John West with Demetri Kofinas
Show: Hidden Forces
Cleaned and reformatted from published transcript or auto-generated captions — punctuation added, filler removed, restructured for readability. Not verbatim. For exact quotes, refer to the original.
Demetri Kofinas
John West, welcome to Hidden Forces.
John West
Thank you, it's great to be here.
Demetri Kofinas
It's great to have you on, John. I originally agreed to read your book under false pretences. I thought this was a book about the internet. It's certainly not fair to say that it's a book about the internet, though you talk quite a bit about the internet. And I feel like it's also not fair to say that it's a book about life or death. Nor is it really fair to say that it's a book about both. It feels like something entirely different and special and beautiful in a really poignant way. Before we get into it — the title is *The Internet Will Die and So Will You*, which is a great, great, great title. Tell me a little bit about you. What's your story?
John West
Thank you so much for those kind words. I am a journalist at the Wall Street Journal. I do computational journalism, so I do a mix of computery stuff. I rarely am out there doing shoe-leather reporting. I do what I call trackpad-leather reporting, where I interrogate databases much more than I talk to people.
I'm also an author. This is my second book. I have a novel coming out in January. I have an MFA in writing, and I studied philosophy and music as an undergraduate. I am definitely a jack of all trades, master of none. But I really like that, actually, and I think it lends itself well to this moment, when so much of what's out there rewards being a generalist as opposed to someone hyper-specifically focused on one niche area. At least I like to tell myself that. I don't know if it's true, but I like to tell myself that.
Demetri Kofinas
Do you mind telling us when you were born?
John West
Sure. I was born in '86, so I'm almost 40. I'll be 40 in November.
Demetri Kofinas
So what role would you say the internet played in your life as a child, as an early adult, and now in your late thirties? And how has it influenced your sense of self and identity?
John West
I was quite a nerdy kid growing up. I had a lot of obsessive interests that didn't make me a total pariah, but I think it's safe to say I was not very popular. I found the internet as a tween. I wrote my first ever website dedicated to anime, in an effort mostly to make friends and to demonstrate to my new friends that, no, I'm serious about anime too. That experience of sitting in the computer lab teaching myself to indent HTML properly was really formative. I did not think of myself as a particularly mathy or sciency person growing up, so it's been a strange shock to find myself doing pretty complex maths at the Journal, writing machine-learning classifiers and that kind of thing.
It was really important to me that I was able to find people like me online, whether in an AOL chat room or on Tripod or Geocities. Those were places where I really found people and became who I was. And then, as a college-age person, I started a blog. By today's standards it would not be a very impressive readership count, but there was community there, and I made close ties with people I would never otherwise have interacted with. Not that I met any of them in person. I loved the experience of writing in response to something someone else wrote, and being read by those people. That was really exciting and invigorating for me.
Demetri Kofinas
I found your reflections on the early web very relatable, and there were moments in the book where your comments felt almost like a eulogy for the old internet we grew up with. When you first sat down to write this book — or better yet, when you first pitched it to your publisher — how did you describe it? What did you tell them it was going to be about?
John West
I said I wanted to write about link rot. I was really interested in link rot because my blog vanished. It was my own fault in a lot of ways. I forgot it. I neglected it.
Demetri Kofinas
Tell people what link rot is, so they know.
John West
Link rot is the phenomenon where a website goes dark — where you click on that blue underlined link of text and, instead of taking you to a new website, it takes you to a 404 page, an error page. Sometimes there's also this idea of link drift or content drift, where it takes you somewhere you didn't expect to go. Maybe it's the homepage of the site, if you're lucky. Maybe it's a Bitcoin scam, if you're not.
That's the mechanism by which a lot of websites go dark. First it becomes a content-drift situation, and then it disappears. My blog, which I'd kept at Oberlin, turned into a German casino and now doesn't go anywhere at all. That was very sad for me. And I got interested in this idea that the internet we think is so permanent — that we have been warned is so permanent — can be that, but it can also be a deeply ephemeral place. That ephemerality is baked into the very structure of the web.
Demetri Kofinas
When we started this conversation I said it wasn't quite right to describe this book as X or Y or X plus Y. I feel like this book is about the finitude of human life and what it looks like to live a good life, but told through the story of the internet — of its physicality, subject as it is to the vagaries of time and nature, just as our bodies are. How would you describe what this book is about, for those who haven't read it?
