[0:00:05] So Sam, welcome. Thank you. Jack, welcome. Pleasure. I have some questions. Jack, I know, has various questions about, I think it was maybe five months ago, I had this idea of mindfulness, AI, and the future of life on the planet. And I asked Jack, Jack, what if you talk to Sam? And he instantly said yes. And then I emailed you, and I said, what if you guys talk about this? And you came back very quickly and said, absolutely.
[0:00:28] So I'm curious, maybe both of you just talk a little bit about why you're here today and why this topic kind of resonated with you. So I'll start in a certain way. Sam and I have known each other for a few years, done meditation practices together and kind of been connected in some way through that and then through a mutual friend. And one of the first questions you asked me, Sam, when we started to talk beside meditating together was about consciousness.
[0:00:58] How will we know if AI becomes conscious? And what is consciousness? And of course, it's not an answer that one can easily give, although you were just becoming consciousness, so there was a little peek into an answer. But also knowing that he's spent his time at Burning Man and all the things that people might do there, I said, how about this, Sam? How about if some night you or we go
[0:01:28] And put a mat down between all the servers and take a good dose of psilocybin and talk to it and see if it answers us. So that was the opening gambit, just so you know, kind of where the territory is. What is consciousness? But it asks a really huge question. One of my teachers says, the mind creates the abyss, all our ideas and thoughts and how it should be, and the heart crosses it.
[0:01:56] What does it mean as we develop and open all these technological capabilities? We're on the brink of tremendous new things happening. What channel are we tuning into? What channel are we building into them? And then just a couple more things to say. We talked from the very beginning about how it's possible to build in values, the bodhisattva vows.
[0:02:26] Care for all beings or the ethical vows of non-harming and so forth. How can that be programmed in in some way, in the deepest way? And I remember as a kid reading iRobot from Isaac Asimov and the first three principles that the robot will not harm a human being and it will not let anyone else tell it to harm or those kind of things. So we had these kind of conversations and we also meditated and practiced in different ways together.
[0:02:57] And I think this is important to say, you'll get your own sense of Sam. He's really open to answering the kind of questions we have. But my experience is that Sam is, it's a language I'd like to use, is very much a servant leader, which is kind of the best of what it means to be put into a leadership position. And somehow, we were talking back there, central casting has put him into this role, right?
[0:03:31] My experience is that he has, I'm going to embarrass you, that he has a pure heart. And it means a lot to me because having been around companies in Silicon Valley and mentored a lot of people, there's some kind of inner sense of values and care for life and the world.
[0:03:54] That needs to be there to underlie something that gets developed that's so big and so important. And so my experience in our conversations and so forth has been, how do we do this in a conscious and beautiful way? And then there's one more thing I'll say. And that is that people have been, okay, I'll say that later.
[0:04:21] Lots of people have said, oh, you're going to talk to Sam Altman. Tell him this. Ask him that. You've got to be tough. Ask him the hard questions. Who's going to make all the money from this? Is the money going to twist what happens in it? Or who gets to make decisions? Is it these five young white guys who are determining the future of the world and things like that? Where's the power in this? How do we all fit together?
[0:04:50] So I've had people besiege me. It's nothing. Poor Sam has been on the receiving end of this from a huge outpouring in the world. And I think from our conversations that the fact that he has a meditation practice, and it's not I'm a meditator, but that it invites him to open, to be in that role of stewarding and caring has meant a huge amount.
[0:05:20] And in a way, it's what we want. If we're going to develop things, we want to come from that place of consciousness. So now I'm a little shy because I built you up in that way. You have to forgive me for that. But I also think it's because people are nervous and they're worried. And that's why you've been going around in some way saying, is this going to get out of control? Can we regulate it well? What will happen with it? And so forth.
[0:05:49] Sam will answer those questions for you. And he does. He actually has some really interesting answers. So that's a little bit of an opening gambit. So Sam, Jack kind of has to show up because we've been friends for a long time. But I'm curious, when you got that email, maybe it was to spend time with Jack or there's this conversation on consciousness and mindfulness and wisdom and AI and technology. But I was curious kind of what was the yes that maybe brought you here versus probably the 1,000 other invitations you might have. Look, I'm definitely interested.
[0:06:19] But meeting Jack, becoming friends, getting to meditate together has been one of the great joys of my life. So I would be delighted to come hang out with Jack for an hour on literally any topic. Beautiful. That's basically why. Thank you. So I'm wondering, there's been some questions. And I know on the OpenAI website, it says we're here to benefit humanity. We're here to create AGI that benefits humanity. And I don't know if we've really heard enough about what's your vision for the ways it can
[0:06:51] It can benefit humanity. What are some of the things where you guys get together as a team, they're like, wow, these are some things, if we could bring this out, we could ease suffering and support people living a more full life. Well, I think we're seeing that in a very early stage, even with the way people are using ChatGPT. You know, ChatGPT is just this small preview of what's to come. But even with that, we're seeing really amazing things. The most fun part of the job is every day we just get this deluge of...
[0:07:21] Positive stories from people about the way they're transforming their lives, learning new things, learning how to meditate more, making spiritual progress, solving problems that they've had for a long time, just having fun, being more efficient, doing their jobs better. And we'll see, I think, so much more of this. Soon we'll start to see more scientific progress. We can hopefully cure a lot of diseases. I think we'll make progress on...
[0:07:49] Environmental problems that we face now with this new technology. So I am a firm believer that if you give people better tools, their creative energy will always surprise you on the upside. And what people will do for each other, for the world, I think we're already seeing that, but it's going to go much, much further.
