Transcript: Is NotebookLM—Google’s Research Assistant—the Ultimate Tool for Thought?

The transcript of AI & I with Steven Johnson is below for paying subscribers.

Timestamps

1. Introduction: 00:00:53
2. The roots of Steven’s obsession with organizing ideas: 00:04:47
3. Steven’s belief in being receptive to organic connections: 00:08:16
4. NotebookLM’s north star: 00:11:16
5. How Google Lab’s note-taking app respects copyright law: 00:14:02 
6. Using NotebookLM to analyze highlights from books that Steven has read: 00:16:17
7. Steven’s personal approach to curating valuable information: 00:19:02
8. Why NotebookLM shows restraint when it is asked to speculate: 00:21:51
9. Using the model to co-create the beginnings of a documentary: 00:29:13
10.  NotebookLM generates the opening script: 00:52:54

Transcript

Dan Shipper (00:01:17)

Steven, welcome to the show. 

Steven Johnson (00:01:18)

Thanks very much. And it's great to be here. 

Dan Shipper (00:01:20)

Honestly, I'm so excited to have you. You're such a great writer. I love your writing. And you're also kind of in this interesting intersection where you're doing incredible writing and you're also working at Google. I feel like I'm jumping the gun a little bit, but for people who don't know, you are the bestselling author of How We Got to Now and Where Good Ideas Come From and a bunch of other books—I think 13 other books. You just wrote a new book, which I have right here, called The Infernal Machine. And if that wasn't enough, you are also the editorial director of NotebookLM and Google Labs—NotebookLM is Google's AI research product. And I'm so excited to have you here.

Steven Johnson (00:01:57)

Oh, this is really great. We actually have some kind of news to break as well. So it's going to be really fun.

Dan Shipper (00:02:05)

Yeah. So, by the time this podcast comes out, a new version of NotebookLM will be out and, on the show today, we will be demoing all those new features. We'll talk about how you used to write your last book and then talk about how we can use the new features to maybe write a new book or find a new book concept for you. So, I'm curious, before we get started, talk to us a little bit about what NotebookLM is.

Steven Johnson (00:02:27)

Yeah. It's a tool that we've been developing for about the last two years at Google Labs, which is this wonderful new division inside of Google that is really designed to build your ideal research and writing software knowing that you have, from the very beginning, a language model at the core of the product, right? So, we really began this in the age of large language models knowing that there was going to be a possibility to build entirely new kinds of software. It wasn't just about adding some AI to a word processor, adding some AI to a photo editor. This is about what can you build that's a genuinely new kind of surface to work with? We can talk about a little bit, but the obsession in my life with using software to help me with organizing my ideas, organizing things that I read, and turning those ultimately into books or into TV shows or whatever. It's just been an obsession of mine for 30 years or longer even, really, and so Google's got very interested in this process of co-creating with people. So not just kind of sitting around and building products and then handing them over to writers or musicians, but actually having a writer or musician in the room from the beginning with the project. 

It's a big part of the Labs ethos in particular. And so they called me a fun day two years ago. They were like, hey, would you be interested in maybe helping us build this thing you've always wanted? And that was an easy yes. And so, yeah, we built a very early prototype and I think because I've been in part because the technology was suddenly available with language models and because I've been sitting on these ideas for so long, we were able to build a pretty quick prototype and that got a bunch of interest internally. And we launched NotebookLM in the U.S. in December of last year. And today we are announcing that we are rolling out to over 200 countries around the world. So we're really excited about that.

Dan Shipper (00:04:44)

That's amazing, and I just want to go back. The thing that really sparked my attention at the beginning of this is sort of this obsession with organizing your ideas and using those to produce previous work, because I feel that same thing. I'm a nerd for that stuff. And I just want to understand. Tell us about that obsession or what some of those ideas are like. What are those things that you've always wanted that are kind of starting to come through here?

Steven Johnson (00:05:11)

Yeah. I mean, it actually dates back— I'm going to really date myself here, which is that one of the things that changed my life, truly changed my life was when I was a sophomore in college in the fall of 1987, Apple released something called HyperCard, which was this crazy app. I always say it was a little bit like the Velvet Underground of software. It never really had a hit, but it influenced all these other people. And it was basically a prototype web-like hypertext system where you could organize information pretty much any way you wanted and make links between it. And I just got obsessed with the idea that I could use this tool as a place to kind of keep all my notes from my classes.