John West
Oh, what a vexing question. I'm terrible at talking about it. I did practise before I came on, in an effort not to make a fool of myself.
I do think the book is about our finitude and what we do with it. On my good days I'm able to bring an awareness of my own finitude, of my own limits — that I am not God, that I have these limits. I'm able to bring that to the things I do during my day, and it makes me more attuned and more aware of what I'm doing. It makes me more mindful, I suppose is the word. Whether it's time with my daughter, where instead of being frustrated that she wants to play *The Lion King* for the umpteenth time that day, I think to myself: this is a fleeting moment, and she will only want to play *The Lion King* with me for maybe a few more months.
That awareness of our finitude, which we bring to our in-real-life lives in the best of times — I wanted to bring that to my digital life too, because my digital life feels so abundant and so ongoing, and yet it is marked by the same exact limits.
Demetri Kofinas
I'm probably going to quote the book a number of times, but I have one quote in particular I want to begin with, because it speaks to a theme that recurs throughout and has to do with choice. You write: 'the story of the internet in its modern incarnation is a shift from human agency to platform agency.' Tell me more about that. Where do agency and choice fit into your personal philosophy and ethics, and why does this play such an important role in the book?
John West
The idea is that there is a force outside of ourselves that has its own incentives, and those incentives are in contradiction to human thriving.
I've probably been most influenced there by Marilynne Robinson, who's best known as a novelist but has also written a number of fantastic books of essays — *The Death of Adam* being the one that changed the way I thought the most, and *The Givenness of Things* being the more recent. She talks about the way modernity, secularisation, and what she calls naturalism have let a kind of disenchantment infect our choices.
I think that disenchantment is best represented by these platforms, which are deeply mechanistic in how they think about our behaviour and how they think about us — and that infects our own thinking about what is possible and what is desirable online. If we were to bring the full numinous self into the digital sphere, we would use social media platforms very differently than we do now. And I think that would be hard for the bottom lines of these companies. That is an upshot of their mechanistic view of human reality: it really does inhibit our ability to bring that full self online.
Demetri Kofinas
Do you feel we have a moral responsibility to live consciously and to take responsibility for our own lives? That we've been gifted this fleeting moment of consciousness, and have an ethical obligation to make the most of it — that there's something deeply immoral about relinquishing control over that agency to others?
John West
Martin Hägglund wrote this fantastic book called *This Life*. It's interesting because it's a secular book — he is an atheist — and it talks about how, because we do not have a life after this one, it is our responsibility to use this one in the pursuit of justice and a rich life.
I find it interesting that I resonated so much with that book, because I'm a Christian, and so much of my philosophy has been informed by the more liberation-theology ideas around this life being an important one — that we have a responsibility to bring the kingdom of heaven here on earth. In the sense that we have a responsibility to make this world more just.
That emphasis toward justice is actually what informs my belief that I should spend time with my daughter. It's not a hedonistic pursuit of my own pleasure so much as a belief that being able to pay attention to the domestic, to our own thoughts, to our own introspection, is actually the vehicle by which we practise moral attention — which is the only thing we have going for us, I think.
Demetri Kofinas
To bring it back to the internet's place in this story: the internet isn't just central to who we've become, it is in some very important ways central to who we are. You're explicit about this when you talk about AI, and how large language models have been trained on the internet — so as more of the internet becomes written by AI, does that in some sense mean we are destroying not just the internet but ourselves?
John West
Whether it's 'the medium is the message', or 'we shape our buildings and thereafter they shape us' — pick your version of the way technology shapes us after we create it. I think that is very, very true. Online, we are the products of what we have made on the web. We are so social, and this is a social technology at its core; it's about communication. The way it has changed us cannot really be understated. It has certainly changed me, and I think it has changed culture — and I don't think those changes are necessarily bad. In fact I think a lot of them have been fantastic and magical.
*Small Is Beautiful* talks about footloose technologies — the idea that certain technologies enable more action at a distance. And what is the internet if not a machine to act at a distance? That's a wonderful thing.
I do think we therefore have a responsibility to make it better, to shepherd it and steward it, and not to view it as something given to us but rather as something we are in the process of curating and cultivating. The metaphor of gardening I come back to again and again in the book. Just as nature will produce things if left to its own devices, we have a responsibility when we garden to make the soil richer, and to shape that natural process into something deeply unnatural, actually. The same is true of the internet. All too often we assume the internet just is what it is, and we bring to it what we bring to it. But we are called to shepherd and guide that process.
