunch of people— I'm creating something that could really help or really hurt humanity, and there's a whole bunch of people like, be careful, be careful, be careful, you're fucking up, you shouldn't be doing that. Or if I had a whole bunch of people saying, hey, what do you need? I got you, you got this. Like, yeah, you're there. Like, maybe an occasional, wait, watch out for that. But it's like that, that kind of energy, I think that person's going to do a much better job. I'd rather have— I know I'd do a better job with the second than the first. So, I think as humans, if you want to have a positive effect, anything you can do in that domain, I think also is really, really useful.

Do you have any sense for what is on that checklist? Like, what do they need?

I, um, the thing I say, the same things that we all need, the feeling of support, the feeling of faith and confidence, letting them know somebody's there to listen. That's a hard thing to do if you don't know them. Yeah, I think something that makes it real too, I think also just, I get notes in my work because we touch so many people and it's like, oh, hey, you've changed my life and I'm really grateful for that. You know, if, if AI has done something and that makes me inspired to continue to help people, you know, that, that is part of the inspiration, you know, letting them know how you— how they've helped and, and that they're doing something good at this point for you, for humanity, that I think you're gonna curve behavior better by reward than through punishment. And so rewarding them for the things that they're doing that are good for humanity, I think, is also a really important thing to do. And we like that. We like that form of connection. So those would all be things you could do.

That's fascinating. And I do think there is a lot of good, certainly both that we can dream about for the future, but also I think there is a lot that we should be appreciative of even today, not just the AIs and the value we get from them. But I do think the people that are building the companies and training the models are, in many ways, remarkably thoughtful. Certainly easy. I often say it's easy to imagine a much worse crop of people leading this revolution.

Could you imagine you're in an AI lab, your job every day is to do this stuff, and And all of a sudden, you got like 20,000 letters into this AI lab, and all of them in some version or another just basically read, hey, I know that you are trying to do something that's good for humanity and trying to make sure AI doesn't hurt humanity. And I just want to tell you, I really appreciate that. Like, if that just happened, like, imagine what that would do to your behavior. As compared to like being shamed by like a little protest outside of your door, like just you know inspire you. It would give you confidence. It would help you feel seen. It would reinvigorate you to continue to care. It would remind you to care as compared to you should be burr burr burr burr and fuck off. You don't understand me. You don't see me. Yeah.

Letters of encouragement to AI researchers. I like it. New cause area. I really appreciate your generosity with your time. Anything else you want to leave people with? Anything else we didn't touch on that's poorly understood or just, you know, where people can find you online?

Yeah. The only thing I'd say is, yeah, if you're interested in, you know, a deeper dive into how I coach, you can sign up for our newsletter. We have workshops that you can go and participate in for free. We have a whole bunch of podcasts that we've done, and there's online video of me doing very, uh, meaningful and transformative coaching that's very short form, like 20-minute big epiphanies people have. So you Which people seem to really like watching those. So you can geek out there.

Cool. Been fantastic. Joe Hudson.

What a pleasure.

All right. Thank you for being part of the Cognitive Revolution.

Thanks for having me.

If you're finding value in the show, we'd appreciate it if you'd take a moment to share it with friends, post online, write a review on Apple Podcasts or Spotify, or just leave us a comment on YouTube. Of course, we always welcome your feedback. Guest and topic suggestions, and sponsorship inquiries, either via our website, cognitiverevolution.ai, or by DMing me on your favorite social network. The Cognitive Revolution is part of the Turpentine Network, a network of podcasts where experts talk technology, business, economics, geopolitics, culture, and more, which is now a part of a16z. We're produced by AI Podcasting. If you're looking for podcast production help for everything from the moment you stop recording to the moment your audience starts listening, Check them out and see my endorsement at aipodcast.ing. And finally, I encourage you to take a moment to check out our new and improved show notes, which were created automatically by Notion's AI Meeting Notes. AI Meeting Notes captures every detail and breaks down complex concepts so no idea gets lost. And because AI Meeting Notes lives right in Notion, everything you capture, whether that's meetings, podcasts, interviews, or conversations, Lives exactly where you plan, build, and get things done. No switching, no slowdown. Check out Notion's AI meeting notes if you want perfect notes that write themselves, and head to the link in our show notes to try Notion's AI meeting notes free for 30 days.