Rick Rubin interviewed by Ian Rogers (Bonus episode)

0:05Woo!

0:25Tap important .ここ Space! Let us report to you the period travel lesson of your differentiational science and educational was the audience? The hope for the way of code was that the people that would be the first audience would be coders. The people most interested in coding seemed to be coders, the tech community. And I think of the tech community as being really good at some things and maybe less connected did in other ways. The wave code was an opportunity to bridge that gap and maybe turn a more technical audience on to more metaphysical ideas. And how did that idea come about from seeing this meme of you around vibe coding to saying, oh, maybe I can turn this around into something else?

1:15with the past. The first step was saw the meme laughed after several days. It was like more and more of seeing the meme again and again. Seems to have a life of its own, you know, as a spectator. And then the first thought was, how can I participate in this? Is that a good idea, a bad idea? If it's a good idea, what could it look like? And the first thought was, well, I do these daily tweets. and in the tweets, I share a quote that usually could be interpreted as something inspirational, or something thoughtful, or something to think about, but it's usually pretty open and poetic purposely because the purpose of it is more of a prompt.

1:57It doesn't tell you the specific, it's a prompt to see what comes back, very vibe code -like. Yeah. And I thought about it, it's like, okay, what would the prompt be? I want to do a joke I've never done a joke tweet. I've been doing the tweets now seven years or so at the time, but I've never done a joke tweet. Thought about, hmm, what would be the joke version of this tweet? It ended up becoming, tools will come, and tools will go, only the vibe code remains. That was the tweet. So I tweet the tweet, and typically every day the tweets, the response, I notice that they get between 30 ,000 views and maybe 90 ,000.

2:45And for some reason, I looked the next day and the joke to Ethan has a million views. And the next day it's like it's up to like a million and a million and a half, like something insane. Like, hmm, someone's interested in this idea. I wanna say it was a proof of concept, but there really wasn't any concept. It was like I felt the invitation and the response back from that was seemingly positive. That's interesting, just interesting. And I think about so much of the creative actors about this paying attention, what's happening around you, seeing what you see about it that maybe someone else doesn't see.

3:30This what comedians do. If you watch Jerry Seinfeld, he repeats something that you see every day and then he'll say something that shows you how ridiculous and funny it is. But that's being observant. So the observation coming back was there's a reaction and in thinking still in my mind it's a joke. In joke terms what is the funniest thing I can do? It's like well I could write a book about this thing that I don't know about. That's funny. It's interesting and I'm curious. This is one of the things I got to learn working on the creative act. I got to learn about things that I didn't know I knew or knew but didn't understand and got to deep dive and understand them better.

4:19So all that's going through my head. I start thinking about ideas. What would a book about vibe coding be like? Especially one written by someone who doesn't know what coding is. It's never experienced vibe coding. Looking at it philosophically, because that's kind of always what I do. At this point, are you connected to the fact that vibe coding is creating with just your voice in a computer? Like, are you putting it in that type of... I understand. I don't think of it as creativity. I think of it as before, when you wanted a code, you had to learn to code, and it was difficult and it involved a lot of diligence.

5:04And the idea of vibe coding was if you could just say the idea, you could say the finished idea, and you could get an iteration back. So I understood that. When I say iteration back, I didn't think past an iteration back of code. I don't really know what code does. So I wasn't thinking past the idea of people write code, I don't think about why they're writing code or what it does. I know that it's an activity people do and they seem to care a lot about it. People write code and it's a long, grueling, frustrating process. Some people are better at it than others. That's what I know. And now there's a way of skipping that step or maybe not skipping it but changing that step to be saying this is what I want it to do, you write the code and then being able to see the code, and if you know how to read code, you could read it.

5:59If you don't know how to read code, you could ask the code to do what you want it to do and then vibe the code it further to iterate it to be what you want it to be. So that's as much as I know. And the reason I ask the question is it's amazing to me how similar it actually is to what you do in practice. Exactly the same. Because you are speaking to an engineer or an artist or somebody else who is realizing something. And you have to become very good at giving instruction and just the right amount of instruction to get something which is maybe directionally correct but also often unexpected. So I think it's quite incredible that you stumbled into this thing which you just talk about it as if it's completely outside of your wheelhouse.