Demetri Kofinas
You often bring up the regenerative nature of the internet — that it requires our attention and our attending in order to persist against the forces of entropy. That's what your comments about link rot were about: if we don't actively participate in maintaining the internet, it will die. Would you say the early web was more aligned with nature than the modern internet, as silly as that might sound?
John West
The category of the natural is so fraught. I came across an amazing example when I was writing the book, about Ansel Adams — the nature photographer — who studiously avoided pictures of the indigenous people who had shaped the landscape in his photos. Behind the nature photos he was taking were generations of indigenous stewards of the land. What counts as natural is oftentimes really just a question of an aesthetics of the pastoral.
Geocities called it homesteading. Geocities was one of those early Web 1.0 sites that really took off, and it enabled anyone to make their own little plot of land online. They called it homesteading, which I always thought was very funny, because it's a mixed metaphor — it's Geocities, but you're homesteading. There is something exciting about that pastoral quality: you're out there, you're creating it for yourself.
We have allowed a build-up of structure — infrastructure largely built by the internet platforms, but also built by norms and communities — and I think that has overwritten some of that pastoral excitement. But I don't know that I'd call it more natural then versus now.
Demetri Kofinas
I think 'the aesthetics of the pastoral' captures what I'm trying to get at. You look at parts of a city that are wilder, where you can feel the presence of the people, whereas other parts are — to use an Italian word — fascist in the cultural sense: very well delineated, very clean, everything designed from a schema. The theory has been imposed on reality, rather than reality giving birth to the theory. That's what I was drawing at with the early web, which felt experimental and enlivening, versus the modern internet, which feels imposing.
John West
That's a really cogent way of putting it. I haven't thought about it in terms of natural and unnatural, but you're exactly right that there was an ethos in the early web — and you can still find it today, but it is not the dominant mode — where people are just trying things.
You go to some parts of New York City and you see Starbucks, Starbucks, Starbucks, a gym, the same stuff over and over again. And then you go to other parts and you see these little stalls where someone has set up a coffee shop — but it's *this guy's* coffee shop, and it's totally crazy and it's awesome, and it's so uniquely the person who set it up, and it's right next to a store selling secondhand purses. There's something really generative and exciting about the smallness of each individual's contributions, and the way those smallnesses rub up against each other and create something really big and beautiful.
Something like RSS — the technology that powers podcasts, but which also used to be the way people read blogs — is basically just an open standard for how you construct a feed that tells a computer: hey, there's a new post on this blog. That technology enabled a flourishing of blogs, and of different ways of reading the web. And then it faded away and was replaced by closed-off feeds — the News Feed on Facebook, or Twitter, now X, which is even more hostile to external links. There was a much more open, willing-to-talk-to-each-other way of looking at the web that has collapsed into clear delineations between platforms, clear delineations between spaces online.
Demetri Kofinas
I've lived in New York on and off since 2000, but I went to high school on Long Island, and I'd come up to City College when I was doing my Westinghouse project. So I've seen the city evolve. I mentioned the work of Iain McGilchrist before we started — his framing of the left and right hemispheres of the brain, and the schematisation of the world versus actual reality. You have a great quote from *Hamlet* in the book.
John West
'There are more things in heaven and earth than are dreamt of in your philosophy, Horatio.'
Demetri Kofinas
Exactly. That tension between the philosophy and the world — the map and the territory.
John West
I'm obsessed by Borges. I think I quote him about a million times in the book. There is something uncanny about a New York City that is overly schematised, because it's almost like a facsimile of itself. It feels hollow. But I don't just mean inauthentic. It feels like someone has taken the map and somehow constructed a real world to look just like it. The arrow goes from map to place instead of place to map, which I think is problematic.
Demetri Kofinas
What did you mean in the book when you said we need a new way of thinking about attention online — one that takes into account our alienated stuff and our alienated selves?
John West
We have become accustomed to thinking of our digital artefacts as not of us, as not something we laboured over. Someone who writes a Substack post thinks of that post as something they wrote. But someone who likes a comment on Facebook doesn't necessarily think of that as an act of creation, or as putting themselves out there into the world. There's a certain way in which we become removed from what we do online — and so much of what we do online creates a record of itself, and that record does live on for some period of time before it fades into oblivion.
