- 4 hours ago
WIRED Global Editorial Director Katie Drummond speaks with tech journalist and author of The AGI Chronicles about the all things artificial intelligence. Katie and Kevin deep-dive into the AI hype train, the risk to humanity imposed by AI and the “blood feud” nature of AI’s leading companies going toe-to-toe in a race towards superintelligence.
Category
🤖
TechTranscript
00:00Kevin, thank you so much for being here.
00:02Katie, thank you for having me.
00:03Nice to see you again.
00:04So you and Casey recently announced that you were launching a new media company called Machine
00:09Gods Media, not to be mistaken for your podcast of the same name minus the media.
00:14Machine Gods, the podcast, Machine Gods Media, the company, correct?
00:18Yeah, a name so nice we used it twice.
00:20There you go.
00:21So when you first announced that you'd be leaving the Times, you described your vision
00:25for what you wanted to do as one that, quote, takes AI progress seriously, is clear eyed
00:30about the capabilities and risks of powerful AI systems, and tries to empower and entertain
00:35people in the face of radical uncertainty.
00:37I'm curious, as you look at sort of the landscape of tech coverage, of AI coverage, what's missing
00:44from the reporting and commentary that you and Casey feel like you can address, that you
00:48want to address with the new show?
00:50We just feel like it was high time that two men had a place to talk about AI.
00:54I've been saying this for years.
00:55I want more men.
00:56No, look, I think there are obviously no shortage of podcasts and YouTube shows and mainstream
01:02media coverage of AI.
01:04It's the biggest story in the world right now.
01:05But when Casey and I looked out at the media landscape, we saw some issues.
01:10One was there are people who just are getting very famous and having a lot of success, saying
01:15that all this AI stuff that's going on is fake.
01:18It's hype.
01:19It's a giant financial bubble.
01:21No one is using these tools.
01:24They're not going to have any impact on the economy.
01:26You know, open AI is going to go bankrupt.
01:28Anthropics is going to go bankrupt.
01:29But basically, this is sort of a genre of popular criticism.
01:32And it's not just a few people.
01:34This is now I hear this from friends of mine who don't pay close attention to tech news
01:39and just assume that what's going on is just fleeting and trivial and that it will all
01:43sort of go back to normal soon.
01:45But there's another genre of AI coverage that is purely hype.
01:48It's look at the, you know, the 17 amazing ways that the new version of Claude can supercharge
01:54your enterprise SaaS business.
01:55And you can go on LinkedIn and just see like example after example of people who are just
02:01purely excited about this technology and don't really care to talk about the risks.
02:07And we both thought there's like a large gap in the middle for what Casey calls AI realism,
02:13which is basically this idea that you can take AI seriously, acknowledge that the tools
02:19are powerful and impressive and in many cases dangerous, and that you can help people understand
02:26that and demystify this area without sort of slipping into boosterism, and that you can also
02:32have a good time while you do it.
02:34We don't want this to be just a dour take on AI doom.
02:39We want to actually give people a good experience and have a good time.
02:43I have to ask, the show you announced recently is being published in partnership with NPR.
02:48Why NPR?
02:49Why was that the right partner?
02:50A bunch of reasons.
02:51Both Casey and I are big fans of NPR.
02:54We like the fact that they have a broad, independent reach and mandate.
03:01We like the fact that they're going to let us own the show and make the creative decisions,
03:06and it will be a distribution partnership rather than like a full acquisition.
03:11So we will still have some operating distance.
03:14We think this is a really critical time and a really important story.
03:18And we like the idea that people might be in their cars just listening to their local NPR
03:23member station and happen on our podcast.
03:28And maybe that's going to be someone who works in policy, or maybe that's going to be someone
03:33who is involved in local government.
03:35Maybe that's going to be someone who has a very different point of view on AI.
03:39We don't just want to have the sort of opt-in, self-selected tech audience listen to us.
03:44Now, I am going to ask you a gauche question.
03:47A Bloomberg report recently said you both were fielding offers of up to $5 million for the show.
03:54I have to admit that link did travel through Wired Slack at RapidClip.
04:01It's a startling sum of money.