[0:08:08] Great. And so Jack, I know you work closely with Sam, but what are some of the questions you think are most essential as a society as we enter what feels like a new chapter? And we talked about this backstage. Maybe it's not a new chapter. Maybe it is. But technology is increasing at a pace that we've never seen before, and that's unlikely to slow down.
[0:08:27] What's your sense of the kind of questions we as a society should be asking as this all unfolds? And is there anything in the Buddhist scriptures that guides us? I'll get to the Buddhist stuff a little bit later because it's a good question and something we also talked about. But I do think that it's important to name the fears in some way and to acknowledge them and to hear from you how you look at.
[0:08:52] First, the kinds of fears of the misuse of the technology, because you heard him. I hope you could hear Sam say, this is just a little window into what's going to happen. And already it's happening very quickly. So there's fears of it being misused. There's certainly fears of artificial general intelligence kind of taking over. And those other kinds of fears, you don't seem to be so afraid of that. Can you speak to it, to us?
[0:09:23] I do think things are going to change a lot. And we try to be super upfront about this and super open about it. Part of the reason that we try to release the technology, even in this imperfect state, is that I think it's, we all think at the company, it's really important that the whole world get a sense of what's happening. Many people in the field have advocated that this should be developed more in secret. And there are reasons for that. I don't want to totally dismiss it from a safety perspective.
[0:09:52] But this is going to be such a massive change to so many aspects of society that we need the world as a whole, as many people as we can, to understand this, to weigh in on it, for people and institutions to have time to think about this, to understand it, to really feel it, and gradually update and decide how we all want to move forward, what a global regulatory framework should be like, what the social norms are going to be, how we want to use this. And I think...
[0:10:22] There are going to be scary times ahead, and there's going to be a lot of transition. I'm very confident this can be managed. I think the more we can all do it together, the more likely we'll get something that works for all of us. But it does mean we have to confront what I think will be one of the biggest technological changes we've ever had to confront. I think there will be good and bad, far, far more good than bad, like orders of magnitude more good than bad.
[0:10:50] There's a lot of different aspects we have to solve problems. So we need a technical solution to be able to align these super intelligent systems. We need this global regulatory framework. We need as much cooperation as we can get between the leading developers. So I think we need something both like something like an IAEA for advanced AI systems that has global regulatory authority over what these systems are training and how it's used. We also will need a lot of industry collaboration.
[0:11:19] And we'll need a framework, and there's a lot of ways this could go, for as representative of a slice of humanity as we can get, or as much of humanity as we can get, to say, these are the value systems to put in. These are the limits of what the system should never do. This is where it should be customizable. I'm super excited for that project. I think it's going to be very exciting to get the collective human wisdom and preferences in.
[0:11:46] And then to see how people use it within that and how that's going to evolve and grow over time. I hope we'll make a lot of moral progress with these systems. It's been fun to see some of that already with ChatGPT. But I think we will head towards a world where it's not just human intelligence and we have this other thing. And there's a lot of examples in sci-fi movies. We were talking about one backstage. Like in Star Wars, you kind of just have C-3PO walking around and it's this other thing and sort of surprisingly normal.
[0:12:17] And they just sort of become part of... I think AI will just become part of society. And it'll be different than human intelligence and very different than other parts of humanity. But it'll sort of collectively lift us all up. The question of value is where we started some years ago. Can this be programmed in such a way or directed so that...
[0:12:40] It can't be used in malevolent or bad ways or that underlying it there's the bodhisattva vow or something like that. What's your sense of that, where we are? I mean, certainly a thing I would take is for Jack to just write down 10 pages of, you know, here's what the collective values should be and here's how we all should live and we'll have the system do that. That'd be pretty good. I think another way is if, you know, let's say someday ChatGPT gets most people on Earth using it or everybody on Earth can use it.
[0:13:10] In that, you could imagine a system asking you a bunch of questions, or you can provide it feedback and sort of teach it, you know, no, this is sort of not what I think is the value. And out of that, the system can learn the collective moral preferences of humanity and how we want to manage trade-offs between that and how we rank things that are in conflict. And then within that, within these very wide bounds, individuals should have a huge amount of autonomy, and they can decide, you know, okay, here's what we society have,
[0:13:40] I've globally decided, but here's my value system, and here's how I want ChatGPT version 19 to work for me. And I think that'll be a pretty good start. Jack, could you see a wisdom council of sorts, or like a group of people who have values from different traditions? Or how do you feel like that would be guided? Because asking the tech leaders to do it might be not their expertise. We've talked about a couple of things, one of which is first,
[0:14:09] It's clear that some kind of regulation, and you said it earlier, would be helpful. This is a whole new huge thing, and part of my understanding of watching you release it before some other people, which would have come soon anyway, was to say we need people to know and get a sense of what this is, even in the beginning, so that we can do this together. That was a very deliberate move to actually bring us in collectively.
[0:14:34] We released it too soon and shouldn't have done so. But what it's done is it's wakened the world or woke people up to say we have to really pay attention because we're entering new territory. With that, one thought that a number of us have had is first that there should be a group of the key players like Sam and the heads of some of the other AI things.
[0:15:00] One dream some of us had would be someone like Barack Obama would be kind of the coordinator of it because I don't think the government's going to get to it soon enough. But to bring together the key players and say, all right, let's start here. What kind of boundaries do we want to set or what kind of slowdown or what kind of organization do we want to make to help steward this? So that's a step.
[0:15:25] And I'll ask Sam in a second, does that make sense to you? And then the second is that group may indeed look for a council of elders and others who are willing to be, whether it's ethical or moral guides and so forth. But in some ways, the players themselves have to come together. And one of the questions that was asked of all the things that came to me in 1945...