Attention paid to what we produce and what we put out into the digital world — we should have the same care that we bring when we create something in the real world. And all too often, the frictionless way platforms want to treat our relationship to our stuff, and to what is already on the web, makes it glide past our own attentions. I think that's a problem.
Demetri Kofinas
Death is a topic you meditate on quite a bit in the book. How much of the appeal of removing friction from our lives — and, as you say, gliding past our own attention — is a way of distracting us from the knowledge that we're going to die?
John West
I hesitate to be this guy, but I do think a lot of it. As we've gained more — mastery is the wrong word, because it seems like mastery until all of a sudden it isn't. It seems like mastery until a hurricane. It seems like mastery until you realise there's a leak in your house. It seems like mastery until nature reasserts itself. We're still much more at the whims not just of our fellow man but of the implacable forces of the world around us than we like to believe.
I live in a small town, just outside Kingston, New York. I slipped on a walk and dislocated my kneecap, and I was unable to move. I was calling for help, but no one could hear me. Fortunately I had my Apple Watch on and was able to call for help, and the volunteer fire department came out and pulled me out on an ATV to the road, where I was taken to hospital. It was a scary experience. It was a loss of control, for sure.
I think about that volunteer fire department all the time as a remarkable institution, built up to provide support in a world full of contingency that sometimes doesn't treat us well. We build up institutions for two reasons. One is to offer real support. The other is to paper over the horrifying — and also beautiful — fact that we are all going to die. And when we are online, and we have sanded away the edges of moving from place to place, or of reacting to someone, we are in essence acting as though we are truly the masters of our lives and fortunes, when we are very much not.
Demetri Kofinas
How long has this been a subject you've contemplated?
John West
My first book was called *Lessons and Carols*, subtitled *A Meditation on Recovery*. It was about my own struggles with mental health and addiction. I was a pretty happy-go-lucky kid — I was not plagued by questions of existential dread — but I have really struggled with severe mental illness, and that changed my relationship to death. It was an open question for a long time whether I wanted to live, and whether I was going to take steps to kill myself. It landed me on the psych ward more than once.
I'm doing great now — I want to preface this by saying I'm doing very well, I'm quite happy, I feel very content in my life. But that struggle changed my relationship to death, because it opened the door. It became hard to ignore that one way or another I am going to die. And that really changed how I thought about my life, as I became more aware that one of the reasons I struggled was that I believed I had no limits — that I could continue to keep going, keep creating. I developed a lot of unhealthy relationships to work and productivity, because I thought I could just do it.
At a micro level it's important to recognise our own limitations. I'm never going to run a marathon. There are all sorts of things I'm just never going to do at this point in my life, and that's okay. And the biggest one is that I am not going to live for ever. That's the one that stands behind all the others.
Demetri Kofinas
What is the connection between depression and this desire to do everything?
John West
I wanted to prove, not just to myself but to the world around me, that I was worthwhile — and in so doing, to bolster my self-esteem. When you peg your self-esteem to something outside yourself like that, it's always risky, and especially so when you have a melancholic mind to begin with.
It was high school when I first got severely depressed, and I struggled with it off and on through my twenties and a little into my thirties. It has changed, and I don't doubt that I will suffer another depressive episode. I like to believe it will be less severe and more weatherable, as I've got better at sleeping and eating and drinking water, and at meditating and praying and being in relationship with people around me. All of those help. But there's also an element that is maybe unchangeable — some part of my DNA or brain chemistry that is just the way it is.
Demetri Kofinas
I found the portion of the book where you meditate on place and belonging to be relatable, both in the physical sense of home and in the digital sense. The internet I grew up on really does feel like a place — AOL Instant Messenger, clunky websites rendering slowly, the sound of my dial-up modem. How important is place and one's sense of belonging in the story? Can people feel at home on the internet?
John West
Part of the reason I view the web as a place is because I am integrally involved in constructing parts of it. I made a homepage for myself. I've made a million little websites that are really only for me and a few friends, and that's been a wonderful experience.
People have places online, whether it's Reddit or Facebook groups. My partner and I are in some parenting groups I've got pretty involved in, and it's been really wonderful to feel like that's a place — even if it's on a platform I don't trust. It can feel like a place in part because you go back and you see the same neighbours, the same norms and tropes replayed over and over again. There's something about the repetition of social norms that feels homey.