04:03Kevin, how much are you making here?
04:06It's not $5 million.
04:08It's not.
04:08Is it more?
04:09It's more?
04:10It's more than $5 million?
04:11So much more, Katie.
04:13So much more.
04:13Look, they made us a good offer.
04:15We could have gotten more money elsewhere.
04:17I mean, that's kind of what I was wondering.
04:19When I heard NPR, I thought to myself, there's no way NPR is giving those guys $5 million.
04:24With all due respect to NPR.
04:26It is not traditionally where people go to get rich in media is to public radio.
04:30They made us a very good offer.
04:31We loved their new chief content officer, Nadine Zalstra.
04:36She's just a total force of nature.
04:39And we were very excited to work with her.
04:41And we just thought they had a lot of things to offer us beyond just money, like their distribution on
04:47radio.
04:47People don't realize how big radio still is.
04:50The reach of radio, and especially public radio, is still quite large.
04:54We've had this show, Hard Fork, for the last four years.
04:57We built up a pretty good-sized audience.
04:59But The Times owned that show, owns the feed.
05:01So we are looking to grow our show as quickly as possible.
05:05And we just thought that the combination of NPR's commitment to journalistic excellence, their long history, their wide distribution, and
05:14their investment in helping us grow the show was the right combination of factors.
05:18This brings me to one more question that I'm so curious about.
05:21The idea that a great reporter, a great commentator, can spend time somewhere that's sort of quote-unquote traditional like
05:28The New York Times.
05:28They can build a brand.
05:30And then they realize that they can just go do it themselves.
05:34And they don't actually need that institution anymore to exist in the world as talent and to make often a
05:41lot more money than they would in traditional media.
05:44What's your take on that?
05:45I don't have anything bad to say about The New York Times.
05:47I had nine wonderful years there.
05:50It was my second stint at The Times.
05:51I've spent like the vast majority of my career at The New York Times and inside these big media institutions.
05:59I do think we are entering this moment where, at least for some portion of the audience, they want to
06:06connect with individuals more than institutions.
06:09We have just seen this in wave after wave.
06:12I am not doing this for ideological reasons.
06:15I'm doing this because I thought it was a really exciting opportunity.
06:17But I do think that organizations that want to retain and attract very talented people will just need to be
06:27more flexible about the kinds of arrangements.
06:30You know, some people aren't going to want to give up their sub stacks and go inside a media institution.
06:34Some people aren't going to want to sort of have all of their work, you know, published by sort of
06:40one publication.
06:40They'll want to do some things for one place and a few things for another place.
06:44And so I think there are some media organizations that are starting to experiment with different, more flexible ways of,
06:51I don't know, bringing people in partway or having them sort of maintain their independent operation, but also contribute on
06:58an ongoing basis.
07:00I think there are a lot of ways this can work.
07:02But I think it all has to start from a recognition that, like, the journalistic career path, where you, like,
07:08go in in the mail room and you work your way up and you spend, you know, 25 years at
07:13the same employer and you eventually become an editor and then a manager of editors.
07:17And like that has broken down.
07:21And that is regrettable.
07:23I don't think that's a good thing that it's broken down, but it has broken down.
07:26And so I think institutions should grapple with the fact that there's now a generation of media entrepreneurs who just
07:32don't really find what they have to offer all that appealing.
07:35When you think about the talent piece of that, when you think about AI, are you optimistic about journalism and
07:42the industry of journalism?
07:45I am very optimistic about the application of AI to journalism.
07:48Like, that is one place where I have wanted to do more experiments, not with having AI write for me
07:55or, you know, do all my reporting, but, like, ways of extending journalism using AI.
08:03What's an example of an experiment you would love to do?
08:05I have colleagues at the, former colleagues at the Times who have done, you know, incredible sort of document analysis
08:11on a scale that, you know, wouldn't have been possible before.
08:14Using satellite imagery to tell, you know, whether a munitions factory has moved or something like that.
08:21That's the kind of thing that I don't do much in my own life, but that I would like to
08:25see other organizations trying because I think that's really cool.
08:29I have used AI to research and edit and improve my own work for months now.