And I kind of built this little application that I called Curriculum that was kind of a way of taking notes for classes and I spent way more time building the tool than actually using it to take notes for classes. And I kind of stopped going to the classes for a while because I just wanted to build the tool, but it was one of those things. It just gave me a taste. It wasn't ready in any way for primetime use, but I got a sense of the possibility. And then obviously when the web came along, I kind of jumped on that maybe a little bit earlier than some people, because I lived briefly in the world of HyperCard and early hypertext. And then about 20 years ago, there was a program that I wrote about a lot, actually, a wonderful program, way ahead of its time, called DEVONthink. And it's still around, actually. It's a really cool application and it enabled me to keep all these quotes from books that I'd read. I would originally kind of type them in. And then when the Kindle came out, I could get the quotes digitally. And you could kind of make connections between quotes, or you could type something and say, what quotes in my research library are related to this? It had this associative early kind of semantic search. and I use that quite a bit on a lot of my books, like The Ghost Map. And I would have these moments where the software would recommend a quote that I had forgotten from an earlier book, and it would make a new connection in my mind that I hadn't thought of before. And I thought this almost feels like a partnership with the software, like I'm curating these quotes. So it's me and I know how to turn them into a chapter or a paragraph in a book. That's my intelligence, but the connection was really made by the software—that seems kind of new and tantalizing and weird, but also maybe very powerful. And so that was another taste that kind of got me along that way. And then when I first started experiencing what was possible with language models, starting with GPT-3, before I came to Google, I thought, oh wait, now it's all really going to be possible. All this stuff is going to get very serious and very real. So that's the prehistory.

Dan Shipper (00:08:18)

And I'm curious, why do you think that every nerd’s dream is to have this interconnected note system that you can use to make stuff. Why is that so appealing?

Steven Johnson (00:08:29)

Yeah. This is true on so many levels, but there are different kinds of nerds. The thing I've always felt, and this is generally true of the way I organize my email as well, which is to say I spend zero time organizing my email. My principle has always been to create one place where you dump everything and then use smart tools like search and now all these language models to find what you need. Don't spend any time organizing anything, just throw it all in one place and focus on having the ideas and stuff like that. And so and I think Notebook—I mean probably to a fault— Notebook, I love, has been kind of designed a little bit with that principle. You can't tag your notes for instance. And we probably should have—people do like to tag things. And I just always like, I'm not going to spend a second tagging anything because I want the software to understand what categories— I don't want to put things in advance into buckets because I want it to be an open-ended connective system where I can make new associations or create new kind of clusters on the fly. And if I spend all this time tagging, I'm going to limit the connections that I can do. So, I'm on the side of emergent chaos, right? Don't organize it and just let things bubble up and figure out tools that will let that bubbling up happen. But then there's a whole other set of folks who really like to organize it and systematize it and have it all in these categories and things like that. And so hopefully we can ultimately make NotebookLM play well with both those groups. I think it's certainly within our power.

Dan Shipper (00:10:10)

Yeah, it's the top-down versus bottom-up folks. I'm definitely in the bottom-up camp with you. So, I'm excited to see Notebook. Let's roll into that. So give us a little bit of a tour of Notebooks. Because we're going to talk about how you used it for your latest book. So tell us a little bit about how Notebook works and then how you used it for the book that you wrote.

Steven Johnson (00:10:31)

Yeah, I'll just give you the basics. The idea is everything inside of NotebookLM is grounded in the documents you provide. So we may open this up a little bit over time. I think it's probably a logical thing to do, but unless you provide Notebook with what we call sources, the documents that are the source of truth for your project, the things that you're working on, and it could be everything from your journals to work documents or to research materials or quotes from books that you've read. You begin each project by opening up a Notebook and uploading sources. And at that point, once they're uploaded, the model, in a sense, kind of becomes an expert in the information you've shared, now this has become increasingly common. It was kind of a radical idea when we were first toying around with the two years ago. But the idea of having documents attached to a model like Gemini or ChatGPT has become increasingly common. But for ours you're always working with documents and the whole interface is designed to let you basically load a lot of different documents, move back and forth between those documents or potentially read those documents while you're working and not get into that flow that so many people I think are finding these days where they have 12 tabs open and they're grabbing some text from one tab and then pasting it into the chatbot and another thing. And then they're getting the output and they're saving it in another document. We want to have a single integrated surface where you can do all that work. So in a sense it’s designed to not interrupt your flow state, if you're thinking or writing or reading, you should just have one place where you can do that and not be like, wait, what was the tab that I had open where I had that quote that I wanted to use and that other thing? So, that's the kind of the underlying model. And if those of you are watching this, I have opened here, a Notebook that— This is kind of cool. This is my crazy Notebook. This is, this is where I have, all of the quotes that I have collected over the years for books dating back to something like 1999, I think it goes back to. So about 7,000 quotes that I've collected. So it's really my reading history, the things that were important to my books in the past. And they're lined up here as a bunch of different sources. This is kind of left over from the fact that we used to have a kind of a cap on how long the sources could be. So now you can have—this just changed a couple of weeks ago. In each Notebook, you can have up to 50 sources and each source can be up to 500,000 words, so you can effectively be talking with 25 million words in a single Notebook, which is just kind of mind-blowing. So these quotes are— I'll just open up one of them. I think there's something like 2 million words total—7,000 quotes, 2 million words, something like that. And—

Dan Shipper (00:13:50)

This is crazy. So, okay, these are all of the quotes from all of the books that you've read since 1999?