Demetri Kofinas
Physical places change — I've seen four or five phases of New York since I was a boy — but the city still has a physical permanence the internet doesn't or can't. Is there something about our increasingly digital existence, and in particular the time we spend inside these walled gardens, that leaves us more dispossessed and homeless today?
John West
I've often struggled conceptually with gentrification. I am often a gentrifier in the communities I live in. There's a housing crisis in Kingston, New York, and there's a lot of anger among people who have lived there a long time that all these people are coming from the city in this post-pandemic moment and buying up or renting all the homes, which means there's less space for other people.
I think about that loss of agency over the community you are in as one of the hallmarks of homelessness. Displacement is another word. And in some ways, online, we are always at risk of displacement, because we do not own the communities we are members of. The blogosphere was a different place in part because people ran their own blogs — although that went away pretty quickly as Blogspot and other hosting providers came into being. But there was a moment when the blogosphere was held down by a lot of small actors operating in community with each other, with the rules of the road set by open standards, which are kind of like the roads. RSS was a great avenue, and Google Reader was 12th Street. And then when Google Reader went away, they ripped up 12th Street, and all of a sudden the people who lived there had to find new digs.
The displacement we feel online is about a loss of agency. When we change our own communities, that feels really great. When we don't have a say in how those communities are changed, that's what feels like displacement.
Demetri Kofinas
I feel the golden age of the internet was post-Google. Until Google, the internet was this rich Amazonian jungle with thousands of tribes living next to each other who couldn't communicate, couldn't even find each other. Google changed that, because it made everything findable. But as Google evolved, so did its business model, and then social media came and — as you say in the book — shifted the centre of gravity on the internet. How important was social media in particular, and the advertising business model?
John West
Social media sometimes feels like an apt name and sometimes doesn't, and that conceptual slipperiness is important to how I think about it. You had different epochs even within social media: the MySpace era, which felt very different from the News Feed Facebook era, which felt very different from TikTok — which people call social media, and yet the follower graph means very, very little there.
The shift I'd want to draw a lot more attention to, if I could go back and edit my own book, is the moment when it went from 'I follow you and I see your posts' to 'I follow you, and an algorithm sits between you and me and decides which posts I see'. That moment is when the centre of gravity truly shifted.
I'll also say that the idea of an open follower graph remains something that would be a wonderful and potentially generative way to experience the web, if instead of being locked in place it were possible to overcome the switching costs. It's one of those classic cases — I am often predisposed to think poorly of markets and to want to decommodify all sorts of things, but I think there's a useful thing competition could offer on the web. It would be wonderful if other entrants could come in and try their hand at social media networks. But that switching cost makes it really hard to overcome the inertia of Facebook or LinkedIn.
Demetri Kofinas
How has that monetised intermediation affected our sense-making abilities?
John West
So much. I've been trying to use an RSS reader — I think it's called River — where the idea is a choose-your-own-algorithm for how it mediates your feed, and you can set all sorts of personalisable things. This post from *The Drift*, I'm going to read first; that one from the random blog I still follow will hang around longer on the feed.
It's very powerful to have these recommender systems, but they are the servants of two masters all too often. They are the servants of the users, who have real desires, but they are also driving revenue for whatever company built them. And that is often in tension with what I actually want.
Demetri Kofinas
That's also a powerful way of describing the shift of the last twenty or thirty years: from an internet focused on servicing the users to one focused on servicing the advertisers. The modern web works the way it does because we're there as sheep to be shorn, as cows to be milked for these giant corporations. Would you agree?
John West
I do think so.
Demetri Kofinas
And does AI make that worse? In some ways AI holds the promise of improving the epistemic outcomes of the internet, because social media is not there to inform you — it's there to engage you. AI, from the user standpoint, is judged on a completely separate set of criteria: quality, predictability, accuracy of information. Is that a function of the technology, or of the fact that we don't yet have a business model that's going to monetise it?
John West
We have not yet seen — to use Cory Doctorow's word — the enshittification of AI in quite the same way we have of the platform technologies, which have pivoted quite hard away from user interest. Although I will say they would argue they are just revealing the preferences of their users in the market system they've set up. I would dispute that, but that's what they would say.
We have yet to see how AI is going to be monetised. But AI can be a really wonderful technology. I'm very torn about a lot of AI, which maybe came through in the book. I think AI is never going to live up to the promises of its detractors or its proponents. I think we'll always be somewhere doing something weirder.
