No Priors Ep. 71: The Best of 2024 (so far) with Sarah Guo and Elad Gil

1. 0:00 – 0:46

   Introduction

   1. SGSarah Guo0:00

      (digital music) Hi, listeners. Welcome back to No Priors. We're halfway through 2024, so we're doing a mid-year best of episode, where we go back to some of our favorite moments from episodes so far, and catch you up on everything that's been going on in AI, from the state-of-the-art in research to hyper scalers and upstarts. We'll list all the episodes featured so you can go back and re-listen to the whole conversation. To kick it off, we're gonna hear a little bit from Emily Glassberg Sands, who's the head of information at Stripe. We talked a lot about how AI can help small businesses make a big impact in the economy. Here, she talks about the intersection of fintech and AI. (upbeat music)

   2. EGElad Gil0:45

      When you think

2. 0:46 – 4:23

   Emily Glassberg Sands on the Future of AI and Fintech

   1. EGElad Gil0:46

      forward on the w- the directions that the overall financial services industry is going, and let's put Stripe aside for a second, because I think Stripe, um, is obviously a core company to sort of the internet economy and it touches so many different pieces of fintech and things like that. But where do you think, outside of Stripe, the biggest white space for fintechs employing AI is? Like, from a startup perspective or even an incumbent perspective, like where do you think this sort of technology will have the biggest impact?

   2. ESEmily Glassberg Sands1:13

      It's a great question. Um, I don't know exactly what others will do. I think, um, having a really robust understanding of identity, who businesses are, what they're selling, um, has always been important. And you know, I think often in industry, we think it's important for marketing or sales or sort of go to market motions. But it's also super important in fintech. Um, yeah, it's important for credit lending decisions, but it's also important for supportability, um, decisions and understanding, uh, where, you know, the business does or does not meet the requirements of, uh, a given card network, uh, or a given bin sponsor. Um, and so I think that that identity piece, like who is this merchant, are they who they say they are, um, but also what are they ... what's their business, what are they selling, and how does that map to this pretty complicated regulatory environment, um, is a really interesting and hard problem that lots of folks are solving in their own ways, but, uh, is, is likely, um, an opportunity. I think there's almost certainly an opportunity to, you know, whether Stripe does it or somebody else does it, to make, um, sort of financial integrations way more seamless. Um, Stripe has a whole suite of no code products so you can use, uh, you know, payment links or no code invoicing. But how does one actually build a, a really robust, um, specific to the user integration without needing, um, you know, a, a substantial number of payments engineers or, um, any complicated developer work? LLMs are proving that they can be very good at writing code. Um, we have a couple cases actually where we're already seeing it work. But as the, as the decisions get more and more complicated, I think there's still a lot of work to do, um, to build the right integration and to build it well, um, in an automated way. And then I think, as I mentioned before, some of this layer on top of the payments data of like, okay, you could build solutions that make payments work better, but payments actually allows you to really deeply understand and improve the business is, is pretty fascinating. And you'd have to think about like, is it a startup that does that or is it an incumbent that does that and what's the, what's the business model? Um, what's the business model there? But you know, if I think about the case of Stripe, um, you know, sort of Stripe has the opportunity to be beneficent, right? Incentives are super aligned. The more Stripe can help its users, businesses grow, the more Stripe grows and the more the economy grows. And so whether it's Stripe or someone else using financial data to help businesses be more successful, to grow the pie, to grow the GDP, um, I think is, is really powerful.

   3. SGSarah Guo4:22

      Up next,

3. 4:23 – 9:03

   Dylan Field on AI and Human Creative Potential

   1. SGSarah Guo4:23

      we have a clip from our conversation with friend and formidable founder Dylan Field, whose company is using AI to change the design process and bridge the gap between design and development. We talk about how bringing AI into the creative process changes the creative job. (upbeat music)

   2. EGElad Gil4:39

      Basically, you're moving from a human to human collaboration company to a human to AI collaboration company over time in some sense, 'cause, you know, what you're describing seems like a really interesting way to have copilots, um, augment humanity or augment creativity. Are there other ways that you've thought about the substantiation of that sort of creativity augmentation or how AI really interacts with human creative potential?