Steven Johnson (00:13:58)

Well, yeah. I mean, not all of the books, but yes, a significant amount. And so—

Dan Shipper (00:14:05)

This is very valuable. Can you just send me a Stripe link? How much do you want for this?

Steven Johnson (00:14:10)

And by the way, I should point out one thing that's really important here. So these are quotes, these are all books that I've purchased, right? And these are quotes that I've clipped increasingly using the Kindle, using the limits that are built into Kindles and the amount that you can quote and use. And it's really important to stress this, it is important for me as an author, we're not training the model on this information. We're just loading the information from these quotes into the context window of the model, the kind of short-term memory of the model. And we're using that to answer questions or be intelligent about it, so there's no chance that this information, which is under copyright, is going to be used to train the model or be shown to anybody else, and so you have this freedom to work with material. If you have the right to use it under copyright, you can use it inside of the Notebook. We spent a lot of time ensuring that that works. So, yeah, this is incredible— I'm just scrolling through this. For those of you who are listening, it's just an endless list of quotes and I'm just scratching the surface. And each source that we put in, we create a source guide that summarizes the source. Now this is normally extremely useful because generally a source is on a single topic and you can get a kind of high-level thing. It's crazy when you can give it whatever. This is probably 800 quotes in this one source on all these different topics. And so it creates a summary—this source explores the intersection of scientific advancement, societal impact and ethical considerations and how it's doing a very good job of trying to make some kind of pattern out of all this. But source guides for just quotes are not quite as useful. So now, at this point, basically, I can ask any question. And so I actually preloaded a question here. I'm going to close this source. I preloaded a question if we view the chat—

Dan Shipper (00:16:00)

And the question will be asked of all of these sources together.

Steven Johnson (00:16:04)

Yes, I can. You'll note that they're checked off here. So you can always tell this is a really subtle thing, but it's really important about it. It'll say down at the bottom, at the chat box, it'll say 15 sources. So that means you are currently talking to all 15 of your sources here. If I actually deselect this one for some reason, now I'm talking to 14 sources. And so you sometimes have moments where you’re like, actually, the information I want is only in this one document and I don't want to ignore all the other information. So you can actually shift the focus of the model to various different things really, really easily. So, I asked this question, “What are the most interesting facts about ant colonies here? Mention authors and specific books” because I actually wrote a book a million years ago called Emergence. We talked about the emergent approach to these things. And there was a big riff about ant colonies in that book, and the model is smart enough to understand this concept of like interestingness too and surprise. I often say, what's the most surprising idea here? Because you think about that as an author and the idea that that can be like effectively a search query is just totally bonkers. So, I'll just read this for people who are listening. So it comes back with, “Interesting facts about ant colonies. Ants use pheromones to communicate in various messages such as danger, food, location, and nest mate recognition. This complex chemical communication system allows ant colonies to function as a single unit or super organism.” That's fantastic. That is a great answer. And now this is new, by the way. So this is a brand new feature rolling out today. We're incredibly excited about it. We now have these inline citations. And so that shows you exactly the quote from the my reading notes that it used to generate this answer. And so you can just roll over them and you can see where the model came up with this. This one is from Norbert Wiener. That's pretty interesting. And you can see there are citations all over the place. And what's even cooler is you can—although a little bit in a way less useful for this project, we'll show it in another thing, but I can always click on those and it takes me exactly to the point in my documents where the original quote came from. So you have this ability to kind of get asked the model to help you get the lay of the land, like what's in here? I'm interested in this topic. What's there? And then because the sources are integrated into your Notebook itself, you can then dive right in and start reading. And so you can go through all these things. And then we also suggest questions based on what you just asked. So there's always something to just click on.

Dan Shipper (00:18:57)

Can I ask a question?

Steven Johnson (00:18:58)

Yes. Go right ahead.

Dan Shipper (00:19:00)

And you tell me whether or not this is a good question to ask. But one of the things I'm interested in is just looking at this and knowing how many things are in here that you've kind of collected over the years. What are the types of things that Steven Johnson is likely to save? What are the patterns in texts or books that are likely to make you want to put them in there? Do you think it would be good at that?

Steven Johnson (00:19:25)

Would Notebook be good at finding those things? Yeah, it's a great question. So, there are two different versions of why I save things, the first is pretty easy and, yes, to this particular use case, Notebook will be great at this, which is, I have a very specific project in mind. I'm in the middle of a book and I'm writing The Infernal Machine, the new book. There are a bunch of themes you've read. I think it has the history of anarchism, the history of forensic science, the history of the birth of the FBI, the invention of dynamite, all kind of woven together in this single plot. So when I'm in research mode for that, I could very easily tell NotebookLM, you know, these are the key themes of the book, help me find passages that are relevant to those themes. And a little later in this conversation, I'll show you that in practice—that's very easy to do. And it's just incredibly good at that. The trickier question is, I often save things without knowing where they're going to go. There's not a specific project that they're attached to. I often save things that I don't understand. I read something I understand, I'm like, well, I already know this on some level, so I don't need to save it, whereas when I read something and I'm like, that doesn't make sense, that's provocative, but I don't really get it. I often try to save that. And you get a little glimmer—I've written enough books by this point, and I 've worked on enough projects—so you get a little glimmer of like, that could be something, and so you save those things, even though they don't have a slot to go in, or don't have a chapter to go in, or even a book to go in yet. And that's the question over time, will software like NotebookLM get to know my general sensibility enough to be able to detect those things? I would have said two years ago, absolutely not. And now I think it probably will be able to do that. And then you start really getting into the agent mode, where you’re like please go scour the internet for fine things that could be useful for me. And I clearly think that's in our future, I think.