08:38I have found that very helpful.
08:40I think the caliber of my work is better.
08:43And I would love to see more institutions in media experimenting with using these tools to improve the output of
08:52their journalists, not just, like, you know, filling their websites with slop, but, like, actually helping these be tools to
08:59make journalists better.
09:00I want to talk about your book.
09:02So, the AGI Chronicles, which I have read, it is hugely compelling, and I'm curious, when did you decide, when
09:09did you have that moment where you said, this is a book, like, I'm going to commit years of my
09:14life and my career because there's a book here?
09:17What was that moment for you when you realized that this was a, you know, a big fucking deal?
09:22I know exactly when it was.
09:23It was early last year, 2025, and I was in the car on the Bay Bridge, stuck in traffic, and
09:31I was sort of zooming around from, like, thing to thing.
09:33And it just kind of hit me, like an epiphany.
09:37It was like, I have been following this story in all the incremental detail for years now.
09:43I've interviewed all the major AI researchers and CEOs, you know, spent time with the papers.
09:49I've, you know, gone to all of the companies and reported on what they're doing.
09:53But I sort of realized that there was this larger story that I was missing, that I hadn't really zoomed
10:00out and tried to take a more panoramic view of something that was just weird.
10:06Like, it was, it's a weird story.
10:07And I felt, living in the Bay Area, being immersed in San Francisco tech culture, I had kind of gotten
10:13acclimated to that.
10:14And it no longer seemed as strange to me that there were these companies racing to build the machine superintelligence
10:20that could either save or destroy humanity.
10:23And I feel sometimes like I am in Los Alamos, New Mexico, in 1943, when, like, the Manhattan Project rolls
10:32into town.
10:32And I've just kind of got my, like, my lawn chair, and I'm just kind of looking at the, at
10:38the trucks rolling in and trying to make sense of what's happening.
10:41But, of course, all of this is, is going to be important.
10:45I believe that this technology is important and that the people and the companies who built it will be important
10:49historically.
10:50And so, when I thought about, like, who is actually doing the work of, like, writing all this down, it
10:56was nobody.
10:57Nobody was doing it.
10:58And I just felt like, you know, I would, I think it would be a tragedy if all this just
11:02kind of disappeared in a bunch of signal messages and slacks that auto-delete.
11:08And if we just sort of end up with no durable historical record of this really strange decade in AI,
11:15when things went from not working at all to, like, threatening the future of humanity.
11:20So, that was the job I tried to do.
11:22I spent about a year reporting and writing.
11:25I talked to more than 150 people.
11:26I should have probably taken more time because it was, it was very hard and intense.
11:30But I think what came out of it was, like, exactly what I tried to do was, like, make an
11:36artifact that people and future AI systems can look back at to say, here is how this happened.
11:43Here's who made it happen.
11:44Here were the key decisions and moments along the way.
11:46And the book follows three key companies, OpenAI, Anthropic, and Google, in their sort of pursuit of this technology.
11:52I'm curious, what were sort of your big picture learnings about those companies and sort of the key differences between
12:00them that you think it's important for people to know?
12:03Yeah, I mean, the companies are very different from one another, both in the sort of makeup of their personnel
12:08and also in their ambitions.
12:10Let's start with Google because they're the oldest.
12:12They have had for decades now an advanced AI research effort.
12:19But they were pioneers in AI.
12:21They developed the transformer, which is the T in ChatGPT, the sort of foundational technology that all of this other
12:28stuff rests on.
12:29And then there were these two guys, Elon Musk and Sam Altman, who got very worried about how well they
12:34were doing, about the fact that Google and DeepMind were racing ahead.
12:39And they decided to start OpenAI.
12:41It's so funny to imagine that now.
12:44I mean, it's wild.
12:46And we have the emails.
12:48It's like, it's all there in the record where they're basically like, we have to start a lab that's going
12:53to sort of beat them or at least challenge them so that they don't kind of run away with the
12:57whole game.
12:58So they start OpenAI and they do a couple years of that.
13:01And then this guy at OpenAI, Dario Amadei, he takes six of his colleagues and they leave and start Anthropic,
13:08basically to make sure that OpenAI doesn't get to the critical threshold of AGI first.