   3. DFDylan Field5:05

      Well, a- and these are just examples of things that I've seen or, or thought about that I think could be cool in the creative space because you asked about. But I think in the design context, um, one thing that really matters a lot is the iterative loop and being able to keep going back and forth, uh, to an agent and give more instructions over time. If you just kind of like go to first principles here, there's so much that you're not able to communicate via a prompt. Like, if you think about great design, it often captures something about the culture, the ethos of the moment. It captures, uh, something about the temporal aspect of the sequence of interaction someone's having or the, the context they will have mentally, something about affordances, what people are used to in terms of the language of design, uh, which is sometimes, uh, similar and, uh, dependent on the platform, you know. But oftentimes, there's something about emotional state too.... uh, there's, you know, videos that the designer's probably watched or, or in-person research interviews they've conducted. And so I, I think, like, fitting all that plus the product requirements, um, plus visual style into a prompt, that's hard. Even if you could just get unblocked by an AI helping you brainstorm and thinking through problems, you know, that's your first sort of draft. And from there, you can keep iterating. From there, you can keep evolving things. I think that could be very, very interesting as a, as a first step.

   4. SGSarah Guo6:32

      What's your response to people who worry that, um, AI like in every role are gonna, you know, eliminate the need for designers?

   5. DFDylan Field6:42

      For all the reasons I just mentioned around, you know, emotions, uh, user context, uh, knowing how flows go, um, having that history of interactions and whatnot, I, I think it's unlikely that that's, like, the world we're seeing in the short term. I think no one knows what's happening in the long term. Uh, you know, if we have, you know, superhuman intelligence, like, I don't know what it means for any of us in this, on this call or podcast, uh, (laughs) or any of our, anyone listening. If we don't try to ask about what that case it looks like and instead ask about, okay, if we assume that there's, uh, uh, continued improvement, uh, what does it mean for design? I think design's actually in a really good place. Probably before you see potential replacement of any part of the design role, you instead see augmentation and you see access. Uh, you see efficiency so that designers can get more done, and I think probably a lot of engineers do more of their, put more of their time towards design than they put towards what we consider coding tasks today, and the abstraction level of coding changes. There's probably still a human in the loop for engineering, but I think that it's not, not clear to me that humans are gonna write, like, every line of code in a year, three years, five years. I mean, o- obviously already we have Copilot, but I think that you could go even further than that and, and a lot of companies are trying to do that.

   6. SGSarah Guo8:05

      I mean, I can't make multi-year bets in the current environment, but my expectation would be that we, um, maybe it's because I'm an optimist, but I think we're just gonna get better and more software, um, and better-designed software versus fewer, fewer designers or engineers.

   7. DFDylan Field8:23

      Yeah, I definitely think that as a metric, like, number of pieces of software that will be created will go up tremendously! Um, and, and it's interesting, like, there's, there's some visions out there of the future where, uh, people interpret the capabilities of AI to mean that you won't, like, have any interface at all. I think it's really cool to see this explored like the rabbit we've talked about, Sara. I haven't used AI, I think you did. Is that right? But...

   8. SGSarah Guo8:49

      Yeah.

   9. DFDylan Field8:50

      ... I think it's a really cool vision, and I think that there will be so much more software in a year, two years, or five years from now than there is today. Like, both could be true that there's demand for that, and there's just way more software.

4. 9:03 – 12:43

   Brett Adcock on Running Figure AI’s Hardware and Software Processes

   1. DFDylan Field9:03

   2. SGSarah Guo9:03

      Next, we talk to Brett Adcock, the CEO of Figure AI. Figure is creating a fleet of humanoid robots to take on the dull and dangerous jobs that humans shouldn't be doing. In this clip, we talk to Brett about how he runs a team with velocity to drive hardware, software, and AI into reality. (instrumental music plays) Big question, but can you describe, like, if you want to run a hardware project, a hardware and software project like this with this complexity at velocity, like, how do you manage product development?

   3. BABrett Adcock9:35

      From, like, a thesis perspective, I, I strongly believe in, like, an iterative design approach. We really don't believe on l- spending a lot of time, like, just, just doing research and analyzing. We spend a lot of time on just testing, bui- building and testing...

   4. SGSarah Guo9:48

      Mm-hmm.