Dan Shipper (00:21:39)

I guess that's what I'm asking is, do you think we could ask NotebookLM what your sensibility is based on these sources and then have it return something that might tell us something about the way your brain works? That might be surprising and interesting.

Steven Johnson (00:21:56)

Yeah, I don't know. Let's try. “I'm Steven Johnson, the author, who—”

Dan Shipper (00:21:59)

Maybe we shouldn't say who you are just so that it doesn't have that to go on. 

Steven Johnson (00:22:04)

Yeah, that's right. That's right, because it's clearly, I'm well represented in the training data. I'm sure it knows exactly—

Dan Shipper (00:22:14)

13 books will do that. 

Steven Johnson (00:22:15)

Okay. So, “These sources are my reading notes from books I've read over the last 20 years. please describe—” What would you say? My interests and sensibility? “... and general sensibility and sensibility. Based on these quotes, what do I care about most and what can you tell me about them?” It's going to get very intimate, Dan.

Dan Shipper (00:23:00)

I know I'm excited for this. We're going right for it. 

Steven Johnson (00:23:05)

Yeah. I mean, this is so much of what our existence is now with these models—talk about emergence. It’s like, what will it do in this situation? No one ever sat down with Microsoft Word and said, what will it do? Okay, so this is interesting. It does a little hemming and hawing. So it says, “It is difficult to draw concrete conclusions about your interest and sensibility solely from the provided source material. However, the content suggests an interest in a diverse range of topics, including literature. You included reading notes from novels by authors like Jonathan Franzen, history and politics, philosophy and economics—lots of citations—psychology and cognitive science. Yeah, I mean, it's right. But it's not going to win any—

Dan Shipper (00:23:52)

It doesn’t want to go too far. If we did the same prompt, but tell it not to hem and haw and assume that if I saved it's because it's interesting to me. Give it a little extra juice.

Steven Johnson (00:24:15)

I wonder if it does have some memory and we're just like, “Can you please try to speculate on my personality based on my decisions to these quotes. I know you don't want to, but I am asking you nicely.”

Dan Shipper (00:24:46)

It's so funny to just learn how to cajole these things too just do what you want to do.

Steven Johnson (00:24:52)

But the truth is this is harder to do with NotebookLM because, for instance, if you try and ask Notebook about something that is not in these sources, it will decline to answer. And so, yeah, it's not going to do it. We definitely we have tried to build this model and I think eventually we're going to have an approach where there's a slider of sorts where you can choose different levels of openness and closeness, but we really started with the idea let's build a model that is fully grounded and then we'll really just stick to the facts. And so what it's straining against here is that the sources don't have any psychological explanation of who Steven Johnson is and that's a tension. I sometimes definitely use it and I'm like, oh, okay, I would like you to be a little bit more open-ended, but I think in the research mode, we're erring on the side of groundedness more than we are on the open-ended side.

Dan Shipper (00:25:45)

I have another one and you tell me whether you think it'd be good for this because it's less about you and just more about the sources. And that's like, what are the two sources that I've saved that most disagree with each other? And what do they disagree about?

Steven Johnson (00:25:54)

Okay, that'll be interesting because there's so many, right? So can it— But let's try. I mean, it'd be interesting, “Of all these quotes, what are the two authors who most disagree with each other, whose positions are the most opposed?” And this is kind of weird grammar, but it'll probably understand. It's just so much, Dan. I'd be interested to see what it says. But it may have an odd— Hey, dude, there's 7,000 quotes here, I can't possibly— Okay, here we go. There we go. That's interesting. Well, it's from the recent stuff from The Infernal Machine. So these are my notes. This is from, so this is Johann Most and Emma Goldman disagreeing about the use of violence. This is interesting because they didn't disagree at the beginning, but Goldman gets into this kind of major fight with him. So it's not a yeah, she ends up— Okay, here's an argument for why this is a good answer. They had a philosophical disagreement about the use of violence in politics that got so intense that most got up in New York City and gave a speech and Goldman jumped on stage and attacked him with a horse whip. So, actually, this is actually a pretty astonishing answer because like it really found arguably the only place in the notes where two figures who were mentioned in the quotes physically attacked one another.

