Sam Altman On AI Revolution, Superintelligence & Future Of Humanity | FO583 Raj Shamani
17 clips from a 68:01 video · top score 94
Get clips like these, freeTranscript
Interesting. And but you wrote a you wrote an essay, 13 rules to become successful. Well, I remember that I wrote that. I don't I don't remember what they were. And I bet I could do a better version now. I bet I could do a shorter version now. Okay. So, instead of 13, what would be I think this point about compounding the one I had first in there. I think this is probably underestimated. Uh a a lot of people I think don't drive themselves to continued forward progress. The joke of this is like you know sure you have 20 years of experience but you have 20 times the same one year of experience and you really the thing that I think is so cool about a career is you can build on each successive phase but you have to continue to push yourself. have to like keep doing uncomfortable, non-obvious new things. And a lot of people fall off that very early in their careers. And so staying on the compounding exponential curve um is empirically not what most people do. And I think that's a very good not obvious piece of advice.
Transcript
Listen to less advice. Do you sometimes go home at some point and be like, "Hey, all this advice is maybe right and maybe I'm wrong." I think YC could probably distill five to 10 very important, highly confident pieces of general advice from all of the incredible wealth of experience. Um, and I think you should listen to those five or 10 things. Uh like I think the slogan of YC or the motto whatever it like makes something people want is the like that is like the single commandment of startup advice to the degree there is one thing for early stage startups and then there's like another list of things that are almost as important but you're not going to become a great founder by listening to like the 800th podcast of advice about how to start a startup. You got to just like do it and figure it out on your own.
Transcript
Was there any point ever in your life where nobody else was convinced but just you? I mean, there have been like small decisions where I was like, "Open I was going to do this thing and none of the rest of the leadership team wanted to." Give me an example. Um, launching chatbt was kind of an example of that. Not that people didn't want to launch a chatbot, but most people thought it was too early and not good enough. Interesting. And what made you go against all of them and be like, "Hey, let's just launch it the YC ground." I think it was the Yeah, I think it was I think it was the YC experience. I think so. Like I had really I had spent so long drilling it into founders that if you're not a little embarrassed of your first version, you're waiting too long to launch because what you need is feedback.
Transcript
Can you tell after meeting a founder in like 30 40 minutes that this person will be able to build a company or not? Not perfectly but yeah I mean I think you can have like a pretty strong intuition. What is that? Like tell me what what do you see? What do you understand? What pattern do you see? Uh one of there's like many obvious things that I could say. I'll try to say the ones that I think don't get talked about as much. If they're like a clear communicator and they can really explain things well and it's easy to follow, this is a huge thing that I look for. um if I leave the meeting like convinced that I want to work with them um that's a huge thing I look for because you know a big part of the job of a of a early stage founder is to be an evangelist in all these different ways with founder with other founders with customers uh with people they're trying to hire um with investors so that's something that I I look for um you know if if they can like come up with new ideas quickly and improve on ideas quickly.
Transcript
why do you think people don't take the moonshot? I have like a bunch of theories on this, but many of them boil down to like people want to stay part of the herd and it's like more comfortable to not do something unusual and maybe fail or maybe look wrong or maybe look dumb. And and there's like a real It's easy to point to the evolutionary biology reasons of why this would be true. like you don't want to get kicked out of the tribe or whatever. Um, but I think people have like really strong pressures to not to not do that kind of stuff just because of the way society is built. I mean, even more than that, I think society is built the way it is because of like evolutionary pressure and and I think that's probably in like very deep for most of us.
Transcript
what's the cost of becoming Sam Alman? Becoming this person who is probably shaping the most important technology of mankind. H it's hard to answer that cuz like subjectively it's just like this is my life now. You know I I I gave you like the trivia example earlier of I I can't go work in a coffee shop on the weekends. Um, but I feel like so incredibly lucky to get to do all this that although it's like tiring and stressful and, you know, a lot of people's anxiety gets projected onto me and there's like all of these just sort of like life trade-offs, this is like all I know now. Everything else is this like distant memory of this past thing. So I don't like I don't have this like real like suffering.
Transcript
I was just like you know I was like I had a very easy life at the time. I had a lot of hobbies. You know I was like free on a random Tuesday night for dinner on short notice with my friends probably. Um, and you know, now I'm like incredibly precisely scheduled and I have not gotten to do anything spontaneous in a long time. And that's very sad. And what was that one moment where you were like, "Okay, now I can't be carefree anymore." Uh, where you got this realization? People like another version of this is people like ask like, "What do you really miss about like your old life?" And the last time someone asked me that, the thing that I realized I really miss is like going to sit in a coffee shop on a Saturday afternoon with like a nice big cup of coffee and my laptop and just like watching the world do its thing and like catching up slowly on email, but mostly just scrolling the internet. And I have not gotten to do that in years.
Transcript
What's the most extraordinary thing you want to build next? Oh, man. Um, the thing that immediately came to mind is I am extremely excited to see how far we can push AI to discover new science. Um, math the math results were quite amazing to me and it really made me think that we are closer to this milestone than I expected us to be. But a machine that can help us understand the world better, discover, you know, the deepest insights in physics, figure out new medical treatments, uh new source of energy, things that I think we can't even imagine but that better scientific understanding will enable. Um I am extremely excited for us to go build that that machine. I I have always been struck by how much of the world getting better is better scientific understanding and the fact that I think we can now really have AI accelerate that is going to be transformative for for humanity and that's something that you want to build.