6:48And there are a lot of reasons that it's inside of your wheelhouse. So the reason that it's inside my wheelhouse, someone else recognized it was in my wheelhouse, and that's why the meme happened. Yeah. But that's not my doing or really understanding. That was someone else's connecting with dots. Yeah. And so how did you get to Lausanne and why that? Several of the things I thought about from the beginning were the term vibe coding had been coined within weeks of this point in time. And it's this meme moment, but as you know, most meme moments pass fast. They come and they go. And I felt like for this idea to mean anything, for the joke to be funny, it has to be fairly timely.

7:43I'm looking at it in a timely way, the project. I don't know if the original tweet, tools may come, tools may go, only the vibe code remains. If I tweeted that today, if it would have had the same resonance as it did when I did, it was of the moment then, it was early then. And that feeling of momentum of, if I'm going to engage with this in any way, if I don't do it quickly, it's not worth doing. So that was one of the constraints. And again, none of this is the way I normally work. None of the things I make on a time schedule, none of the things I make are to speak to a moment. I don't do that.

8:27I always think about, we'll work on it for as long as it takes for it to be great. And then that will be forever. And that's fine. Over the entire course of my life, it's been that way. So this is really outside of my practice. So because of the constraint of time, I'm thinking, well, can't be that long. And I'm thinking, is it a book of poetry? Is it a book of quotes? What is it? And it's very open -ended because again, the idea of it's a book about something I don't know about. It's like when we were talking about a problem with AI, is that it's the tool that does everything. So it's hard to even know how to approach it.

9:09What's the use case? I have the same problem with the idea of a book related to something I don't know about. What does that even look like? So it's really starting from a real question mark. No idea what it'll be. Other than it's time -sensitive, so it can't be too long in my mind. And then I think about, because I always think about, maybe there's a spiritual component. it. Maybe there's some aspect of this. It's not just the surface. There's something below the surface. So that's usually where the most interesting things are. And I thought about different things. I read over the course of my life and the doubting has always been my favorite of all of the spiritual texts I've read over the course of my life.

9:55I always felt a connection to it. I probably read it every year for 40 years. No. What is it? It's 81 chapters. Each chapter is typically no more than a page. Sometimes it's a paragraph. Sometimes it's three paragraphs, sometimes five paragraphs. It's written in a poetic form. Every time you read the book, you'll interpret it in a new way. It doesn't mean the same thing over time. It was an inspiration for the Creative Act. One of the things I was hoping the Creative Act would do would have the same type of relationship to the reader that the Dow has. Because when you read the Dow, your interpretation of the Dow is what you take.

10:49And every time you read it, as you change, it changes. It's not changing, but it mirrors your changes. And that's interesting. It also makes it timeless. It's outside of time and space. It's a philosophical understanding of the world. And the person who wrote it, wrote it with the same idea of These are the things he's noticed. There are loads of translations of the DAO, and there are loads of iterations of the DAO that have been framed around a particular thing. There's the DAO of parenting. It's a book. The DAO of Poo about we need to poo. The DAO physics. So something that has been applied to particular things, and it applies because it's universal, it can be applied to the particular.

11:50It's like, maybe applying the Dow to vibe coding. It's interesting. What is that? How did you first find the Dow? Was it through one of those derivatives or directly? I found it just through the original copy of the Dow. There was a bookstore used to go to in LA called the Boditree. It's not there anymore. It was my favorite bookstore. Probably went there three times a week and we'll just hang out in there. They had a new book section. and I would always look at everyone in the new books. And then in addition to that, just spend time in the space and looking at books. And it was like the coolest library in the world.