Dan Shipper (00:27:50)

So, okay. I have one more follow up and then, and then I'd love to move on. But so the follow up is something like, of the other sources in this corpus, who is the author, or what is the set of ideas that these two should have come in contact with, that would have helped to mediate their dispute?

Steven Johnson (00:28:10)

Okay. It'd be interesting to see. I'm just going to just make sure it keeps the dispute in its mind. So, “In the dispute between Goldman and Most over violence, which other author in these quotes would have been most helpful in resolving and mediating their dispute? And why?” You're pushing the boundaries of this. I love it. And we shall see. That's exactly what I would have said. That's so funny. So Peter Kropotkin is the anarchist philosopher who Goldman was heavily inspired by who had a middle-ground position on the use of violence and was— Yeah, I mean, that is exactly the right answer as well. Okay, good. Well done.

Dan Shipper (00:29:36)

I love it. This is very cool. I know we have you know another segment, which is what I really love doing as part of these shows is to go on a little exploration together and use these tools to do something new that neither of us know where it's going. And so I think what you have prepared is another instance of Notebook with a bunch of preloaded documents that we've never really gone through, and we're going to use that information to help us find what could be a new concept for a book or a documentary. Can you set the scene for us here? What are we looking at?

Steven Johnson (00:30:12)

Yeah. Okay. So this is a new Notebook that I've created. One of the things that NotebookLM is incredibly good with, which is useful across a lot of different domains, is interview transcripts. So, many workflows involve people—if you're a reporter, if you're a documentary filmmaker, or if you're a user researcher or a market researcher, you do all these interviews and you're trying to discern what the patterns are, figure what the lessons are, and you dump a huge number of words of interviews. It's very disorganized. And NotebookLM is just amazing at going through those interviews and finding the relevant quotes and things like that. So, this is an example where the scenario is— I have been thinking about this legitimately, which is there's a great oral history project that NASA created on the history of the Apollo missions and others. There are thousands of interviews, and I've gone in and I've taken about seven of them from people like John Glenn and Gene Kranz and some famous people. And it's about 200,000 words worth of transcripts that are in this particular Notebook. And I'll show you one other thing. This is also new as of today. We're very excited about it. Up until recently we've supported Google Docs, PDFs, text files, and so on. And now we have Google Slides. And we have true image understanding built into Notebook as well, so you can have slides in there as well. And in a sense, kind of talk to your images and do queries and it will understand the images. So it's amazing. It will do handwriting analysis and things like that. It's pretty powerful. So we're really excited about that. That's a brand new feature as well. So, in this, what I was thinking we could walk through is this idea. I'm thinking about the project that might involve the history of the tragic Apollo 1 fire that killed three astronauts in 1967. And thinking about could you make a documentary about that? What would that look like? And the question is—I've dumped a bunch of sources here—what do I have that I could use to kind of build the beginnings of a script maybe for a documentary? And I want to show you one other new thing that is related to this. And in a sense, this process of, I have a lot of information and I need to get my bearings. Just what is it? What are these sources? This is not a situation like my reading notes where I've already read it. It's like somebody gives you a bunch of files. You're trying to make sense of it. You've downloaded all this stuff. You don't know. You haven't listened to the interviews yet. And so we've added this. This is a really cool thing. We previewed this at I/O in Sundar's keynote a couple of weeks ago. This is a new feature called Notebook Guide and Notebook Guide basically gives you a summary across all your sources in the Notebook and gives you some suggested questions that are kind of higher level, how do I really understand everything here? But we also have these templates that are kind of pre-created FAQ study guide table of contents, timeline, and briefing doc. And so those are ways to get kind of the big picture view of what's in the documents. And they take a little bit of time to generate. So I preloaded a couple of them, so they're back here. By the way, I forgot to mention this before, you can save anything as a note in this kind of noteboard area. So you have this area that can be filled with notes. You can write your own notes, if the model says something interesting, you can pin it to this noteboard area so you're able to kind of capture this stuff as you're having conversations, as you're reading, you can capture everything to this noteboard. I'll show you a little bit more about that as we go on. But here's the FAQ it generated based on these sources. I went through 200,000 words worth of material and decided to figure out, okay, here's some good questions. What motivated individuals to join NASA? What kind of rigorous training and selection processes did aspiring astronauts endure? How did NASA manage the immense technical challenges and risks? And there are answers for all of these? And the other thing that I love to do—because this is something as a writer you're constantly dealing with is it'll create timelines, which is just incredibly useful. Creating a timeline is the most laborious thing, but it's really something you need. If you're writing a book, you need to know what the sequence is. And it just will go through all these disparate transcripts and pull out. This is not based on its training data. This is based on the information that's in all these transcripts and it goes through and it breaks it up into like the early pre-NASA era, the Mercury program, the Gemini program, the Apollo program, they’re bullet points, you can see for each of these that break it down. And then we also have a cast of characters. So it pulls out all the people who are mentioned, who are important, and gives them a brief description. I mean, it's just like, it's so useful to get to understand what's in the material. So that's a start. The point here is that these transcripts are not at all focused on the Apollo 1 fire. So the it's not a needle in a haystack, but it's like there's a giant haystack of NASA-related information in this Notebook, and then there's something, I don't know, something the size of a shoe—not a needle—about the Apollo 1 fire, but it's spread out throughout all the documents. And so what I'm trying to do in putting together ideas for this documentary is figure out what's there that's relevant to this particular topic. And so I pre-wrote this prompt and again, I'm trying to give it a little bit of context about the kinds of things I'm interested in. So, “I'm the author and TV creator, Steven Johnson. I'm interested in making a TV documentary about the Apollo 1 fire in the multidisciplinary style of my books and shows like The Ghost Map and How We Got to Now with a focus on surprising scientific explanations and compelling narratives. Give me a reader's guide to the most important sections of these interviews that I should read in getting started with this project.” And actually, I think it's still in the chat, actually, if I bring it up. So this is the reader's guide that it created. And it just goes through interview by interview and pulls out the most relevant sections, and so it talks about the Kraft interview and explains who he is. And then he talks about the Apollo 1 fires, a turning point in the lunar program, and then it moves on to this Yardley guy and then Frank Borman. And so let's say I'm interested in the Borman quote, so I think it's this one. Let me see. Yeah. Okay.So there's a citation next to this quote and I can click on that citation. It takes me straight to this passage and the passages talking about the trouble they were having with the spacecraft before the fire, and so let's say I'm thinking this is an interesting quote. I'm working on a documentary. I like that. And so I can just select this quote and then I can say, add to note. And so it'll take that quote and just add it right there to my noteboard. So I've got this saved response from this interview. So then I can kind of go through, I can go back to the chat and I can see if there's, okay, here's a quote from Neil Armstrong, and he's talking about— Let's see, I can flip through these ones. I think the last one is probably the best. Yeah, okay. So he's talking about the Apollo 1 fire, some very traumatic times. “I suppose you're much more likely to accept the loss of a friend in flight, but it really hurts to lose them in a ground test.” So I'm taking that quote. That's doubly, doubly traumatic. So I could also save that I can add to another note. So I'm kind of starting to collect ideas here that could be useful. And then I can obviously ask anything else. Anything you want to do?