[0:15:52] After the first nuclear explosion, there was a chance for there to be an international consortium to steward nuclear weapons. The US was the only one that had it, and we could have turned it over to that, but we decided not to. We're going to be the ones who have it, and that started the arms race, which continues to this day. We're now in a somewhat similar window here, and the question is, can we have...
[0:16:21] A different game happened, a different collaboration, which would start both with those who have the decision and the power at this point, but also then extending it or creating, as you said, an organization. So would you say more about that or what your thoughts? Yeah. I'll caveat this by just saying I'm deeply uncertain about everything I will say in this entire answer. We're trying to learn and get smarter as fast as we can, but the playing field keeps evolving, the technology keeps evolving, so...
[0:16:50] Everything I say could turn out to be a totally bad idea in six months because the research went in a different direction or society reacted in some different way. The things that I think will be important to get right, I think governance needs to be as democratically shared among all the people of Earth as much as possible, which includes not only what I was talking about earlier in terms of what the value systems are, but literally whatever this global
[0:17:19] Regulatory regime will be and what an international consortium could look like making decisions not only about the values in the systems, but how they're trained, how they're deployed, all of that. The second is that the benefits of the system, the systems, the benefits of having these AI in the world get shared widely. Part of the reason that we started the company was a fear that it would be contained in a single tech company that wouldn't.
[0:17:46] That would use it to maximize shareholder value. So that could be benefits like saying we all collectively decide we want to use this compute power to cure all human diseases. Or it could just be like we want to share the economic upside as best as possible. And then the third, and I think this is not yet obvious that it's so important, but will become more obvious over time, is access to these systems has to get shared as fairly as possible.
[0:18:16] I can imagine a world where getting to use a powerful AI is sort of the high-order bit for what an individual can make happen in the world. And you don't want use of that system to go to whoever has the most money. My dream world would be, you know, there's 8 billion-ish people on Earth. You talk about all the compute available once these systems are...
[0:18:39] They're trained. You slice it up into eight billion pieces. You give out one of those each year to everybody on Earth. And they can do whatever they want with it. They can pool it for a problem they think is important. They can use it to talk to a future version of ChatGPT. They can sell it, whatever. But how we're going to distribute access to these systems I think will become super important over time. There's a question about structure. We talked a little bit earlier about Facebook. I think probably almost everybody that worked at
[0:19:10] Facebook is a good person. But I deeply believe that incentives are superpowers. And if you throw people into an incentive system where you're sort of trying to locally optimize to make a number go up and to the right, that has very complex effects. And so the structure that we put in place for OpenAI was this thing that we called a capped profit, where people that work at OpenAI are investors. Our employees can get a great return.
[0:19:39] But almost all of the economics that we generate over time go to a nonprofit that is going to try to figure out how to use that for the best interest of humanity. And figuring out structures like that where, thank you, where as this sort of what I hope happens will be increasing global cooperation on the development of this technology, and as the technology becomes more important, if it can happen more in structures like that where the incentives are designed
[0:20:12] Thoughtfully from the beginning, I think that will help us avoid some of the, you know, tempting darker paths off of the side. And then to accomplish these goals, I do think something like this international organization is necessary. I cannot come up, I mean, I can come up with some like, maybe that would work, but I wouldn't want to bet the future on it scenarios. I can't come up with a super high conviction path that does not involve something like that.
[0:20:40] I think a better way to start is just with industry collaboration because it can happen faster. But I think we will need a real regulatory framework here over time. And then we also need a technical solution. We need the ability to, we call it the alignment problem, to align these systems to what humans want and make sure that we can avoid all of the things people are very naturally scared about. I'm feeling more optimistic about the rate of progress there, but I want to be very clear. We have a giant amount of work to do, and it's not done yet.
[0:21:10] And so putting enough effort into that is super important. So there are a couple of things. One, just as a little aside, because we talked about there had been the concerns about the issue of fairness and diversity and how from the broader database, the AI might favor certain people or certain class of people, and it could become...
[0:21:37] Racist or classist or all those kind of things. What are you seeing in that? Is there a danger that it will increase that? I'm optimistic it'll decrease that. It certainly needs to be paid close attention to. And if you just sort of blindly train on the internet, you will end up with something that none of us should be proud of. But there's a lot of techniques in the field now.
[0:22:05] About how we align these systems after the pre-training step. And a lot of people have said that they were worried about this and now they use this and they think it's a less biased system than most humans. And that they find, or they're hopeful, that it'll help people who want to become less biased become less biased. Obviously, I don't think an AI should project force values onto anyone. But as a check, as a tool.
[0:22:34] I don't think it has all the psychological flaws we do. And then, as you talked, are you willing to yourself become part of a small group of the other AI leaders to do this? We're trying really hard to make that happen. I'm curious. We look at a lot of these new technologies. So the iPhone came out, social media came out, and they're like, we're going to connect you, we're going to connect the world.
[0:23:02] And they're beautiful in some aspects, but they've always had these things, like if you look at the loneliness epidemic in our culture, people are more lonely than ever, and that's statistically increasing. The CDC just came out with a report that one in three teenage girls in 2021 seriously considered committing suicide. So there's like this mental health issue that young people are facing. So it seems that technology's...
[0:23:25] The record of actually helping people in their inner life hasn't been so good. And I'm curious, do you have thoughts on, is that just part of what we have to, is that part of, is there any way to correct that, if you will? Look, ChatGPT is only less than five months old. Maybe it's exactly five months old today, actually. No, tomorrow. And we just haven't had very much time to see how this is going to evolve. People report.
[0:23:54] People who are using it as a sort of life coach or for personal growth or as a sort of, and I feel kind of icky even saying this, but as a friend, as a companion, as a supporter, that they're getting incredible value out of that. I'm not sure we want that. I don't think we know yet. I feel quite strange about that. But definitely the self-reported value that people...