   5. BABrett Adcock9:49

      ... uh, here, and, um, that helps us really shake out all the problems. It helps us learn. It helps us recursively add it into a continuum of product that's coming down, uh, coming out. And, um, so at first, that's our strategy. We, um, we want to be continuously updating the hardware and software forever. It'll- I don't think it'll ever be good enough for us. Um, so we have a whole process built around building a robot from a des- like a basically hardware and software design that we run here. We first set out with understanding who are the customers, like what does the robot need to do? From there, we, uh, we basically set requirements, like, okay, we need the robot to lift this much pounds. It needs to run this long. It needs to charge here. The safety requirements are that it can't, battery can't burn down. The building, there's a bunch of stuff we have to, um, the environment on IP rating has to be done on all the actuators. There's just a bunch of requirements that come from there. From there, we look at those requirements, and we, we do engineering design, and we have basically, like three big phases. Uh, we have a conceptual and preliminary and critical design review that we do here throughout the year. The whole company's involved. So we, um, have these, like, design gates that we work through, similar practice that I instituted, exactly similar, well, similar practice I instituted at Archer from a, from an engineering design perspective or philosophy. And, um, yeah, we work, we work through in a very methodical way, like, uh, all the way through that serially.

   6. SGSarah Guo11:17

      And how, how does, um, integration and testing work in a way that's different from a software company since you've also done that? I'd imagine-

   7. BABrett Adcock11:23

      We try-

   8. SGSarah Guo11:23

      ... really differently.

   9. BABrett Adcock11:24

      Yeah, we try to, we try to test, and we try to prototype and test as fast as we can to see if we were right.

   10. SGSarah Guo11:30

       Mm-hmm.

   11. BABrett Adcock11:30

       Same with software. It just happens on a longer timeline.

   12. SGSarah Guo11:32

       Okay.

   13. BABrett Adcock11:32

       Well, software, you'll come in one day, um, and you'll, and I'll say, "Okay, um, we talked to the client. We believe the cli- we've, we talked to the client. We believe we have all these things on the product backlog list we wanna do." You'll somehow have some heuristics where you'll score those, and you'll basically comb the backlog, and you'll say, "I'm gonna go... We're, we're gonna add these, like, six things to the sprint." You'll do story points, and you'll basically, you'll, you'll assign those out, and you'll basically manage that whole process. And then you'll launch it...

   14. SGSarah Guo11:57

       Mm-hmm.

   15. BABrett Adcock11:57

       ... and you'll get feedback, right? You'll try to either A/B test things. You'll watch the analytics, and you'll say, "Did that work? Did that work?"

   16. SGSarah Guo12:03

       Mm-hmm.

   17. BABrett Adcock12:03

       Uh, you really wanna do that, and you wanna have a, kind of a scientific method around it saying, "Okay, was that, did that actually..."... help, you know, fix this problem. Uh, same here. We have the client. We have requirements that we set, like they need to do this. We are designing things, say like, we're har- designing hardware from scratch, like in, um ... so we take our ... if we're designing an actuator, we're gonna take our CAD system and we're gonna, from scratch, design it. We're gonna make, uh, assumptions on ... and trade studies on, like, what the different trade-offs are of how we can do it up front so we don't spend a lot of time designing something that just didn't work.

   18. SGSarah Guo12:36

       Mm-hmm.

   19. BABrett Adcock12:36

       Uh, so we can be pretty methodical about it, like much more methodical than you are with software because the timelines are, you know, order of magnitude plus longer.

   20. SGSarah Guo12:42

       Up next

5. 12:43 – 17:43

   OpenAI’s Sora Team on Artists’ Creative Experiences with their Model

   1. SGSarah Guo12:43

      is a snippet from a conversation we had with the OpenAI research team, building Sora. Here, we talk to this team about their generative video model and whether or not video is on the path to AGI. (instrumental music)

   2. EGElad Gil12:59

      Do y'all have a favorite thing that you've seen artists or others use it for, or a favorite video, or something that you found really inspiring? I know that when it launched, a lot of people were really stricken by just how beautiful some of the images were or how striking, how you'd see the shadow of a cat in a pool of water, things like that. But I was just curious what, what you've seen sort of emerge as people, more and more people have started using it.