Dan Shipper (00:38:30)

I have a lot. I have a lot of questions. I'm full of questions. So I guess one of the things that strikes me a lot about your books is if you find a pivotal moment where everything sort of changed and then you sort of trace how different technological innovations or different ideas led to that moment and created it. And I'm wondering if there's a way for us to find some of those in here in ways that we might not know already. So it's like, what are the sort of key technical ideas or key underlying innovations or lack of innovations that led to one of the pivotal moments in this narrative? I don't know how you would frame that exactly, but that's sort of what I'm getting at.

Steven Johnson (00:39:09)

No, I love that. Okay. So let me show you one little kind of low-tech NotebookLM hack that I often have a written note with prompts because sometimes you're reusing your prompts, right? So, inside the Notebook, I'll have just a prompt note. And so for instance I actually want to continue this idea of I'm interested in the Apollo 1 fire. I don't want to have to rewrite that. So we can just take that and then create a new prompt with that kind of introduction and then say. “I want to find a scientific idea or scientific or technological idea that is central to the Apollo fire that I could develop into a major set piece for this project. What would you recommend based on these—” You don't really have to say based on these sources, but I sometimes like, yeah—

Dan Shipper (00:40:23)

Can we add a little bit more like, ideally, the idea is seemingly unrelated and surprising, but in hindsight, inevitably led to the fire.

Steven Johnson (00:40:34)

Yes, it's good. It's tricky again because it's limited to the sources. But let's try it.

Dan Shipper (00:40:45)

You modify it in a way that you think will work best.

Steven Johnson (00:40:45)

Yeah. “Ideally the scientific concept will be surprising and involve an unusual connection that the viewer might not originally thought of.” Okay, so then I'm just going to copy-paste that. So we have it, we can reuse it too, and let us see what it comes up with. It came back with a note called “Pure oxygen environments.” And it says, “The use of a pure oxygen environment in the Apollo command module, while seemingly counterintuitive, played a significant role in the Apollo fire. The unusual choice stemmed from a desire for simplicity and weight reduction in the early spacecraft designs. This is what made it so flammable.” And so here's the crazy thing, Dan. When I was thinking about this a little while ago, just kind of as an early idea independent, I mean, I was kind of researching with Notebook, but I saw this stuff about pure oxygen. And I, Notebook doesn't know this, so this is just fortuitous on some level, but I wrote a book called The Invention of Air that was about the discovery of oxygen, Joseph Priestley the 18th century chemist who kind of isolated and isolated and first identified oxygen. And I thought, oh, maybe there's kind of a version of the Apollo 1 story that connects to that history of oxygen that I've already written about. And that could somehow be set up here. So the fact that it likes to pull that out now I suppose maybe there's some training data that it knows that I've written about oxygen?