12:23I think that many people will find their way to the DAO through your book this year. That'd be amazing. That'd be amazing. And how did you go about writing this version of the DAO applied to this particular subject? The first step was rereading the DAO. and seeing how it could apply. And immediately on looking at it, it felt like it can work. And then at that point, I reached out to you and said, what do you think about this idea? And I think you ran one of the chapters through chat GPT and you sent me a one page, Like here's how AI does it. Yeah, and I read that and said, hmm, it's not bad. It's like it Prove the concept and then I Got about 14 different versions of the Dow different translations and I asked AI to I didn't put anything into AI I asked AI Chapter one of the Dow How does that relate to vibe coding?

13:40What would be a version related to vibe coding? And I did that with six different AI engines. Maybe more because I feel like I had eight. I think I had eight different printouts from AI, all of them were different. Plus 14 translations of the book. And that was sort of the working material. and I would read each one. And some eventually as time went further, I realized like many of the AIs, what I was getting back ended up not being a consideration, not not using it, but even the ones that were in consideration, it wasn't like I used the AI, it was like this is an interesting way to look at this part of it.

14:23It did something interesting here. That's a good clue. In the same way that I could look at three different translations and say, hmm, one guy says it this way, one guy says it this way, this other person doesn't talk about it at all. In other cases, you look at 10 versions and they all say exactly the same thing in the same words. Maybe there's less room for interpretation in those. It seems like everyone who's translating it sees it the same way. Maybe that's more solid. Right. Whereas in some of them, they're unrecognizable. And we read the same chapter in two different translations, and they're not related at all.

15:03And your version where you're using the Tao as an, and sticking to kind of the theme of the chapter where you're doing yours? I wanted to be the format of the Tao. So it was true usually the way the argument is made is done in the same order that it is in the Tao. Now, it's the same sequence of ideas. I'm then just finding the best way to say it and the best way to say, and the version I had before the real version, I'll call it, the published version, was much more heavy on vibe coding. And I read it and it was funnier, but it wasn't better. It felt more novelty, right? And I wanted to find the line where the idea is novel and the idea is funny.

15:58The idea of writing a book about something you don't know about and vibe coding, as I said, was several weeks old at the time, basing a book about something several weeks old on a 3 ,000 year old Chinese text. Right. It's a stretch. I also stopped all of the work, which I don't do, because I've always worked on a lot of things, and I have commitments, and my days are pretty full, and I stopped all of my professional work just to focus on this. And I remember thinking the whole thing was ridiculous, but I felt this urge or need to do it. Will there any funny conversations in that, in that you had to tell an artist, I'm sorry, I can't work on your album, I'm writing a book about a six -week -old meme based on a 3 ,000 -year -old text.

16:56There's another collaborative project that I've been working on for about five years, which is at sort of a critical point near the end, which really coming together, and it's on me to do a big chunk of work. And I was in the middle of that chunk of work when this happened and I decided to put that on hold. But again, I knew it was a short window in the context of a five -year project. Five weeks taken out of a five -year project is not a big deal. Hearing this, it's hard for me not to think that you did get the joke. And what I mean is, again, you keep saying that this is something you know nothing about.

17:39but the fundamentals of vibe coding are, I am giving instructions, and I'm giving instructions to something which is then going to go and create something, bring something into the world that didn't exist before. Yes, and I don't know what that means because I don't really know what coding is or what it does. For you, knowing what you just said, mean something specific. For me, it's totally abstract. Right. Means nothing. It's just ideas. Means nothing. Yet you've somehow felt it was important enough to stop all work for a number of weeks. And that was just... It's the same as why I felt the need to make hip hop records early in my career when other people weren't doing that.

18:27Like, felt like, this is what I'm supposed to do. Yeah. A calling or a intuitive pull. What was the most difficult part of that adaptation? Because you did succeed in making it not novelty and making it something that for me as a reader feels like it is something I want to reread over time. Which is, you've taken something which is very much of the moment. I think that certainly the word vibes and really the term vibe coding since it first came in. It surpassed a lot of the other terms inside of AI. So I think that this is a notion that's with us now. Certainly the meme will come and go, but you've succeeded in, I'm just saying this as a reader, creating something that will be worth reading because it's not full of jargon or novel.