   3. ARAditya Ramesh13:19

      Yeah, it's been really amazing to see what the artists do with the model because we have our own ideas of some things to try but then people who, for their profession, are making creative content are, like, so creatively brilliant and do such amazing things. So Shy Kids have this really cool video that they made, this short story of, uh, Airhead with, um, this character that has a balloon, and they really, like, made this story. And there, it was really cool to see a way that Sora can unlock and make this story easier for them to tell. And I think there, it's even less about, like, a particular clip or video that Sora made, and more about the story that the- these artists want to tell and are able to share and that Sora can help enable that. So that is really amazing to see.

   4. SGSarah Guo14:07

      You, you mentioned the Tokyo scene.

   5. EGElad Gil14:09

      Yeah.

   6. SGSarah Guo14:09

      Others?

   7. TBTim Brooks14:10

      My personal favorite sample that we've created is, uh, the Bling Zoo. So we ... I posted this on my Twitter, uh, the day we launched Sora, and it's essentially a, a multi-shot scene of a zoo in New York which is also a, a jewelry store. And so you see, like, saber-toothed tigers kind of, like, decked out with bling, and you have-

   8. SGSarah Guo14:28

      It looks very surreal, yeah.

   9. TBTim Brooks14:29

      Yeah, yeah. And so I love those kinds of samples because, as someone who, you know, loves to generate creative content but doesn't really have the skills to do it, it's, like, so easy to go play with this model and to just fire off a bunch of ideas and, uh, get something that's pretty compelling. Like, the time it took to actually generate that in terms of iterating on prompts was, you know, really, like, less than an hour to, like, get something I really loved. Um, so I had so much fun just playing with the model to get something like that out of it, and it's great to see that the artists are also enjoying using the models and getting great content from that.

   10. EGElad Gil14:59

       What do you think is the timeline to broader use of these sorts of models for short films or other things? 'Cause if you look at, for example, the evolution of Pixar, they really started making these Pixar shorts and then a subset of them turned into these longer format movies, and, um, a lot of it had to do with how well could they actually world model even little things like the movement of hair or things like that. And so it's been wa- interesting to watch the evolution of that prior generation of technology, which I now think is 30 years old or something like that.

   11. ARAditya Ramesh15:25

       Mm.

   12. EGElad Gil15:26

       Do you have a prediction on when we'll start to see actual content, either from Sora or from other models, that will be professionally produced and sort of part of the broader media genre?

   13. ARAditya Ramesh15:36

       That's a good question. I, I don't have a prediction on the exact timeline, but, but one thing related to this I'm really interested in is what things other than, like, traditional films people might use this for. I do think that, yeah, maybe over the next couple of years, we'll see people starting to make, like, more and more films. But I think people will also find completely new ways to use these models that are just different from the current media that we're used to, 'cause it's a very different paradigm when you can tell these models kind of what you want them to see and they can respond in a way, and maybe there are just, like, new modes of interacting with content that, like, really creative artists will come up with. So I'm actually, like, most excited for what totally new things people will be doing that's just different from what-

   14. EGElad Gil16:21

       Yeah.

   15. ARAditya Ramesh16:21

       ... we currently have.

   16. EGElad Gil16:22

       That's really interesting because one of the things you mentioned earlier, this is also related world modeling, and I think

   17. BPBill Peebles16:26

       Yeah.

   18. EGElad Gil16:26

       ... that you'd been at OpenAI for something like five years and so you've seen a lot of the evolution of models in the company and what you worked on. And I remember going to the office really early on and it was initially things like robotic arms and it was self-playing games and things, or self-play for games and things like that. Um, as you think about the capabilities of this world simulation model, do you think it'll become a physics engine for simulation where people are, you know, actually simulating, like, wind tunnels? Is it a basis for robotics? Any use there? Is it something else? I'm just sort of curious where some of these other future forward applications that could emerge?

   19. BPBill Peebles16:58

       Yeah. I, I totally think that carrying out simulations in the video model is, is something that we're gonna be able to do, um, in the future at some point. Um, Bill actually has a lot of thoughts about, uh, this sort of thing, so maybe you can ...