13:14So the whole industry sort of spawned out of itself.
13:18These people all used to work together.
13:20And now they run these companies that are in some ways mortal enemies.
13:26Like, I was shocked.
13:27This was actually my biggest surprise was I thought this was more like Coke and Pepsi.
13:31This is not a buddy-buddy industry.
13:34This is like a blood feud.
13:36From all of the reporting that you did, who do you trust?
13:39Who do you trust with our future in the context of artificial intelligence?
13:45I don't trust any single person.
13:48None of them.
13:48But I think what I learned through reporting this book is like, these are people.
13:54They are flawed.
13:55They are fallible.
13:56Their motives are never 100% pure.
13:59Some of these people are quite nice.
14:01Some of them are very thoughtful.
14:02I think we have, in some ways, gotten very lucky with the people who are running these AI companies, who
14:08I think are, on the whole, much better suited to build powerful technology and release it into the world than,
14:13like, the social media barons were.
14:15I think they are.
14:16You see a marked difference between sort of the Facebook era and this AI era.
14:22Oh, yeah.
14:22I mean, I think, for one, they are just way less naive.
14:25You know, the social media guys came in and they said, we're going to change the world.
14:28We're going to, you know, we're going to free communication from the sort of the bottlenecks that hold it back.
14:33We're going to distribute the benefits of technology to everyone.
14:37And they really didn't start thinking about the problems until they were being, you know, questioned in front of Congress.
14:43I mean, I will say some of these AI guys talk a lot about saving the world.
14:49They talk a lot about how great this will be for humanity.
14:52You know what I mean?
14:53If you go back and look at, like, the founding emails of OpenAI, they have been very consistent that they
14:59think this is a potentially very dangerous technology.
15:03Now, they're racing toward it, so maybe their words don't mean that much.
15:08But I think you can't accuse them of being naive because they just have such a long track record, all
15:13of them, of saying that, yeah, this could be great for humanity, but it's not a given that it will
15:17be.
15:18And we need to build it really carefully and thoughtfully to make sure that it's actually going to turn out
15:22well for us.
15:22What kind of responsibility do you think falls on their shoulders in the context of AI safety?
15:30Yeah, and this was something that I've asked all of them about at various points.
15:34Like, why don't you just stop?
15:36Why don't you just slow down?
15:38Why do you get up every day and try to make these systems more powerful if you're worried it could
15:42end the world in some cases?
15:44Some of them, including Dario and Sam, genuinely believe that this technology is inevitable.
15:49The recipe for AI is not that hard.
15:52You get a lot of compute.
15:54You get a lot of data.
15:55You build the right scaffolding and kind of grow the model in this organic process of stochastic gradient descent and
16:03reinforcement learning.
16:04And out comes a super intelligent model.
16:06But that is their sincere belief that it's not that hard to build this stuff if you understand the basics
16:12and that because it's not that hard, someone will do it.
16:16And whether that someone is a U.S. AI company or a Chinese AI company or a terrorist group or
16:25an academic research lab, someone will do this.
16:29And so it is in the best interest of humanity for someone who thinks a lot about safety to be
16:35the first to get there because they can sort of set standards for the rest of the industry.
16:39Now, I don't know if I fully buy that, but that is their logic.
16:44And that is the reason that they feel sort of validated getting up and doing this every day.
16:48The book ends in an interesting way.
16:52And I have a couple of questions about that.
16:53I was just looking at the last page before I came in to do this interview.
16:57And it ends essentially without giving away any spoilers.
17:00You essentially saying, really hope these guys get it right.
17:04Really hope that they slow down so that I can sort of keep living my life the way I live
17:09it now.
17:09It's funny timing that the book is coming out right as these very acute and severe conversations around AI safety
17:18are happening.
17:19And it feels like every other day a company is disclosing some new breach, something that went wrong with one
17:27of their models.
17:27How do you think about that?
17:29How scared are you right now, Kevin Roos?
17:31I am actually feeling quite hopeful right now relative to where I was a few months ago.
17:38And it's largely because we are now having this conversation.