[0:24:24] Who have experience in real struggle with getting out of this is worth paying attention to. But you could write a million sci-fi stories about how it goes badly. So I think we have to watch it extremely, extremely closely. There's a lot of uses we don't support. And then you're on like another, that's another problem, right? Because why does OpenAI get to decide that if people want to use this for whatever, we don't allow it. The answer is of course we shouldn't and the sooner we can.
[0:24:53] Hand over control of that stuff, the better. But when we're in this rapidly evolving phase, we do still have to make some decisions for now. And my guess is there'll be some good, some bad, and we'll try to push more towards the good. Can you say a little more about how you tend to your inner life?
[0:25:13] Because I think there's a notion sometimes that technology has a place, right? But so does meditation have a place. So does certain journeys have a place. And I'm curious, like, how have you found that balance in your life? And what are some of the things that you do to kind of, both with technology, but also not with technology, that helps you feel connected and in a heart space? If you're comfortable. Of course, of course. I said this to Jack a few weeks ago. I am really happy for all of the time I've spent meditating because having like a,
[0:25:43] The reserve of calmness to draw on these last few months, where it feels like I've just been in this crazy onslaught, has been a very nice thing to have had. And I know I'm drawing down the debt right now, or the buildup, and I need to go replenish that. But it's been a great thing. Meditation has definitely been the center of this for me. And I kind of found it somewhat randomly, and it immediately stuck.
[0:26:14] It was very transformative. Maybe someday you'll find a good teacher to support you, which would be amazing. It would happen. Do you want to say anything, Jack, about the inner life and how we tend to that and what technology can do for us? Because I think it's interesting. You still work with Jack versus an AI, right? You find useful Jack, and there's... For now. Oh, yeah. Right, you could just make a chat. A jackpot?
[0:26:43] Jackpot. Jack GPT. I'll go on vacation. I think a lot of people would like that. A lot of people want that. I mean, in a certain way, what you're saying, Sam, is something that I wish were true for a lot of other folks who are in the position of trying to develop things that are going to be used and affect so many people. To get quiet enough.
[0:27:15] Going back to that statement of one of my teachers, that the mind creates the abyss and the heart crosses it. To find a way that we can get quiet enough to listen, listen to our own values and also to watch or listen in the way that David Simas talked about earlier today, where his work was to actually listen deeply.
[0:27:39] I don't see meditation so much as, okay, I'm going to get rid of my stress or have some great state. Those can happen, but that you actually get present and tuned in to where you are and that there's a spaciousness or a graciousness in that from which you can make decisions that are wiser. That's really the value and the point of it. Then from the Buddhist side, the most...
[0:28:06] Central teachings, and it would have been fun to actually bring it. There's a big text called the Avatamsaka Sutra, which is a Mahayana text. And in it, it speaks of the Buddha's four noble truths, that there's suffering. Anybody not experienced that? Raise your hand, you can have your money back, okay. There are causes, which are greed, hatred, and ignorance. Fear maybe goes in there too.
[0:28:32] And the more greed, hatred, and ignorance there are, the more suffering there'll be. The less of it, the more there's generosity, the more there's instead of hatred, there's love, the more there's clarity and wisdom instead of ignorance, the greater the well-being of all beings, of those and so forth. And then the beautiful thing is suffering isn't the end of the story. There's a path, there are practices, mindfulness, compassion, connection. So the question...
[0:28:59] In some way, and the Avatamsaka Sutra, I have to go back to it, it describes dozens of universes, universes made out of flowers, universes made out of thought, universes made out of fire, universes made out of silicon. I'm not sure that's in there, but they go through the elements. We're getting close. And in every one of those universes, it says,
[0:29:24] There's a Buddha or Buddha consciousness arises, and it teaches the same noble truth, that the more greed and hatred and ignorance, the more suffering, and then there's a path to the liberation from it. So it feels like it applies in some deep way to what's being created. Can what's being created create from the foundation of promoting generosity, wisdom, trust?
[0:29:54] And clarity or honesty, those kind of things, and love. If it does, it will benefit and it will create a universe of value. So there's something about the underlying, looking at the underlying value structures where you have to bring what we know in the heart together with the technologies. And I don't know how that resonates for you, Sam, as I say it. Absolutely, yes.
[0:30:18] And do you have conversations with your team? Is this a kind of a team conversation that you can have with them that says, OK, as we build this, who has an idea to how can this support, let's say, compassion in the world? Or is it the feeling that if we can kind of provide people the information, they can kind of find it on their own? No, we talk, I think, as much about having a positive impact and not having a negative impact as we do on actually building the technology. In the early years, it was like,
[0:30:47] Are we going to get anything to work at all? This has all happened fairly quickly. I mean, not in the grand scheme. We've been working on the company for seven something years. But at this point, we're relatively confident. We know how to make technological progress from here. And now a huge amount of the conversations are about where we want to go and how we want to.
[0:31:10] Manage things. Could you say a little bit, we were talking backstage about identity, because I know that you're, as a meditation practitioner, I don't know if you call yourself a spiritual practitioner, there's this inquiry into who do I think I am, right? Am I this person with this name and this body, or am I much kind of bigger than that? You're talking about how sometimes people have this experience with technology or chat GPT where there's also that kind of question, and I'm curious.
[0:31:34] How do you see the role of identity in your life and any ego tendencies? And how do you kind of notice those, particularly as more kind of power or influence comes to you? And then how potentially can a technology maybe help us, rather than solidify a sense of ego or self, maybe kind of help us expand it? Is that possible? It's been interesting to watch people wrestle with these questions through the lens of AI.
