YouSaid · the spoken record

Eric Steinberger

lines on the record: 42
first: 2024-08-30
most recent: 2024-08-30
sittings or episodes: 1
sources: podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

“You and yeah, I mean, first and foremost, thank you again for supporting Magic, and thank you for giving me the opportunity to speak here.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“It that we will not view it as a one to one integration, that we will view it as all these tools are being adapted for AI the way websites had optimizations made for Google search, crawlers. I think there will be a tool optimizations made for AI. And for those that don't have it, the models will just use it natively. And the agent, the model, is the main thing that matters and everything else will get solved for you by other companies. It's the same way how this army of rapper companies is going to get swallowed by AGI companies doing their own agent stuff. The same thing is going to happen there. If you build your own tools, it's going to get swallowed by simply the model learning to adopt everyone. This is just, I just like to think in the end point. So I think the end point is the model users thinks the way a human does. And then maybe also more.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Great question. Very good question. I changed my mind on this like four times. Again, I also just like to read nobody has a real things. So it really just matters what the market wants in terms of product. We'll just do whatever the market wants. Our current state of belief is that you want the system to behave like an employee. So it uses the tool set that you give to your staff members in an interface that is either the same or specifically crafted for AI to be better, then a human could use a tool. For example, Grafana login ingestion. I'm sure we can come up with better ways for AI to use this than humans are using it. So maybe I anticipate that companies will integrate into magic and maybe others, hopefully, I'm guessing there'll be competition once this is a thing. But I think it'll be such that systems will be, AI systems will be on a level with the human and tools will be below.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Think about anything else. This is just the responsibility. And people look back in history and all these questions are answered, and that's amazing. But great that this stuff did go wrong, you know?”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Because I can't come up with a mediocre AGI future. It doesn't exist. I don't really get all the stuff we need, and we find a great way to lymph together and not use it as a weapon. And then everyone has all the things. Nobody is starving. And we all have infinite computer stuff. We find new meaning. And we're not dead. That's a good start. That world is amazing. And so all the failure modes are terrible. So really, I'm sorry, it's just the only thing I can think about is the bimodal nature of this distribution that we're rushing into. And really what's happening is that this smooth distribution of this cloud of uncertainty is slowly collapsing in this bimodal thing. And everyone is sort of going to progressively understand this more. And we just as sitting in a chair that I'm sitting in, I just don't feel like I have the right.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“I'm curious about a number of things, but it's just not the North Star, it's a nice side effect that I think is just, I mean, in a way, it is the North Star, right? Like, what are we building? We're building this automation engine that can answer all our questions. But in a way, I think this is just going to happen. So you don't need to try. The thing you need to try is to make it go well. And if you make it happen and go well, all these questions are just going to get answered. Like this is a side effect. Someone's going to ask, oh, what's three? I'm going to look it up. I'm going to try to understand the proof. I'm going to fail because I'm not smart enough, but I'm going to try and this will be interesting. I'm going to spend like a few weeks on it. And it's going to be fun. But that's just not my North Star at least. I just want everything to be fine in 30 years. If it is, the world would be amazing. Because all the ways in which it could not be amazing are terrible. So if we simply keep it not terrible, I think it will be amazing.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“My honest reply is that I think all of these questions are going to get answered. And my North Star, at least personally, and I think most is true for most of the company at least, is that I just want World to be in a good place in 30 years. And after all this is done and thus said, and this is the past, and we talk about it the way we talk about mobile phones. I just want the world to be in a good place. This is the largest transition we have ever faced. Work will get automated in this crazy. We'll have to find new ways of finding meaning, and that's crazy. The economy will be like, I don't even know. Governments are going to have to figure that out”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“We won't be able to contribute to society as much because the work I feel like at least I can speak for myself like I'm doing I feel tremendously is fulfilling and meaningful and that's going to be deleted as much as it sucks to say and hear this I just it is going to be deleted and then and then my ability to be competitive is going to be deleted because you know deep pollutes everyone in chess and like hopefully magic or summing AI system will beat everyone at coding and then like what am I you know and then like there'll be some CEO system and then like”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So just do your thing and hold through if you don't get automated. And maybe you do, actually, but your company doesn't. So that might be one way this could go. I think games will be huge. People who are competitive, like all three of us, I would guess, will be deeply frustrated by the fact that they can no longer fulfill their desire for competitive interaction through work. Look, I care much more about the positive outcome of what I do than I care about winning personally. But this is a hell of a lot of fun. Like I love what I do every day. I get to build AGI and be, holy shit. And like, I mean, isn't it, it's great to compete against other competent people. If it wasn't in such a serious environment, I would just be enjoying it. Now I have to be conditionally enjoying it. Enjoying it and it's going to go away. So weirdly, I think we're the ones who aren't the most on a meaning level. Because like...”