   20. TBTim Brooks17:11

       Yeah. I mean, I, I think you hit the nail on the head with applications like robotics. Um, you know, there's so much you learn from video which you don't necessarily get from other modalities, which companies like OpenAI have invested a lot in the past, like language. You know, like the minutiae of, like, how arms and joints move through space. You know, again, getting back to that scene in Tokyo, how those legs are moving and how they're making contact with the ground in a physically accurate way. So you learn so much about the physical world, uh, just from training on raw video that we really believe that it's gonna be essential for, uh, things like physical embodiment moving forward.

   21. SGSarah Guo17:42

       Up

6. 17:43 – 21:06

   Scott Wu Gives Advice for Human Engineers Co-Working with AI

   1. SGSarah Guo17:43

      next, we talk to Scott Wu, the co-founder of Cognition, the company behind Devin, which is building an AI engineer. Here, we talk about the design for Devin and what it means to work with AI engineers. (instrumental music)

   2. EGElad Gil17:57

      What do you think is gonna be important from a human software engineer, or just, like, human technology person, five years from now? I realize that's a really long timescale in AI, um, but it's certainly not, like, encyclopedic knowledge anymore, right?

   3. SWScott Wu18:12

      Yeah. Yeah, and I mean, I think there's, there's, there's a meme that, you know, the hottest new programming language is E- English, right? And I, I mean, I think there's a lot of truth to that. Um, but with that said, I think that, you know, the software engineering fundamentals are obviously still super, super valuable, right? Um, people, you know, um ...For example, like, I think, you know, the internet today is, is something that we all kind of are able to use and kind of take for granted. But people who work with these networks, um, it's certainly very helpful for them to understand the details of, of TCP, right? And I think similarly, I think, um, you know, I, I think we'll be able to communicate our ideas in English and work with all these things. But, you know, understanding the internals of, um, of how computers work and understanding logic gates and, you know, a lot of these core kind of pieces, like these core foundations, I think will still be very useful, right? And so, you know, um, whether that's, um, you know, algorithms or technologies or, um, logical reasoning or, or things like that. Like, I, I think the, you know, I think the role of a software engineer, um, five or 10 years from now looks something like a mix between a technical architect and a product manager today, you know? Where, where a lot of what you do is, you know, you take problems that you're facing or that your business is facing or whatever, and you're really thinking about and breaking down what exactly the solution should be.

   4. EGElad Gil19:29

      How do you think about it, uh, on an even farther timeframe? Because when I ... If, if it was five years ago, I would have told either my kids or people who have kids, um, you know, "You should study computer science and math." 20 years from now, I'm not as certain. So I'm sort of curious how you think about the future of, uh, of this field, if much or all the work, including a lot of the planning, is actually done by machines at some point.

   5. SWScott Wu19:54

      Yeah. Um, I mean, I love math. So I, I have to say, it's, uh, (laughs) it's a worthwhile experience even if it doesn't, uh, (laughs) it doesn't end up being practically useful. But no, I mean, I think these, I, I think a lot of these fundamentals will, will stay useful for a long time. There's obviously a lot of questions that come up about, um, you know, super intelligence and singularity and, and all of this. Um, and you know, it's very hard to predict. Uh, I, I think everyone in AI, it's, you know, we've, we've all made our, our own predictions and, you know, tried to make our guesses, but I, I think it's, it's hard to be very high confidence. Um, but with that said, I do think that, um, you know, we'll, we're gonna see AI's concrete impacts on, on work and the economy and, and people's lives, I think a lot sooner than now. You know, I, I think, um, the, the way that, that we think about the problem is that, uh, you know, even with the tools that are available today and the technologies that exist today, there's so much that's possible to, to really impact people's lives, right? Um, and, um, you know, we're still very, very early in, in this whole AI revolution. I mean, you know, ChatGBT was, um, about a year and a half ago at this point. And there's a lot more to do and, and a lot more to build, you know, both on, on the research side and on the product side.

7. 21:06 – 25:55

   Alexandr Wang on How Quality Data Builds Confidence in AI Systems

   1. SGSarah Guo21:06

      Finally, we have Alex Wang, the founder of Scale AI, the data foundry for AI. Here, we talk about what's next for Scale as models, uh, approach and go beyond human abilities. (instrumental music) With great power comes great responsibility. Um, if, uh, you know, if these AI systems are what we think they are in terms of societal impact, like, trust in those systems is a crucial question. Like, how do you guys think about this as part of your work at Scale?