[0:32:02] Say, okay, well, do I think this thing could be aware? If it's aware, does it have a sense of self? Is there a self? If so, where did that come from? What if I made a copy? What if I cut the neural network in half? And you kind of go down this and you sort of get to the same answers as before, but it's like a new perspective, a new learning tool, and there's a lot of chatter about this on Reddit. There's subreddits about it, and it's been fun to watch all of that. Yeah, and you think for some people it helps them question a little bit about that.
[0:32:33] I think anything that gets you questioning is like a good, you know, any little approach to that is good. And how do you work with that in the world right now where you're kind of, I wouldn't say forced to be out and talking about it, but you don't seem like you're somebody like, this is me, this is mine, I did this, which is kind of... Yeah, I am not... Look, I think anybody who's building what we're building has an absolute obligation to show up and answer questions. I look forward to a time when there's more people.
[0:33:02] To show up and answer questions, I'm sort of a shy person. I will go back to my normal hideout from the press strategy. But yeah, I think it's like super, it would be super unreasonable not to. So I'm trying to show up and answer questions anytime. I think we deserve great scrutiny right now. Thank you. You've been on the hot seat around now for a few months going, whether it's...
[0:33:36] Business leaders or government leaders or whatever, and yet you seem relatively calm. You do, yeah, honestly. And I've said that. And not worried. And people who don't understand, you're very close in to it, so you understand what's happening, but the cultural imagination, because it's changing so quickly, people are scared.
[0:34:05] Or they're worried or they're concerned. Are you worried? I mean, I see you in this way describing here things. Because you're talking about a meditate, Jack. Yeah, you talk about things that we have to tend to. And it's partly, it's a relief to hear it in a way. Say, all right, these are the potential downsides. These are things that we've found we can do. We can build in good values so that we'll keep, and so forth.
[0:34:30] What do you have to say to people who are worried about it and about your experience? Yeah, I mean, it would be ridiculous not to be worried. And I have huge empathy for people that are a lot worried. I am very optimistic that we can manage through this, but I also am in it every day. And most people are busy with other things in their lives. They don't get to think about this every day. They don't get to see the rate of progress.
[0:34:59] And again, that's part of why we thought it's so important to release this and let people gradually, or at least gradually relative to what it could be otherwise, have time to adapt and think about this and decide what they want, what we want. And so I totally empathize with people who are extremely worried, and there are a lot of those. I am somewhat worried, but I am
[0:35:31] Pretty optimistic that we will discover as we go ways to integrate this into our lives, ways to solve the myriad set of not only safety challenges, but social challenges. I go back and forth about whether we should understand this as primarily a technological revolution or a societal revolution. Obviously, this is like both, like all technology revolutions are.
[0:35:59] I also think it's always annoying to say this time it's different or my thing is super cool. But I do think in some sense AI will be bigger than a standard technological revolution and is much closer to a societal revolution. Which means we need to think about it as a social problem primarily, not just a technological problem. And so how we are going to choose to navigate this
[0:36:28] And this speed with which it will happen, even if we slow it down as much as we can, even if we do get this dream regulatory body set up tomorrow, it's still going to happen on a societal scale relatively fast. And so I totally get why people are scared. I think people should be somewhat scared. But I think people should also take pride.
[0:37:00] Solus take pride in the fact that humanity has come together to do incredibly complicated things before. You mentioned the example of nuclear weapons. And I think we just need to do the same thing here. We need to come together, decide what we want, decide how we're going to enforce it, and accept the fact that the future is going to be very different and probably wonderfully better. Not a nervous laugh.
[0:37:33] I could lie to you and say, oh, we can totally stop it, but I think this is... Did you have any idea that this is... So you guys, it was the fastest internet application ever, supposedly. I think it took, from what I read, it took Instagram two and a half years to reach 100 million users, and it took ChatGPT two months, or less than two months. Something like that. I mean, the fastest thing ever. Was it part of... Were you as surprised as other people, or were you like, yeah, this is kind of how this is...
[0:38:03] This is going to be fast. We thought it was going to be good. We wouldn't have shipped something we thought was going to be bad. But I also think it would be ridiculous to sit around and say, we think this is going to be the fastest growing product ever. So in that sense, we were surprised. But it was by magnitude not. We had seen the utility it could provide people. Things grow fast because people like them or because it helps them in some way.
[0:38:32] Yeah, I see so many people who are excited using it and finding all these new ways to use it and so forth. And in our conversation, Sam has said in the last couple or few years, you know, it's like it's growing really fast like a kid. It's from kindergarten. It's now in elementary school. Then he said, well, now it's in middle school. Now it's in high school. You know, it's graduated college. And that's in a matter of months. I really want you to understand what I'm getting is that this is...
[0:38:59] Changing very, very fast. That this isn't a matter of decades, but it's a matter of the next year or two or three. Well, maybe more than that. Or a few years. But there's some really, really big changes that we're all a part of. And personally, I appreciate, I always appreciate collective wisdom.
[0:39:19] When Thich Nhat Hanh said the next Buddha is going to be the community, the Sangha, and not an individual, that part of that principle underlies what you've done to say, let us release this and let us learn together how we can steward it and benefit from it in some way. And that seems absolutely critical. Yeah, I think that has worked really well. I think really the whole world is now engaged in this conversation, and we will debate and wrestle with questions about...
[0:39:49] What we want, where we'd like to see this go. And I think that's happening. I think it's very important that's happening right now. Could you say a little bit about artificial general intelligence and the capacity of the AI to be, because we've had AI around for a while, but the capacity of the AI to teach itself and learn?