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“The same, but it's not made by the human, you know. So that's a thing, I think. Like, that will grow really big. Whoever owns Etsy, I don't know, but you will get rich. Josh, are you listening?”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“The reality is I really truly believe that capitalism and competition are the only chance we have to provide an optimizer that is capable of getting us to the right place. I think we need to write guardrails to do that. Not stupid guardrails. I'm not saying any guardrails. I'm saying the right guardrails. And then by the end of it, I'll work on a computer at least and probably like robot factory stuff. I know less about that. I haven't run the cost structure, but probably that too. I just don't know the cost structure will be automated. And humans will do other things. I don't think they'll do, I don't think we'll be required to do work for financial gain, but we will probably be able to. Property will be a thing. I think if you own apartments, that won't go away. There's Etsy. Etsy is a great proof of concept for what happens after AGI. This is completely useless. Like you could just buy the made in China product. It looks”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So, I think there are like 10,000 possible futures. And which one we end up with depends entirely on which answer we choose or whether we fight the sensible middle ground and we manage to have a rational debate.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“And this is just one example. Centralizing power is terrible, so therefore you should open source everything. If I stop now, this is reasonable, right? Please don't make a quote out of this because I don't believe this. And then you could say, well, I should not, I should learn how to do PR. I should not say these sentences. Anyway, you can keep this in. So there's one way to say this. And then the other way to say this is, well, this is like nukes. We all fear this existential risk thing. Like maybe we care less about the, you know, or we care, it's not that we care less. It's just that we think these intermediate problems are completely solvable. But you can't open source how to build a nuke. This is just terrible. So the problem is that both of these things are true. And the problem is that there are 10 questions like this. And both of the answers are true and all of them. And so what you get is people arguing on X, claiming one side and ignoring the other.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“It's going to take longer. I'm going to try to compress it, but it's actually quite complicated. One of the biggest problems with this question is that everyone tries to simplify it by picking one side of the argument. For example,”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Just do our thing. We have our plan, and we have brilliant people who I'm delighted have trusted us and spent their energy and their best years on our company.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“And he didn't say this in an arrogant way. He just said, This is a ton of work. I've done this for the last few years, and I just need to be an environment where I am sufficiently challenged to do this. And he's been grinding every day. And just having that level of ambition, but not with the typical San Francisco. But it's quiet with humility drive. You just come into the office every day. You're not working until 3 a.m. because you proved that you look like this is our culture where to grow. You do it sometimes because you're just so obsessed and you try to be healthy. You do work all the time because you just care so much, but we're not buying the IP by poaching someone from like a lab who tells us how they train GPT. Never done this. We'll not do it.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Brain power and understanding of the world to comprehend that this is the right thing to do and or a lot of trust, to trust an organization with doing that in the first place. And everyone cares deeply about this. But we don't do it for the, you know, here is like your marketing sign or whatever. We don't have the whole we are. At the same time, everyone is deeply productive. They all, you know, when one of our primary, I go one of our core engineers who writes the inference engineer, it's one of the two people writing the inference engine and Colonel Nelson. When he joined, I was like, why do you want to join? What do you want to do? And this was before we raised this giant stack of cache that's getting an else now. It was like the tiny amount still compared to other labs, or I guess compared to any lab by a large margin. He was just like, well, you know, he saw this as an opportunity to, he wanted to be one of the best, he said he wanted to try to be the best kernel engineer outside of NVIDIA.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“And so that's the type of person we have, and we have a decent number of them now. We've gotten really good at identifying them. I would like to have roughly four times as many as we could. But that said, again, I think with the series of announcements that's going out in the batch that this podcast is going live, that again will get easier and better. But I love our culture. It's just... Everyone cares about the mission deeply. What I said earlier about why we do what we do, that safety bound at AGI recursion, just takes a lot of”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So now this is really easy because we've raised an unbelievable amount of money from great people and we've got things to show that even risk averse people who don't think from first principles can understand that this makes sense or who need that initial seat of trust. But I did find it very hard to recruit when we got started to be honest because when Dario Amode goes out and starts a company, this is trivial. Hi, I made GPT3. And when there are others like this, it's easy to establish that trust in the outcome. So the strategy we adopted at the start was to hire actually kind of people that like you do once you're referencing, people who might be, we have one guy who is just like depressed that Amazon.