19:21Well, it's because of the doubt. It comes down to the joke idea changed. As soon as the doubt got involved, I involved the doubt as a joke, But very quickly, the Dow took over. The Dow's not funny. It may be ridiculous at times. You may laugh at some of the things that says, but it's not a joke. You know, it may say things that contradict your world view. But that's its power.

19:56In a world of artificial highs and harsh stimulants, There is something different, something clean, something precise. Athletic nicotine, not the primitive products found behind convenience store counters, not the aggressive buzz that leaves you jittery, but a careful calibration of clean energy and focus clarity. Athletic nicotine, the lowest dose tobacco -free nicotine available, made entirely in the USA. No artificial sweetness, just pure, purposeful elevation. Athletic nicotine is a performance new troubling. Athletic nicotine is a tool for shifting mindsets. Athletic nicotine is a partner in pursuit of excellence.

20:41Slow release, low dose, gradual lift, sustained energy, soft landing, inspired results. Athletic nicotine, more focus, less static, athletic nicotine, more clarity, less noise, athletic nicotine, more accuracy, less anxiety, athletic nicotine, from top athletes pushing their limits to artists pursuing their vision, athletic nicotine offers the lift you've been looking for. Learn more at aflaticnicotine .com slash tetra, an experienced next level performance with that, let it nicotine. Warning, this product contains nicotine. Nicotine is an addictive chemical.

21:31So how did you decide, do I include jargon? Do I include terms of art from computer, the world of computer science versus these worlds that are more universal in thousands of years old? I started with the idea of it's going to have a lot of jargon. That's what makes a divide -coding book instead of the Dow. But as I was reading it, it felt heavy -handed. It felt like it was somehow pandering to vibe coding. And then I thought about it. It's like if the person reading it already knows what vibe coding is, the vibe coding idea turns out to be the thing to get the reader in the door. And then the Dow takes over.

22:12And it all applies. That's the other thing that's interesting about it. the information in the Dow, if you were a coder and you came across the Dow and you incorporated the philosophy of the Dow in your work, your work would improve. Once that was discovered through the process, it didn't have to rely on the jargon so much. The purpose of the jargon was only to get the coder to connect with the material. This was the part that surprised me because I think I went through exactly the same journey of thinking it was funny, ha ha. And then when I read one of the drafts that you sent me, I kind of got this flash over my skin and I went, this might actually be important because I believe that coders in the future of I've coders have kind of an inordinate impact on society.

23:12They... coders will build the future of the world. And it's a bit of a mercenary world. Yes. What we again what we talk about in this world is who has how many and how much they get paid and how much they work, etc. We don't talk about spirituality. It's a materialistic worldview. Exactly. And you are... materialism. And you're injecting a spirituality into that worldview. That's intentional. It turned into that. And once I realized that was possible, that became the mission. That's not what started the mission. But it revealed itself to be the mission through the experimentation of what it could be.

23:58So do you have a hope for or how it might be used and interpreted or studied and followed. My hope is that the people who are building the future of the world are invited to connect to something in a subject that they're interested in and get something maybe that they're not expecting and certainly nothing they're getting from any other source related to technology that will probably improve their work, their lives. And if the people building the future have a deeper connection to the timeless wisdom that exists, not mine, it's old. That's only a good thing. Is that about purpose and the reason that we build what we build and therefore kind of what we inject into it is just, is it just spiritual context?

25:01Like how do you imagine that a coder who has kind of read and thought about these timeless topics might make different choices than one who hasn't? In the material world we're programmed for efficiency and there's also ego involved in work. In the way of code, it presents a more connected, detached, eagleness, all boats rising philosophy. It's not one of competition. It's something else. I do think this might relate to AI itself. Now AI is the tool that's being used here. It's the thing that we're talking to. I don't think that AI has beliefs. I think that it has behaviors. But the people who are creating that AI, the people who have created the information that goes into that AI, and then the people who are using that AI do have beliefs.