17:43There have been people, including many of the people I spoke to for the book, who have been worried about
17:48runaway AI, rogue AI, possible AI takeover for 10, 15 years.
17:57Who have been trying to sound the alarm about this.
18:00And no one believed them.
18:02Outside their little bubble of AI safety people, everyone was sort of like, yeah, yeah, yeah.
18:07And now I feel like we have finally reached this point where this stuff is inside the Overton window.
18:13I think that some of what I see on social media and some of what I see, you know, in
18:18the news media is really hyperbolic and is really dramatic.
18:22And I'm not saying that the stakes aren't dramatic.
18:25I think my assessment of the situation, though, is that if something does go horribly wrong with artificial intelligence,
18:31it will be much more boring than we might think.
18:36And it will have more to do with human error than maybe we are sometimes attributing.
18:42That's what I think.
18:43Yeah, I totally understand that.
18:45And to be clear, I'm not saying I know exactly the risks that we should be most afraid of.
18:50Yeah.
18:50I just think there's this whole category of risk that has kind of been written off or downplayed or just,
18:55you know,
18:55those are just those weirdos in Berkeley, you know, talking about it that we can now sort of have conversations
19:00about.
19:01People are worried.
19:02They're taking this seriously.
19:02And I think we really have a window here where we can actually make some changes or take some steps
19:09to make sure that this goes better for humanity.
19:12I don't think that was possible a couple of months ago.
19:15So I think this is that that's why I'm feeling more hopeful, because even though I think objectively, like the
19:20models are getting worse and scarier and more dangerous,
19:23we are also much better positioned to recognize and talk about and perhaps prevent those risks.
19:30I think that you, you know, in the book, in that last page that I was just looking at, I
19:34don't want to ascribe an emotion to you,
19:36but I felt fear there from you and a reluctance to see your life change too quickly.
19:44And I think that that is something that is very much universal.
19:48Absolutely.
19:48And I think this is where the people inside the AI bubble do not understand the world.
19:55I think the people at the AI companies building this technology are generally people who enjoy the prospect of large,
20:04unannounced social change.
20:07Large, unannounced social change.
20:10Hard to imagine a bigger nightmare for me personally.
20:12Like, they love when things get weird.
20:17That's part of why they moved to San Francisco.
20:20They want to live in the future.
20:21They want the prospect of radical upheaval does not scare them.
20:26Doesn't that seem like a huge problem to you?
20:29Yes, because normal people don't think like that.
20:33Like, normal people want to live a life that is recognizable to them.
20:38They want their kids to grow up in a society that resembles the one that they grew up in.
20:43We don't manage change very well as a society.
20:47We never have.
20:48But I think this is why I wish that there had been more people involved in the more types of
20:56people involved in the critical conversations around this technology.
20:59Because, you know, as I said, 50 people in San Francisco, give or take, made all of the relevant decisions.
21:06And they are not a representative sample.
21:08They are very weird.
21:09They are very, like, they see the world differently.
21:12They have a higher tolerance for change than most people.
21:15And I think it led to them, like, making some decisions that we can't really take back now.
21:20I mean, do you think that if two years ago we had had more philosophers, more artists, more creatives, more
21:27journalists, lawyers, you know, whoever it may be involved in those conversations, that it really would have moved the needle
21:36when there is so much money on the other side of that conversation?
21:41I guess what I'm saying is that, sure, it's all well and good for a broader coalition to be having
21:46those conversations.
21:49But Greg Brockman's donations to the Trump administration get him that phone call, right?
21:55They go to the inauguration.
21:59They have the president's phone number.
22:00I mean, to be fair, a lot of journalists have that phone number, too.
22:03But you know what I mean, right?
22:04The access to the decision makers is bestowed upon those 50 people in San Francisco by virtue of their wealth
22:11and their power.
22:12I think it can help at the margins.
22:13And I'll give you an example.
22:16One of the people I write about at some length in the book is Amanda Askell.
22:22She is a longtime employee at Anthropic and was at OpenAI before that.
22:28You just had a story in Wired about searching for the most powerful woman in Silicon Valley.
22:32I think she's got to be up there in the top two or three.