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“No, I think if you can, the leap to doing all use cases is small Like you can build a UI builder, and then it's like a normal UI builder, or you can build a true great UI builder driven by AI with some added features for that vertical. But then you can do the same thing for all the other verticals. Just add the features, you know? And so maybe your product team needs to do one by one. That's feasible. That's totally imaginable. And maybe your go-to-market needs to be one by one. But the model, I don't think so.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So that's like the fully automation thing, right? And then you can launch something where you're like, I'm going to do code review, but it's not frustrating. If you have to do code review and it's really taxing and you have to fix half of the problems, like half of the PRs or whatever. So I think the bar for this product is just high. It's not that we are so ambitious and I dig that too. I just genuinely think that there is a gigantic market that gets unlocked in a step function moment where users decide that they're no longer going to use VS Code to write code and send it to their colleagues. They're going to use magic or whoever ends up doing this well first to write their code for them and then briefly look at it and correct every now and then what has been done and then eventually not, right? But that is a step function moment, I think. It's not like you're not going to use this for like 5% of your tasks.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Do you just trust it? My engineers, when I have one of them write a piece of code and another one review it, like looking at it, why would I look at it? It went through these two guys. So what's the point?”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Have they asked this right now? Like, there is no amazing can do everything. It just feels like a Frugenius colleague on your team. But if we don't hit it by that time, we'll launch the other thing. But I would prefer hitting it. It's just the honest, you know, the reality is. Things are hard, some projects are going great, and some are delayed, and some are just, you know, there are a hundred fires all the time. This is just how every hard engineering project goes. Everyone who's listening to this and has ever worked on an engineering project was like, this is just how it goes. I think we, you know, there are a lot of things we learned along the way. There's nothing we're stuck on. It's just things are god, there's this thing we didn't think about. Okay, let's fix it. So I feel very optimistic”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“You can see signs of life, but you guys tried it when you decided to invest. So it's like, would you be using this to write your no, not yet? But we can train the next model. And then that model can do it. But then that model can also do all these other things that would go into Vincent. So you just enter this stupid recursive loop. Until the point, we got together to the point where we're like, okay, what's the final UX? Let's just make sure this never happens again. We're trying to meet the bar of that UX now. Which I hope we will sooner rather than later. The closer you get to it, the sort of dumber it feels to launch the thing before it, because you're going to replace it. It's going to a few months. So if it's good enough for that, how hard can it be to add these last level few things? So I do think there is a difference between like you can launch an extremely capable assistant before you launch full automation. That's fine. But launching a sort of mediocrely capable assistant, we might do it. We have a deadline eternally by which we don't.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“As we have built the UX, each iteration of the next UX internally and thought about launching it, we were at this interesting stage of it feels like an Uncanny Valley where Like, well, you know, clearly you can see signs of life. First of all, I mean, completion is a trivial one, right? We decided in auto launch completions because it's just obviously going to get killed by the next thing. And then we're like, okay, it's going to take us a few months to get a prototype of the next thing. We got a prototype of the next thing. Looking at this, it was like,”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“That's interesting, right? Because there is enough compute for search for Google search. I don't think if you took Gemini 1.5 Pro and you put it into Google and you did all the fine-tuning properly and then you swap it with Gemini 4, I don't think anyone would notice unless you go like prove Riemann, right? But it's good, like 99% of users use cases will never notice the difference. So this does the job. If AI overviews work snap. I think this is going to similarly be the case for each thing. There's going to be a model that can prove Riemann. You can make it 100 times smarter. You have your proof, right? So for each thing, there's going to be improvements after a certain amount of compute. And I think I underestimated that number. And I underestimated how much others would focus on code. So anyway, just to clear up, I think our 101, you were right.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Think a lot of it is direct competition. Think that, like, there are totally things I could say here that sound entirely believable as major differentiators, but I think fundamentally are, and I'm going to talk about them, but I just also want to highlight that I think at the core, the wider world has understood that code is very helpful and there are ways to deploy compute to improve coding performance. And so therefore, because a lot of compute is deployed in this domain, the need for compute is large. That said, also in parallel with the release of this, we're going to be announcing what is going to be one of the largest clusters to ever be built. So we are correcting for that. I was wrong when we initially talked. You just definitely need a lot of compute. Now there's still, I think, a notion of enough, but it might be a lot.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Product that's a new generation of thing that just doesn't really exist yet. It just, there is no thing that does your work for you. I think ChatGPT was one of these monumental moments of a new type of thing. The first AI assistant, this is like mind-blowing to all of humanity. And this will happen once more. And then maybe once more when you can just query the solution to Riemann. And then that's probably it, at least in this domain. And so I hope we can be a part of the second one, maybe the third one. And on the product side, but we chose the domain we chose because of what I said previously.