Transcript
after that essay during that May you wrote you wrote to Elon about it and you said something that it would be good if someone else apart from Google started it right. So you said like the biggest threat for continued human existence the biggest threat is going to be Smi right. Why do you think that if you started it makes less of a threat? Well, I believe in pluralism. I don't think I don't think like only us having it is good either. Um the there are like I think there are two kind of cartoonish ways this goes really wrong. Okay. Um, one is that there's like a loss of control to AI and we intentionally or accidentally sort of seed control of the future to systems that are unbelievably more capable than we are and we can't get it back. Um, and the other is like one person or company or country gets way too powerful relative to everybody else.
Transcript
when we started, we said we're going to try to build AGI and this is the direction that we're going to push in to try to get there. But the main thing is that like we have a bet on AGI and if our initial direction turns out to be wrong, we'll change. There were not many other efforts in the world that were still willing to say they were trying to build AGI. I think even Deep Mind at that point had stopped saying they were trying to do it. And so although most people thought we were crazy and didn't want to be associated with us, we were effectively able to concentrate all of the really talented people that also believed in this. And as I've looked at some of the other kind of great science and innovation projects in recent history, many of them have had that in common. Interesting. So you just pick up like incredibly far-fetched big problem and be very clear about it that you're going at it. That's how you build a great culture company. Maybe far-fetched is not what I would pick, but like pick pick a problem that is like point at something that some really talented people really want to work on. Be very clear and focused that that's what you're trying to do and don't hedge.
Transcript
You know, you'd look at the mistakes that these neural networks were making and you would say there's no way that something so dumb can just fix all of its problems by scaling it up. There's it just it seemed like all these smart people were saying you have to have symbolic representations or you have to do this or I can give you this very good reason. I remember someone articulating why our approach was never going to work and it sounded so convincing. What was it? Um it was like it went something like all right what these models are doing is they are learning to predict from what's in their training set what's likely to come next. But that means they're never going to be able to ex make new knowledge because they can only model what's in their training set. And like that sounds very you know reasonable.
Transcript
wishful thinking um but you can articulate why it's a good bet you should cherish those moments like that's awesome as a builder that's when you should be the most excited. How do you decide that? How do you find a non-conensus bet? Uh good question. I I like my own experience of this was this was one thing that just no one ever taught me. And you know like very little of my experience in school was figuring out what questions to ask. Um you figure to like answer other people's questions and then at some point I think like when you become a PhD student or whatever you get to start asking more questions and then eventually you do more and more of that. Um, I I think we should focus much more on teaching people, teaching very young people to figure out what questions to ask and what things are interesting to explore that we don't already know. And I think getting good at that skill was the best way that I figured out the sort of non-conensus bets to take. Just figuring out how to ask good questions. Figuring out how to ask good questions.
Transcript
It does feel to me like it's gone a little far. Interesting. What do you think is the biggest misunderstanding about you in the world? Well, I I mean, you touched on one of them, but like the I think not taking not having equity and open eye has been a mistake. I did it for what I thought were good reasons. I think it's just been hugely confusing to people. I like it's it's kind of like interesting to me, right? is like if a if an artist works super hard on their art project or a sports star like practices super hard and doesn't make a ton of money from it, no one has like a problem with that. That's kind of understood and this is like my equivalent of that. But yeah, like it's it's been Why is it misunderstood? Um what do people not like what do they not get about it? I think it's just very outside of the norm of why people assume that people start companies or, you know, work as business people.
Transcript
I I think what some people take that to mean is don't introspect. Just like be sure in what you're going to be sure and don't question it. I try to like question everything all the time. Uh, I still try to like act consistently and with conviction until I find a reason to change my mind. Um, but I think this is really important. Give me an example. How do you do it? My process often is just to keep thinking about the same question from every different angle, throwing out a bunch of bad ideas, trying to like break them apart or find problems with them in my own head, then going to other people. Um, taking more input, trying a different version of it out, adapting it again. I think it's like an iterative process like that.
Transcript
they really helped there. Interesting. But you must be doing something right. You said not sure about how to run the company but you're doing something right about the company because when you were fired there were 700 people who threatened to resign. What do you think you did right as a founder that 700 people like almost the entire company is ready to back you? Um I don't think it's any like one thing but and I don't think this is I don't think it's even mostly about me. I mean maybe the thing that I I did is I got to like you know pick the leadership team and sort of set up how we what we run at and how we how we work. But uh I think the thing that we have done most right is we've just built an extraordinary collection of people really motivated by the same mission um with a very healthy culture for the right combination of being sure and high conviction where we need to be that and being open-minded and exploratory everywhere else. And that that that is a special culture.
Transcript
Something like that. But we were we were sitting on it and we were looking at it and we knew it was extraordinary. But I don't it took us a little while to realize just like the new avenues of research this was going to open up and how quickly the models were likely to improve post GPT4 and that we could do what became the chain of thought work and the reasoning models. And we then decided we were going to go make a very big bet on compute because we saw that we could continue to scale models and that the value of these models would be really important. But it was new and the rest of the world wasn't there yet.
Transcript
Do you get nervous usually? You don't get nervous, right? I was reading somewhere. Um, I mean, I have gotten nervous before, but now stuff has just gotten so crazy. Feel like my activation energy is very high. But now, do you get nervous at all or no? Yes, it's just rare. Like it's that's a like the threshold for it is high. Nice. So, tell me it's Tell me more. When do you get nervous and when do you not? It's a good thing. The last time I remember it uh was when we were about to announce the Navier Stokes result. I was just like this is like such a crazy I mean this is like the real deal. This is like a you know some of the hardest things that are harder than something that people have been able to do it and AI is now doing and I was just like this feels like a crazy moment and I was I was definitely nervous about that one.