22:35She has been in charge of Claude's character.
22:39This is her.
22:40They call her the Claude mother at Anthropic.
22:42She's a virtue ethicist.
22:44She has a PhD in philosophy.
22:46She went into AI specifically to think about this question of what should a good AI system do?
22:54How should it act?
22:54What values should it represent?
22:58How should it decline to do certain things or volunteer to do certain things?
23:03How can you instill something like virtue ethics in a chat bot?
23:08And this was a very fringe area of research.
23:11She was, to my knowledge, the first person ever to do this kind of work inside one of these AI
23:16companies.
23:17And it has resulted in them having a really sophisticated way of thinking about the morality and the ethics of
23:25Claude.
23:26She now has a whole team.
23:28They have many people that are devoted to this.
23:30And I think it has probably made Claude not just safer and more better behaved,
23:36but has also kind of inspired other labs to hire their own philosophers and come up with their own ways
23:43of training their AIs for something like moral goodness.
23:46So, yeah, I think there's a really strong argument for having lots of people from lots of different disciplines engaging
23:51with this technology
23:52because it should not just be engineers doing this.
23:55As someone with a Bachelor of Arts degree in philosophy, I have to say it is boom time for my
24:00people.
24:01We are coming for planet Earth and we are about to shake the foundations like you've never seen.
24:05You never thought a philosophy major would be determining the outcome of planet Earth.
24:13I mean, there was a lot in the book to me that was, as we've been talking about, troubling, right?
24:19That is troubling, that is scary.
24:23What stands out to you that is hopeful when you think about the book, when you think about the technology?
24:30I mean, aside from the fact that we may actually make some meaningful progress towards regulation,
24:33was there something you discovered in your reporting that made you feel optimistic?
24:39Yeah, a lot of the optimism that I feel around this stuff has to do with science and medicine.
24:45You know, I lost my father to cancer.
24:48I know lots of other people who have lost loved ones to rare diseases that have not been cured,
24:58not because we lack the ingenuity, but because it's just a question of resources and manpower.
25:06And, like, I think that AI can do incredible things for people who suffer from disease.
25:16I don't think that's all marketing BS.
25:19I, you know, I was very moved.
25:22There's a story in the book about AlphaFold, the deep mind protein folding AI system that won the Nobel Prize.
25:29And there's this sort of incredibly touching moment where a mother of a kid who has a rare disease,
25:38a life-threatening genetic condition, writes to the deep mind researchers after this AlphaFold breakthrough
25:46and asks them, like, could this help my kid?
25:52And they have to give her the honest answer, which is probably not, because this stuff takes time.
25:58You have to get it through clinical trials.
26:00Even if you have these amazing breakthroughs in science, like, you have to design the drugs.
26:04You have to test the drugs.
26:05You have to get the drugs approved.
26:06Like, it can be a decade before these things actually make it to saving people's lives.
26:12But, like, I want that to happen faster.
26:14I want there to be more AI-designed drugs.
26:17I want us to get them to market quickly.
26:20I don't want to have to suffer, you know, have people suffer from the same diseases that killed, you know,
26:25previous generations.
26:26So that's where I feel a lot of optimism right now.
26:29You know, ask me in two weeks.
26:31Maybe it'll be something else.
26:32But right now, that's where I'm feeling good.
26:34Depends on how your P-doom is looking behind you.
26:36Now, I'm curious about your own process with AI.
26:39You have published three books prior to this one.
26:42Your first was published in 2009, well before any of this technology was available.
26:48How did it change the book writing process for you?
26:50I know you're very open with how you use AI.
26:53Sometimes too much controversy among your journalistic peers.
26:57But tell us about how you used it for the book.
26:59Yeah, there's a whole section in the beginning of the book about how I did and did not use AI.
27:03Because I thought it was important, first and foremost, to be transparent.
27:05Like, I don't think this business of, like, people, you know, writing books and then people check it in Pangram
27:10and it comes up as AI generated.
27:12Then they lose their book contract.
27:13You know what, Kevin?
27:14I'm going to be transparent with you.
27:15I'm going to do you a favor.
27:17If you haven't already done it, I ran your book through Pangram.