Dan Shipper (00:42:40)

I mean, now we're getting excited because we can take this and go back to your readwise notes and try to pull some sources, right? Do you have oxygen sources in there?

Steven Johnson (00:42:50)

Yeah, let's see. I could. That would be interesting. I can't remember what's in there from that book because that was a long time ago, but it probably has some stuff in there. Let's try it. We'll go back over to that other Notebook and we'll say, “I'm writing about the use of a pure oxygen environment that caused the Apollo 1 fire. What quotes in these sources could be relevant to the use of oxygen and the history and its history? Explain how I could use those ideas.”

Dan Shipper (00:43:50)

Should we ask it to be surprising or do you think we should just do straight up first?

Steven Johnson (00:43:56)

Let’s not overtax it, Dan. It’s a very sensitive model.

Dan Shipper (00:44:02)

I’m not going easy on it. It can handle it.

Steven Johnson (00:44:06)

Yeah. This does raise an interesting question. I have to jump to another Notebook to do this. And there certainly there's this idea that perhaps you would want to have, okay— That one started right away. So it just goes straight to Joseph Priestly, who I just mentioned and Scheele who also discovered oxygen independently. Oh, that's really cool. It brought up Piccard. Yeah, this is great. This is another idea I actually never wrote about. I think we're building something good here, Dan. Are you going to want a piece of this project?

Dan Shipper (00:44:42)

I’ll take 10 percent.

Steven Johnson (00:44:44)

So this guy Piccard, I wrote about this because I wrote about the discovery of the ozone layer, because I read about this guy who invented the freon CFCs that caused the hole in the ozone layer. And so there's this explorer Auguste Piccard, who went up to the kind of stratosphere for the first time. And so this product just blows me away so, so, so many times. So look at what it says here. So it reminds me of this, Story, which I thought was fascinating, but have never used and briefly describes it. I've got a link back to the original citation so I can go and read more about it. But this is what Notebook says, “This source provides an example of an early enclosed environment that relied on a pure oxygen supply similar to the Apollo 1 spacecraft and notes the importance of respiration in such an environment. Oh my god. And it gives this quote. As the professor remarked, when you face the possibility of shutting two men up in an airtight space of such small dimensions, you must study very carefully the problem of their respiration.

Dan Shipper (00:45:48)

That's so cool. That's the opening to the book or the documentary or whatever. That's amazing. That's so cool.

Steven Johnson (00:45:57)

I'm actually going to just go and copy that into the other Notebook. I'm shifting tabs here, but we gotta add that. So I’ll close the chat briefly. And so here I'm just like creating a new note, but it's like a written note, so I can paste that in there, okay. Do you have another one?

Dan Shipper (00:46:10)

So now we've got this, this sort of Auguste Piccard idea, and I'm curious if there are any other parallels between the two stories. So we've got pure oxygen and a sort of sealed environment. Do you think given this as context, this note that you just put in there, do you think we could search through the sources to see if there are other parallels?

Steven Johnson (00:46:36)

Other parallels to Auguste Piccard in the NASA transcripts? Yeah. The thing that's actually annoying to me about NotebookLM right now, is that the contents of your notes and the contents of your sources are kind of distinct. This is a key feature that people actually, most Notebook users, I think probably don't fully understand, but if I select all my notes. I focused on the model you see down at the bottom says six notes. I'm now focusing the model on the notes and not on my sources. And so I can do things: I can summarize, I can suggest related ideas. That's what I could do. Actually, I could actually— This will work, but what we really want to be able to do, we just haven't built it yet, is have the option of talking to all of your notes and all of your sources at the same time. It doesn't quite work that way, but there is a hack that will work actually. So what we're going to do is can we just go to the prompt? Yes, we could, but I have an even better version. 

We have a dedicated prompt that we've built called Suggest Related Ideas. And so I've selected the note about Auguste Piccard, right? Or Auguste, I don't know how to pronounce it, and now based on that note, I'm going to click on Suggest Related Ideas. And basically what it is saying is take this note and find things in my sources that are related to this note. And you know, eventually you can imagine where this is going. You could be writing a note and be in the middle of a paragraph and you're just like, okay, I just wrote this paragraph. Tell me other things that are like that. And it's just an extension of your memory. That's obviously where we're headed, but right now you have to kind of select the note. So here we go. We suggest related ideas, I actually haven't even tried this one. I wonder if it comes in with citations, this is kind of an older feature. I wonder if we've updated for citations. I'm learning about the product as we use it here. 