   2. AWAlexandr Wang21:35

      A lot of what we think about is how do we utilize ... How does a data foundry, um, enhance the entire AI lifecycle, right? And that lifecycle goes from, you know, A, ensuring that there's data abundance as well as data quality going into the systems, but also being able to measure the AI systems, which builds confidence in, in AI, and also enables fur- further development and further adoption of the technology. And this is, this is the fundamental loop that I think every AI company goes through. You know, they, they get a bunch of data or they generate a bunch of data. They train their models, they evaluate those systems, and they sort of, you know, uh, go again in the loop. And so evaluation and measurement of the AI systems is a critical component of the lifecycle, but also a critical component, I think, of, of society being able to build trust in these systems. You know, how are governments gonna know that these AI systems are, are safe and secure and fit for, uh, you know, broader adoption within their countries? How do, how are, um, enterprises gonna know that when they deploy an AI agent or an AI system, that it's actually going to be good for the consumers and that it's not gonna create greater risk for them? How do, um, how are labs gonna be able to consistently measure what are the intelligences of my, of the AI systems that we build, and how are we gonna, you know, how do they make sure they continue to develop responsibly as a result?

   3. SGSarah Guo22:49

      Can you give our listeners a little bit of intuition for like what makes evals hard?

   4. AWAlexandr Wang22:53

      One of the hard things that ... You know, because we're building systems that we're trying to approximate and, and build human intelligence, um, grading one of these AI systems is, is not something that's very easy to do automatically. And it's, it's sort of like, um, you know, you have to kind of build IQ tests for these models, which in and of itself is a very fraught philosophical question. It's like, how do you measure the intelligence of a system? And this is, there's very practical problems as well. So most of the benchmarks that we as a community look at for the-

   5. SGSarah Guo23:25

      The academic benchmarks.

   6. AWAlexandr Wang23:26

      Yeah, the academic benchmarks that are what the industry use to measure the performance of these algorithms are fraught with issues. Many of the models are overfit on these benchmarks. They're sort of in the training datasets of these models. Um, and so ...

   7. SGSarah Guo23:38

      You guys just did some interesting research here.

   8. AWAlexandr Wang23:39

      Yes.

   9. SGSarah Guo23:39

      Like, you published them.

   10. AWAlexandr Wang23:40

       Yep. So we, one of the things we did is we published GSM1K, which was a, uh, a held-out eval. So we basically produced a new, um, evaluation of the math capabilities of models that there's no way would ever exist in the, in the training dataset to really see how much of the perf- how were the performance of the models, um, uh, were the reported performance of the model capability versus the actual capability. And what you notice is, some of the models perform really well, but some of them perform much worse than their reported performance. And so this whole question of how we as a society are actually gonna measure these models is, is a really tough one. And our answer is, we have to leverage the same human experts and the, kind of the best and brightest minds to do expert evaluations on top of these models to understand, you know, where are they powerful, where are they weak, and, and what's the sort of, um, what are the sort of risks associated with these, with these models? So, you know, one of the things that, um, we're very, w- you know, we're going to, uh, we're very passionate about is, there needs to be sort of public visibility and transparency into the performance of these models. So there need to be leaderboards, there need to be evaluations that are public that demonstrate, uh, in a, in a very rigorous, scientific way what the performance of these models are. And then we need to have, build the platforms and capabilities for governments, enterprises, labs, to be able to do constant evaluation on top of these models to ensure that we're always developing the technology in a safe way and that we're always deploying it in a safe way. Um, so this is something that we think is, you know, just in the same way that our roles in infrastructure provider is to support the data needs for the entire ecosystem, we think that building this layer of confidence in the systems through accurate measurement is going to be fundamental to the further adoption and further development of the technology.

   11. SGSarah Guo25:21

       Thank you all so much for listening. We've really enjoyed talking to people reshaping our world with AI. Uh, to listen to any of the full episodes, please find the links in the description for this podcast. And we'll be back with new interviews next week. (instrumental music) Find us on Twitter @nopriorspod. Subscribe to our YouTube channel if you wanna see our faces. Follow the show on Apple Podcasts, Spotify, or wherever you listen. That way, you get a new episode every week. And sign up for emails or find transcripts for every episode at no-priors.com.

Episode duration: 25:55