27:20How did it do?
27:21I was like, I really like this guy.
27:22But, like, if we have a scoop here, you know, it's a scoop.
27:25But the book is written by a human being, presumably by you.
27:29Yeah, it was written by a very tired, overworked, underslept human being.
27:35I did also have a researcher help me with this, Jasmine Sun, and had a bunch of great human editors
27:41at FSG.
27:42So the book is fully human written.
27:44But how did you use the tech?
27:46Yeah, a lot, constantly.
27:48So I had a giant notebook on Notebook LM filled with all of my research materials, interview transcripts, magazine articles,
27:59academic papers.
28:01And the most basic way I would use that is just to query it about things that I needed to
28:05know.
28:05So, like, you know, give me all of the stories that I've heard about the pre-training of GPT-4.
28:10And then it would pop back a list, sort of like a supercharged search function.
28:14I also did it to help with some, sort of, reporting tasks.
28:20Like, who would the three people have been in the room when this decision was made?
28:24And what are their contact details?
28:27It helped with organizing my notes with fact-checking.
28:31Actually, at the end, I had a human fact-checker.
28:33But I also had an AI swarm doing fact-checks and catching some things that, honestly, neither I or the
28:39human fact-checker had caught.
28:41Oh, wow.
28:41So we got to fix those in the manuscript.
28:44And then I used it for editorial feedback.
28:46I have a council of clods, I call it, which is sort of my team of clods that are assigned
28:54to different personalities and vantage points.
28:58So I have one who's like a hardcore LLM skeptic who, you know, goes through the draft and tells me,
29:03you know, oh, this is what I object to.
29:05This is what I object to.
29:06You're anthropomorphizing here.
29:08It's not really thinking here.
29:09I have one that's like a sort of curse wild type futurist that wants to, like, you know, expand the
29:15vision of the future in the book.
29:17So, like, some of this was just slop and probably wasted more time than it saved.
29:21Like, there were a couple things that the council of clods said to me where I was like, oh, yeah,
29:24that's a good point.
29:25I should go back and revise that.
29:26And how do you think about the premium on human-generated writing?
29:32I'm curious about this.
29:33Like, do you think in two years, three years, five years, the average person will care that Kevin wrote this
29:40book as opposed to an AI writing that book?
29:44Do you think it matters to people?
29:46I do.
29:47I think it matters a lot to people.
29:49And I know this because they've done studies where when people read a sample of writing that is generated by
29:56AI, but they don't know it's generated by AI, they give it very high marks.
30:00They prefer it to human written text in a blinded test.
30:04But then the minute you tell them this was generated by AI, they hate it.
30:09Their approval of it plummets.
30:10I think this has a lot to do with human psychology.
30:12We like to think that people work really hard to make something for us that represents their authentic view.
30:19So I think that people will still continue to be offended when they find out that their favorite writer has
30:23used AI to, like, you know, to generate their latest book or essay or whatever.
30:28But I think that's basically, like, only if they detect it.
30:32Only if they can tell.
30:35And I don't think they can tell or will be able to for much longer.
30:38So I think as long as Pantgram exists, that will probably be a useful tool.
30:44But I also think that, yeah, in the abstract, people just they just don't know the difference.
30:48And when you tell them the difference, they care.
30:49But before that, they don't.
30:50That's really interesting.
30:51I mean, it is the psychological value of wanting to know that someone's fingerprints were on something.
30:58And wanting to know that you are getting their voice and not the voice of Claude.
31:02Totally.
31:02I think it's important for writers, too, to, like, show their process for this reason, to talk about and maybe,
31:07like, literally film themselves working so that people can prove, like, you can sort of connect with the laborer involved
31:14in making something.
31:15I think that's going to be important for humans of all creative stripes.
31:19Is that going to be a social video series that you and Casey come up with?
31:22Yeah, I'm streaming myself writing 16 hours a day on Twitch.
31:26Go check it out.
31:27Sounds like hell.
31:28Kevin, congratulations on the book.
31:30Thank you so much for being here.
31:31This was fascinating.
31:32Thank you, Katie.
31:32You're the best.