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Of our capital to do it is to say, like, look, if this type of thing works, it's not really an assistant. It's just like you just put it there and you talk to it and it's like a colleague. And that's great. I think the economy does the same thing, same outputs with less input or even more output with less input. This is fantastic for the world if we just do it well. Capitalism and competition can be great and progress. The entire history of progress comes from this. We'd all be farmers if we weren't in favor of automation. It's scary, I get it. And if you don't think about it all the whole way, it's very scary, especially if you're affected and you're not in one of the lucky seats I'm in, for example. But ultimately it's good. So yeah, I just basically like when we started tried my best to think through this and this seemed like the most productive path forward. And again, I think just like we're lucky to be in a position where we both get to pursue an incredibly valuable”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Thousand brilliant researchers in your computer is better than having 100, especially if you can connect all their brains. And then if you can say, okay, we think this is safe to do, like let's go do it. And then you have the next thing. And then you're like, okay, like maybe we should use that thing to do some more safety research. You can bootstrap. And so just very pragmatically, even if there was goodwill, I just don't know that we have the mechanisms to prioritize safety without automation. So yeah, that's the primary goal. Now the good thing is as we pursue this, if we're successful, if we're not successful, then I'm deeply sorry to all of our investors, including you. But if we do succeed, this will, especially if we are first or early to succeed, drop a beautifully profitable apple of a tree that happens to automate a good chunk of what we call work today, which is another thing that I think is a responsibility.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Like the sort of recursive approach that you're mentioning here and hinting at, and then obviously, you know, this is what we're founded to do, is exactly that, I think the only way to sort of reasonably approach this is to iteratively ask your model to solve alignment and safety at that stage. Surely you can also ask it to solve your product level problems, but that's nice, but that's not the fundamental objective. The fundamental objective is to iterate towards AGI with a safety boundary. And there's just no knob like this. There is just no knob like this in the human world. You can say, oh, I'm going to spend X percent of my resources on this, but that doesn't indicate an outcome. You can actually control it. But if it's compute, you can somewhat do it. So I think it's just the most promising path we have, both on the progress towards better models, because if you just have 100”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, that's a great question. So I remember one of our first conversations. We were talking about sort of the step function relevance of safety risks, let's say, right, where like there's a lot of stuff people are panicking about that really doesn't matter in the short term for the grand scheme of society. And like some people will get pissed from me saying this, but it's just what I actually believe. I just don't think this is like the current complaints are similar to what we have seen in all other technologies and totally resolvable. But then there comes the evolutionary one. And I was like, okay, shit. Humans succeeded in the world because we're smarter than chimps and we're smarter than bunnies. And, you know, bunnies do not rule the world. And that would be cute probably, but also it probably wouldn't be as nice for us. And here we are turning ourselves into bunnies and apes and creating this thing that, you know, we're all thinking is going to be waste-minded and we are, this is insane. So the reason I think”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Simply because it is a shortcut. Like, you can also train this model for a quadrillion dollars. Maybe you can't actually because there's no data. But say you could. Well, what if I just need a billion? You get the idea. So I think it's like this fundamental trade-off that you want to be able to bake into, and the humans can do this exactly as you say. And this applies to all parts of the workflow that you ask a model to do.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“And there are things you have to learn during training. Like if you are asked to write a piece of code and you've never learned coding, you can spend the inference time compute you have to spend is ridiculous. Like you're going to have to, at inference time, learn programming, which maybe I actually think this is possible. But it is also crazy. And this is clearly this is shared among everyone who posts a query. So it's stupid not to make this into training. Even the best mathematicians in the world for the frontier of mathematics require a long time to solve the problem. So I would love to have a Terence Tao in my computer, but I would then still need to run Terence Tau for a year of human thinking time. And Terence Tau will not just token by tokens spit out a proof for Riemann or whatever. So I think to achieve things like that through pure training compute deployments, inference can compute work would lead to drastically fail.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Curve in between. It seems strictly beneficial to be able to provide that choice. So instead of training putting all the computer to training and having that $1 million inference performance be purely from the training compute, which is just hilariously inefficient, you can allow the user to choose their thing. Or rather, you can just deploy multiple things. The reason I could talk about this now is because everyone gets this. But basically, you clearly want to be able to regulate the amount of compute you server. Now, it turns out this is actually not trivial. Like doing this is hard, like finding the right algorithms to do it. That being said, people have done it in RL for a decade. And so to those, you know, I don't want to name people. I don't know him obviously as one of them, but there are others in other labs as well. There's a set of people to whom this is the opposite of a surprise, but it's still not trivial because it's the general domain. There is no game”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So you can think of model performance as some function of training compute times some function of inference time compute. Now those are specific functions that are just scaling law things that you can model, but the general way to think about it is some function and some function. And then you would want to Estimate how much inference you're going to do and what your total budget is, and then you would want to create the optimal trade off in your allocation of money. You also want to consider the distribution of outputs. There will be users who will want to spend less money.