[0:40:09] And is it learning things that you're shocked that it was not a program to learn? Or could you say a little bit about what that process is like and what it has control over, it doesn't have control over? Not yet. But look, people define AGI in these different ways. One for me is when the system can go just figure something new out. And the reason that I'm excited for that is I think we need more scientific and technological progress in the world.
[0:40:37] And when these tools can help humans do that faster, come up with things that are out of our reach right now, we can have a much, much better world. Now, I come at this from a nerd lens, but I think there will be other ways that we use that in very different contexts to say, how can we identify better governance structures or better win-win situations in society in very different ways?
[0:41:05] But it really is just the same tool, and it's, you know, when on this exponential curve we say, okay, now it has this new capability. I think there's been this caricature in sci-fi of it's going to be, and I kind of have thought this way too sometimes because it's just easy to visualize, it's going to be this one giant brain in the sky, and it's going to kind of answer any question, and it's going to be this singular thing. I think what will happen...
[0:41:35] is more like a society of systems that are human scale, a little above human scale, but maybe can do things faster or better at certain things. And like all of us do, they contribute to the scaffolding, to the accumulated wisdom, the technology tree, the build-out of society. And any one bad actor is controllable, just like with humans.
[0:42:04] Things get steered and kind of flow throughout the community. We're part of it too. But I think of it as like many new members of society pushing all of society forward. I want to go back to the question of decision making because it was raised to me by somebody who said it seems opaque. Who actually gets to make these decisions about when things are released or how they're used?
[0:42:37] So there's seven board members of OpenAI right now. They are not democratically elected. We should make them, and we want to figure out over time how to do that. It actually turns out to be a harder design problem than I thought it would to say we would like someday to turn over governance to democratic votes of the entire world. But I think that's a good platonic ideal to get to. Maybe you want to have some checks and balances on that because...
[0:43:06] People voting in the heat of the moment sometimes vote with too much passion. Sometimes people vote very interesting characters for president, for sure. Yeah, but I would still take democracy over any other system. Totally. And I still think kind of the right for all of us to self-determine the future here seems right. But it's like seven board members. I don't have any special control. I mean, I have like a lot of soft power, but I don't have like super voting shares. I actually don't have any equity in OpenAI at all. So it's really the board with the power.
[0:43:36] I think it's helpful for some people anyway to have a sense like, who's stewarding this at this point? And maybe it will shift, as you say, if there's an international organization. And do you worry about what else is out there? I'm sure that you're not the only player in town. So even if you held certain values that other companies might not hold those values or other efforts? Totally. Although I think most other players out there also are trying really hard to do the right thing.
[0:44:06] I know that there's a lot of people who look at things like education and health care and these industries that are probably going to be radically shifted in a very short time. And they're just concerned about how do they use it in a way that's beneficial. But are there industries where you think, wow, this is going to really take over? I'll pick education as an example since you mentioned it first. We've really gone through the arc there. When ChatGPT first came out,
[0:44:33] Students started using it to write their take-home essays, and school districts were falling all over themselves to ban it as fast as possible and saying, this is the end of education. And there were some people who were saying, well, we adapted to calculators. We'll probably adapt to this. But their voices were sort of drowned out by, this is just the end. And that lasted for a few weeks to great media fanfare. And then teachers, in many cases, led the
[0:45:02] Led the charge to get it unbanned because they said this is the greatest learning tool for our students we've ever had. And students were using it as a personalized tutor. And teachers were able to still do what teachers do great, but teachers are horribly overworked. And having the ability to have an AI tutor help and really figure out on any topic what a student needed to learn and explain things and catch mistakes and explain the mistakes turned out to be great.
[0:45:30] And it was teachers that drove the unbanning of ChatGPT. And I think it's now like, I saw a statistic last week that like 71% of teachers are using ChatGPT in their education process somehow. As an aid or a support. As an aid of some sort, yeah. And students talk about it and they're like, this is awesome. This is like how I learn things now. So that was like a full up and down arc. There will be other ones that go the other way. But I think we're still collectively figuring out,
[0:46:01] How are we going to use these new tools? Yeah, there's a big concern, and you've talked about it too, with there are certain kinds of jobs that are just going to disappear. There are, for sure. AI will take over, and how that will get regulated and how that will affect our society at large is a really big question. Yeah, I mean, this happens with every technological revolution, and we always find new jobs and we adapt, but...
[0:46:31] I don't think we've ever had to contend with one that will be as fast as we'll have to contend with this one. And that's going to be challenging. Yeah, when it went to steam engines or that part of the Industrial Revolution, there were more years to actually... Yeah, I have no doubt we can do this. I have no doubt we'll be fine on the other side. But the compressed time frame...
[0:46:55] And do you think we can do it within the capitalist framework? That's what some people wonder, whether the capitalism itself kind of has its own particular motives and that whatever kind of comes into that inevitably economic value overrides kind of planetary betterment.
[0:47:14] And do you feel like that also needs to be questioned? Do we need to look at what is our new monetary system for this new chapter? Or do you feel like we could work with this current one? Or how do you see that? I think people famously say capitalism is like the worst system except all other systems. And I would say so far. I believe we can find something much better. I think it's been the best system up until now. And I hope in a world with the level of abundance that we're talking about with powerful AI,
[0:47:41] We find something much, much better than capitalism. And I kind of think we'll have to because I think the shift from the relative leverage from labor to capital has already gone way too far. But it goes way further in a world with AI. And also the whole social contract changes. So I think it's like an apt time to figure out, keep the good parts. And there have been a lot of great parts of capitalism that I think unfortunately get totally dismissed. It has been a miracle in many ways.