We're doing it live, folks. Yeah, this is the beauty of launch day. Okay, yeah, I did. I'm sure there's gonna be a great answer, but it didn't have citations because of that. I got a note from the team here. Okay, so it says, “The original passage focuses on his innovative use of a sealed gondola. It makes a connection to the perils of a pure oxygen environment,” which is great, but it actually quotes from Gene Kranz. So it gives you a quote from Gene Kranz, the famous flight director. “We had become very complacent about working in a pure oxygen environment. We all knew this was dangerous.” So that's going to be helpful, then it talks about human factors in spacecraft design, and it says Jonathan Yardley's oral history. And it gives a quote and then it says, “This quote, while seemingly humorous, reveals the initial crude understanding of the challenges of human spaceflight, Yardley also discusses the challenges of atmospheric control, which is really important for Picard's flight, and he highlights an incident where a test engineer passed out due to unexpected nitrogen buildup.” And then the Notebook brings up the importance of redundancy as another idea, and it quotes Glynn Lunney talking about redundancy, which was really important in the Apollo fire that they didn't have enough redundancy built into the system. 

So you see what it's doing here is that it's basically been told the prompt is basically saying find three related ideas. And explain what they are and explain how they connect to the original thing you've given it. And it was such a high-level form of intelligence of like making those conceptual leaps and again, it gets back to what we're talking about at the beginning. This is a fusion of so many different separate intelligences coming together here. So you have the original sources, whoever wrote about Piccard that I quoted from, that is an author somewhere who's come up with this idea or written about this person or biographer. There's Steven, who has curated these quotes and put together this idea of the NASA project and gathered all these transcripts. There are the individual astronauts or flight directors who are talking and their expertise. And then, in the middle of it all, there's Gemini Pro and NotebookLM synthesizing all these things and making these connections possible, so it's like, I used to talk about a duet between human and computer, but this is a full chorus, right? It's so, it's so extraordinary. So let's say we're going to pin. So, let me show you this. I haven't shown this before. This is a key part of it. So this is where you save interesting responses. You just pin them, and so that becomes part of your noteboard here. And so we've got some stuff about Piccard. We've got some quotes from Borman and Neil Armstrong. So we've got the beginnings of framing a documentary about this. 

So we're going to select these specific notes that we've curated. And you could imagine this would be a much larger collection, but this gives you a sense of how the new workflow that's possible. And this is the kind of thing I really think that 1 percent of Notebook users actually understand that you can do this. So once I've selected these notes, not only can I summarize or suggest related ideas or create study guides or create an outline based on those specific notes, I can do whatever I want with them by typing into the chat. And so what I'm going to do is say, I'm going to take that original quote about the thing I'm interested in making. “So I'm interested in making this documentary. I'm Steven Johnson.” And I'm going to add, “Based on these notes, suggest an opening script for a documentary episode about Apollo 1 and the pre-history of space flight,” just adding that for the Piccard. I will try, in a little extra twist here, “Suggest images from these sources that could be relevant.” Let's see what it does. So, oh, shoot. Actually, it's not going to work because it's not going to be able to do the images because it's focused on the notes right now. But we'll see what it does. But it should be able to generate a pretty interesting script. 

Yeah, there we go. Okay. So opening script. “Visual: open on black and white footage of high altitude, balloon gondola. It looks surprisingly modern. Narrator: In the 1930s, Swiss physicist, Auguste Piccard, looked at the skies and read the fantastical novels of Jules Verne to imagine a new way to explore the world.” And then we cut forward to, like it says, “A visual black and white still photo of the Apollo 1 crew in January of 1967, tragedy struck on the launch pad at Cape Canaveral.” And look at the transition it does between these two. Narrator says, “Decades later, a new generation of explorers would borrow from Piccard's vision, sealing themselves into metal capsules, breathing pure oxygen as they slipped the bonds of Earth. But the dream of space travel would soon be met by a devastating truth.” 

Dan Shipper (00:54:07)

That's so good. I love this whole journey of Apollo on fire, Piccard, and then we get a little script. Like it's a beautiful idea and it's a beautiful parallel.It’s so cool.

Steven Johnson (00:54:30)

Yeah. This will be coming out in two years, folks. A Dan Shipper, Steven Johnson production. You heard it. Dan only gets 10 percent though. That's what he said.

Dan Shipper (00:54:34)

You're writing the rest of it. I'm just the idea guy. This is wonderful. I really appreciate you taking the time to show us this product. Taking the time to go on this intellectual journey with me. I would love to have you back anytime. For people who are looking to read your latest book or to find you on Twitter or anywhere else on the internet. How can they find you? And tell us about your latest book or can they find that?

Steven Johnson (00:55:00)

Yeah. The new book is called The Infernal Machine: A True Story of Dynamite, Terror, and the Rise of the Modern Detective, so it is very much in the spirit of what you've just seen being created here, weaving together a bunch of different stories about the history of science and technology, but also, political violence and it's kind of a thriller in a strange way in the second half of it. It's really fun. Just in bookstores now. I'm @StevenBJohnson on Twitter and I write the newsletter Adjacent Possible on Substack where a lot of information about NotebookLM will be there. And of course, NotebookLM is NotebookLM.google.com and now available in over 200 countries around the world, as of today. So please do check it out. 

Dan Shipper (00:55:46)

Incredible. Thanks, Steven.

Steven Johnson (00:55:48)

Thanks, Dan.

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn, and Every on X at @every and on LinkedIn.