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, I mean, so you okay comment this from two perspectives. I can explain it mechanistically, and I can, the other cheap way is to just point that Richard Sutton's bitter lesson, retrieve also, like, say, subset of data for one completion, our model sees all the data all the time. Clearly, a subset of data for the whole completion is a subset of all the data all the time. So if retrieval was optimal, our system could learn it. And it just turns out that it's not optimal. That's the sort of mechanistic, I guess, explanation or logical explanation as to why non context will be better. You could make arguments around the quality of long context if it weren't sufficient, if it wasn't sufficiently high quality, maybe having a short context window and pulling in some data. So you obviously have to evaluate this. But in principle, yeah, on the assumption of Richard Sutton's better lesson, you would want the thing that can learn your heuristic rather than a heuristic.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“So I'm sealing this quote here instead of bringing the data to the compute, we're bringing the compute to the data. So you have a set of stuff and our model acts on that stuff rather than having a giant model that you have to sort of work around. The whole system is designed for this. So yeah, ago, we announced right million. By the time this is out, we may have announced a larger number. The domain reason being that you would want to deploy these things. For a very long horizon trajectories, and you want them to spend a lot of time thinking, and you want the model to remember all of that. And you can't really do that by fine-tuning because you'd have to fine-tune every like whatever many thousand tokens your contact spend is law.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Thank you. Yeah, it seems important for models to have the ability to learn from long histories of their own and their collaborators' actions, as well as take into account a large amount of fast-changing data. And so if you imagine having 10,000 employees or everyone on earth having their own model and wanting to feed in all their data, you can now fine-tune everyone's model, maybe do some Laura tricks. But in practice, context just works better. And that's like in context learning is the magical part that came out of Transformers. This is what makes them great. I think of that as some sort of as an online optimizer in a sense that instead of compressing a set of data, you're trying to learn an optimizer. So the perspective we take on models is instead of, this is one of my colleagues put it this way. I find it very”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“It sort of came from a place of working backwards from AGL. If your end goal is to have a system that can do everything, you can reduce that to building a system that can build that system. And so that minimal system is a system that writes code and comes up with ideas and can validate those by writing code and running experiments, which is still in the same order of complexity as the full thing, but at least we don't have to train Sora. And we don't have to think about 10 billion other use cases that everyone building general domain products has to think about. We only have to think about code. So it's a lot simpler in all aspects except compute and slightly simpler on the aspect and slightly cheaper on the aspect of compute. I think it's not a lot cheaper. I probably overestimated how much cheaper it would get on the compute site at the beginning. But the other things are simpler, I think.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Was like, could you do a mini PhD thing where I'm a complete newbie? But if you can bash me, just please bash me every two weeks and tell me how to be a good researcher. And so eventually I got like reasonable and then did some actual research work and worked with a few other people, including Non Brown, who had meta on developing new RL algorithms to be more sample efficient and just generally better and faster or whatever was the goal at the time to solve whatever environments we were interested in at the time. So yeah, that's how I got into it. I have no background in language models when we started magic at all. It just seemed, I just was totally not on my radar. I was like, oh, wait a second. Like if you take this and this and put it together, like maybe this works. And so then I sort of, it felt like this huge relief of uncertainty relief of where GI would come from because you just put those two things together and then it will work with the sort of hope.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, thank you. So I guess when I was 14 Had my midlife process and thought I had to do something important with my life and spent a year trying to look at everything. I was pretty stupid. And basically I'd look at things like string theory and like all the things a 14 year old would look at. I mean, like, okay, what can I spend my life on? And eventually my mom got me a book on AI and I didn't read it. I'm sorry. But it was like the idea was sufficient. So it's like, okay, this could do anything. And so you should just do that. And then it does everything. Then it seemed plausible that you'd need to do reinforcement learning. So I didn't know how to code at the time, then learn to code over a couple years. This was sort of in high school times. And then it seemed plausible that you'd need to do reinforcement learning because otherwise you'd not be unbounded. So I sort of just started working on RL, played around with things for a bit, and eventually reached out to someone at DeepMind to basically pitching this multi-page email.”
2024-08-30 · No Priors · Building toward a bright post-AGI future with Eric Steinberger from Magic.dev · IDENTIFIED FROM THE TRANSCRIPT