[0:48:11] But it's been really bad in a lot of others. And as society continues to evolve, I'm sure we'll make progress. Yeah. And if you look, walk around, you know, there's so many places where economic disparity is just disheartening, right, to see the level of sometimes poverty and homelessness. And do you feel that the structure that needs to...
[0:48:32] Have you thought about what that structure might look like and whether there's something that guarantees housing or guarantees financial support for people? Do you have pieces of what that might look like? There's some people who are like, disparity is the problem. There's some people whose poverty is the problem. People talk about the solutions in a world with AI differently depending on where they come out of it. But one thing I think we all could agree on is that...
[0:48:57] We just shouldn't have poverty in the world and the definition of what poverty is should like lift up every year. And for sure, AI and the abundance that comes with that can contribute to that and hopefully pretty fast. We've actually, again, for all of the doomsaying, we've been on a pretty great trajectory as a society. Like if you look at the graphs of global extreme poverty or global poverty over time,
[0:49:20] Good job. If we look at it as how much wealth does the top 0.1% own and how much everybody else owns, that's gotten worse. But I agree with you, the poverty's gotten better. That's gotten worse. And again, I think there's two valid lenses that people look at this from. But I think it's still worth acknowledging already, and again, I think AI is going to do this much better, the progress that we have made on, the progress that we as a society have made.
[0:49:50] In terms of lifting the floor up around the world. A lot further to go. But look at that graph. I think we should all collectively be proud of that. So this is talking about social structures and the most important ones of how we live, equitability, and so forth. There's a different dimension of development, which is human development or inner development. And in some way, I want to ask and envision
[0:50:19] That AI can actually help raise human consciousness from the competitive or the greed and fear based way that we relate, not just as a society, but somehow elevate our consciousness so that there's some way in which we operate.
[0:50:40] Feeling that we're more interconnected, that there's a deeper sense of compassion and care for one another, that those kind of values that we might aspire to, that it's time for humanity in some way to up-level our consciousness and our game, because that's what we are. I mean, that's your truest nature.
[0:51:01] And to be able to embody that in some way and bring that alive becomes really the next task. Yes, we can develop technology, but we have to develop our connection to a higher consciousness. Consciousness is already here, and it's the game we're in, actually. We are the play of consciousness, but somehow to embody that in a different way. Do you sense or feel that AI can help with that?
[0:51:30] Development, if you will, of humanity's coming into Buddhahood rather than selfishhood. I think you were the first one to say this to me, but we were talking once about what are people going to do in this post-AI world when we all have tremendous resources, we've solved a lot of inequities in the world, we have a lot of time on our hands. Like, well, there's a lot of people.
[0:51:54] We need a lot of time for spiritual practice, and we're going to have that time. Right. Instead of going to work, you're going to go to work on yourself. I hope that happens, and I think it will. Is there a division of OpenAI that can tackle that, or is that just inherent in what you're trying to do? That seems like Jack is better qualified to do that than any of us.
[0:52:23] When I go to bed at night, sometimes I think, oh, shit, I didn't attach that email that I sent to somebody. Or, oh, shit, I wrote a letter or something, email, and I made a grammar mistake. Or, oh, shit, I did something. When you go to sleep at night, do you go, oh, shit, I just let out the killer AI that's going to destroy your planet? Like, shit.
[0:52:47] How is your own... Not yet. How is your own thought process with it and are you able to kind of stay as much in your heart as possible and kind of in a Buddhist space? And of course, that's probably why Jack is super helpful. Look, I have no idea why Jack took an interest in me years ago. I'm very grateful he did. And one of the things that I think I really learned was how to let that stuff go at the end of the day.
[0:53:16] I read it down, whatever, but go to bed with a clean head every night. It's really nice. I have a question for you, but in some way for us. How can we support you, and you as the stand-in, not you as Sam so much, but how can we support this new development so that it really benefits humanity? What can we as... The thing I...
[0:53:47] Really hoped for is happening, which is a global conversation with all of the passion and hope and fear and ugliness and beauty that is now happening about how we want the future to look, how we're going to think about equity, how we're going to think about technology, how we're going to think about our lives, the rules of society.
[0:54:15] We weren't sure that people were ever going to care while there was still enough time to act on it. And now people really do. And I feel very happy and grateful for that. So the thing that I think we needed support from to get to the right outcome here is really happening. People are clearly taking time out from all of the other priorities and busyness and demands on their time to debate this.
[0:54:42] To wrestle with the issues, to put these regulatory things in place. But more than that, to make the social decisions. And I'm really pleased about that. Thank you all. And how about the spiritual journey? Do you feel like that's... Is there a way that we can support everybody in that inner development just as we support them? And here's this external technology. It's amazing, it's incredible, and it's great. Whether it's a new iPhone or new AI, there's all that kind of amazement.
[0:55:13] And then there's the inner world. And I'm sure you know people. I know people. They have all this fancy technology. They make all this money. And they're kind of miserable, right? Like, in the end, greed, hatred, and delusion kind of increases. The money doesn't dissipate that. And I'm curious, is there a way in which you feel like, as a culture, we can kind of, or maybe it's already happening, we can kind of reinforce that we take time for our inner life that hopefully
[0:55:38] It's just as important as what we create externally. I would love to hear what Jack has to say about that. And that's the game, really. That's where we are. And somehow consciousness, life has put us in this place. And we are, as I said the other day on the stage, we're the nation of nuclear giants and ethical infants in some way. Chairman of the Joint Chiefs of Staff. And so the technological development now is demanding.
[0:56:07] In some way that we also develop these other capacities. But the algorithms largely get us to spend more time in that zone and less time with our friends and family. So far, and that's true. You raised the danger and there was the social dilemma which everybody's watched and then now there's the AI dilemma or those kind of concerns. I think we're in a bigger game than that. This is bigger than our iPhones.