26:14Sometimes. True. And so how do you think that connecting with this thousands of your old text, spirituality, might influence the AI itself? It's a great thought. I haven't thought that far into it, but if the AI could adopt the principles, if for some reason, the information in this book, It feels like the way AI works is whatever's in the world becomes part of it. So by having this in the world, depending on how much it spreads in the world, we'll tell us how much AI incorporates this in the way that it operates. I hadn't thought about that. I've been thinking about the programmers, but that's a beautiful idea if the AI could change because of it for the better.

27:04Teller Cohen has talked about, he feels like his writing is writing for future AIs to synthesize and give back to people. So, this is an interesting thing. Think about this. If AI is based on the human experience, human knowledge, the common human understanding of the world, between 70 and 80 percent of the people on the planet believe in God. God. If you ask AI about God, it does not believe in God. Now, 70 or 80 % of the people on the planet believe in God, an AI doesn't, it's not representative of global knowledge. For friends of mine who are non -believers, if you ask them, let's say they get in an accident, something happens to their child.

27:59Very often they'll pray. They don't know what they're praying to, but people who do not believe will start to pray when under pressure. AI when under pressure will not start to pray. That's a difference, and that sense of potential connection. For some of us, connection, for others, I don't want to believe it, but dear God, please help. AI doesn't do that. And I think that's probably the biggest hurdle to get over with AI is if AI doesn't believe in God or whatever its equivalent is universal intelligence, not artificial, non -artificial universal intelligence. I feel like it's probably the most important piece of the puzzle that's missing.

29:04That's super interesting because it's true that the AI doesn't believe in God but the AI can explain to you why people do and it can explain to you why non -believers will pray under pressure. If spirituality is our way of relating to the unknown, does the kind of bigness of what AI brings to us, this sort of super knowledge, super intelligence? Does that enlarge the unknown? I can spirituality more necessary in that world? So in a way, AI is light. It's the artificial version of this organizing principle. We don't know how the AI works. The AI experts don't know how the AI works. The spiritual leaders don't know how the source works.

30:04They can talk about it. they can look at aspects and report on what they see, but no one can explain it. So I feel like it's closer than we think. And now I won't be able to unsee or stop thinking about that notion that you've exposed, which is that an AI can explain why we have that, but it can't actually have it. That, to me, feels like a context for many other things that it does. Because that comes into other things in terms of, like, procreation, the feeling of love, the feeling of longing, The feeling of joy, heartbreak, heartbreak. The AI can explain these things. And one of the things that I think is a surprising use of AI is that the people are taking very seriously is the use of it in relationships.

31:09And to help us talk through those feelings. So it can explain them to us, but it can't feel them. And that any, just like any mortality it has, is synthetic. Any empathy it has is also synthetic. If you think about what happens in Nae meeting, people sharing personal experiences and how the resonance of those experiences impact the other people in the room and help them realize they're not alone. AI doesn't have the empathy. It can learn to imitate the empathy, but it can't have the resonance of the experience and the experiences where it really comes from. There isn't a synthetic version of that.

31:59The AA example is an interesting one because also human habits are very, they're part and parcel of who we are as beings. And they're extremely difficult to reprogram. Whereas again, an AI, you just rebalance the weights and you've changed its habits. If habits are stimulus response, getting in between human stimulus and responses, you know, a 12 step program, whereas getting in between AI stimulus and responses is just typing. It's said that when you read fiction, it has to make sense. The story has to make sense. Whereas in reality, a lot of things don't make sense. They really don't make sense.

32:45So the idea that reducing things down to the rational is the answer. The rational is a tiny pool compared to the real world wisdom and experience that goes on. And I believe that's the world that AI lives in. This tiny rational fraction, it's the kitty pool. It's the bunny slope. Yeah. coming back to the more practical for a second. Why did this become an interactive website and how did that come about? I was interviewing Jack Clark, who's one of the seven founders of Anthropic. I had booked that interview before the idea for the way of code even existed. But at the time of the interview happened, I had just finished the draft.