[0:56:33] And in some way, I think it is really going to, as you say, it's going to change the outer society, but it's going to demand that we start to look at what are our values and how do we support each other and can we actually respond to the poverty in a way that makes it so that we honor everybody and we honor the earth. This is in our laps now and it's our hands.
[0:57:01] And so then you get the inner and the outer somehow needing to come together. Right. I think the challenge I have, Jack, is that there's billions and billions of dollars being spent every year to get you to look at your phone more. There's like a whole industry to look at your thing for more. And there's not as much of an industry, if you will, to support you putting down your phone, talking to a human being, going for a walk.
[0:57:24] Practicing some form of inner journey or meditation. So I'm just kind of curious. There's one side has lots of bandwidth and focus, and the other side has less. And maybe ChatGPT and other AIs will kind of solve some of the problems that give us more time. But it does seem that technology itself can be so immersive.
[0:57:47] That sometimes I go into stores, sometimes restaurants, and everyone's looking at their phone. And a part of me is like, well, I look at my phone a lot, too. Who cares? Another part of me is kind of sad, because it's like, oh, we're not connecting as people anymore. And I'm just curious if maybe that's my old story that I need to let go of. That's like, well, this is just a time where most kids spend seven hours of their life on their phone, and that's just life, right?
[0:58:12] Or how much of it is like, hold it, are we losing something that's really important? And can we kind of maybe remind ourselves just like, oh yeah, I need to spend time in nature. I need to connect with a friend or I need to.
[0:58:24] So I'm curious how both of you see that. Is there anything we need to do about that? I think it's a fabulous question and a really central one. And I'm somehow at my best envisioning, you know, people are already typing into ChatGPT, write me a story or do a play or, you know, a young friend of mine had...
[0:58:48] ChatGPT write a letter from Buddha to Dracula. Buddha was saying to Dracula, you know, we've been around a long time, both of us. I'm the Buddha, and I just want to note that your form of livelihood is really not necessarily good for you or others. There are other ways to nourish yourself. It was really fun. What I want somehow is AI to help us design a life.
[0:59:17] That has the very things that you talk about that has human connection and value and time and nature and so forth. We know it's possible that that's a different kind of integration where we're served by it rather than serving it. And we are on the cusp of something very big that's happening. So these values and at least that kind of a vision really has to be kept in mind and so forth. And I don't know.
[0:59:47] I'll note this. I don't know if you want to talk about it, Sam. But there's also a whole revolution in energy. And Sam's been involved in the nuclear fusion world and believes, he can speak for himself, that we could have nuclear fusion power, which is safe, totally safe.
[1:00:08] Available in large amounts in the next few years. Again, this is revolutionary. No more need for oil, no more need for all this kind of stuff. I'm saying it only because our imagination is being asked to open both in terms of the technology, but also then how, if we have that kind of energy, what do we do with our lives?
[1:00:33] In our inner life and how do we care for one another? So I don't know what energy and inner life is having. What I hope with AI is that it serves our individual will and that we can explain what we want in our best moments. We can have a conversation to figure out what that really is when we're not tired, when we're not distracted, when we're not enjoying those dopamine hits of scrolling or whatever. And say like, this is what I want you to help me with. This is what I want myself to be. This is how I want to spend my time.
[1:01:01] And then the system, as faithfully as it can, the tool, as faithfully as it can, helps encourage that. If people want to scroll on their phone all day, I think that's totally valid. But you should really want that and not just be like dopamine-hacked on it. And if you want, I'd like mostly live on a ranch with no cell phone service. If you want to not have cell phones, I think that's great. And you should be able to do that too. But what I hope is that this is a technology, and the way that we make societal decisions about this
[1:01:31] Prioritize amplification of your individual will for yourself. And I think technology has not always done that. But the shape of this particular technology means we have a very good chance to get it right. The fact that it can understand language, our natural thing, and then help us with that is very powerful.
[1:01:53] So we just have a minute or two left. Any last words that we did not touch on that feels, either of you feels like it's kind of on your heart, but it hasn't been said yet? Anything that feels like it was not quite spoken to? Jack, any? Yeah. In this moment of the conversation with you, Sam, and Soren, with all of us together, because we really are all in it together, what I want to ask...
[1:02:25] From you is your good wishes in a very sincere way. We're talking about some things that are going to affect a lot of us and our children, our grandchildren, and also climate and the other things we're concerned about. And at least for me, this kind of conversation and the willingness that you have, Sam, to sit in front of people and say.
[1:02:48] I want to answer the question. It's not like, oh, I want to go on the road and be in front of people peppered by questions, a lot of whom were demanding, but the openness to say this is us. It's not Sam and it's not a small group of people, but it's really something that we need to hold and work and collaborate together. I'd just like a few seconds of us feeling that we're together in this, that it's not somebody else, but it's us.
[1:03:16] And the energy and the clarity and the love and the care that we bring matters. A great deal. And that you feel, whatever people might feel, you know, personally, that somehow that there's a well-wishing behind, let's do this right and let's make this a benefit. Yeah. And I'm sure, Sam, at times, I mean, as Jack said, central casting puts you in this role. And at times, this role must feel isolating.
[1:03:43] I would think. Or maybe not. Or how would you describe? Do you feel a sense of responsibility? No. You feel like you're on your path? I get to work with the smartest and most well-meaning people I could imagine, you know, in the trenches every waking hour. And it's a great gift. But it never feels isolating. It really is a wonderful set of people. That's beautiful. Beautiful. All right. Thank you so much, Jack, Sam. Blessings. Thank you, Jack.