That said, one of the things I'm most excited about is the way it has empowered a more experimental and creative way of using the web. I made this thing called Mixtab, which I would never have made without AI. I vibe-coded quite a bit of it. I'm an experienced software engineer, but I don't have the time to make this. What I made was a tiny social media site where my friends and I can have bookmarks, and we click a button and it creates a new edition of a newspaper that we share and can talk about in our own space online. I made that in about a week, and I would never have done it otherwise. A friend who's not a coder made a little way of talking to me and another friend about what he's watched — it's like Letterboxd, but just for three people. He just made it. Why not?
There's this 'well, why not just make it' feeling a lot of people are having right now that really feels to me a lot like the early web did.
I shudder to think what will happen when some enterprising AI company decides: what if we just tweak the model to hawk whatever product we're advertising, in line with the recommendations it gives in a chatbot? Or what happens when we actually have to pay the full token costs of what we're doing? I think it's going to be very expensive. That said, I have a pretty top-of-the-line MacBook and I'm able to use local models quite effectively. There's something really beautiful about the idea of a vector space with all of our human knowledge, accumulated and put online for decades, being not just accessible but queryable. It is an exciting time to be a technologist, even if it is a somewhat terrifying time to be a technologist.
Demetri Kofinas
How do you think this changes search and discovery on the internet?
John West
I hope search as we knew it isn't dead.
Demetri Kofinas
Wasn't it already dying?
John West
It was already dying. I used to go on Google and get great search results. Now I just get a bunch of sponsored crap, and I always try to remember to scroll to the bottom of the page where the best results are, rather than the top, which is how Google used to work.
I think about the way people have been appending 'reddit' to their search queries in an effort to get the Reddit results, because that's more trusted than what's natively given back from Google. One of the reasons that appeals to me is that it gets at the idea that the internet is still a social place. We are way far away from PageRank, the algorithm that originally revolutionised Google's search results. There's a lot more going on now when you make a Google search. In some ways, the more social and authentically human you can peg things to, the better it'll be.
I read a fantastic book about gamification — I can't remember exactly what it was called — about how the reductions we make when we metricise and quantify things make it possible to have large-scale institutions, because they make decisions portable from context to context. But they also have this flip side, where they reduce everything down to numbers, which is itself really horrible.
In the best case, the web — and AI — could help synthesise and make human-scale conversation and meaning more portable. Unfortunately that's probably not the way the AI companies are thinking about their products. They're thinking much more about how to win, take-all, the whole thing. But if I'm going to append 'reddit' to the end of my search to make it good again, the reason that works is because I'm trying to find what people are actually saying, and not what a heavily gamified algorithm is going to tell me is the right thing to choose.
Demetri Kofinas
Speaking of meaning — what's the difference between information, or structured information, and meaning? And what role do humans have to play in all that?
John West
Meaning is something we bring to the world. It is not an independent fact of the world outside of it. Data or information could be something external to us — there's information theory about what can be carried — but we make meaning. I don't know that that's a falsifiable distinction. I don't know if there's a way you could convince me it's not true, or a way I could convince someone holding the other position that it is. But I do really think it's an important facet of what makes us essentially human: that we are the meaning-makers.
Demetri Kofinas
That feels central to the debate about whether AI is conscious, or intelligent in the way human beings are intelligent.
John West
There's a Wittgenstein line — I'm going to mess this up — where he's basically saying that the philosophical problems that used to vex him are actually answered by not asking the question.
I think it is not a very useful question whether AI will achieve superintelligence. It is a very useful question to ask what the specific capabilities of artificial intelligence systems are going to look like. Emily Bender wrote a book with someone else called *The AI Con*. I disagreed with parts of it, but one of the things I took from it is that AI is a set of technologies rolled together and packaged as artificial intelligence — a bunch of very different capabilities, oftentimes not even in the same systems, that we're talking about when we say 'artificial intelligence'.
It is very useful and generative to try to shape those capabilities. I don't know if it's actually an actionable question to wonder whether it is a superintelligence. I think it might be a category error, and a lot of the fights we have over whether it's like a person or not are missing the point — maybe I'm revealing my own bias against it being a person — but they're missing the point of what it's actually going to do.
Demetri Kofinas
You wondered openly in the book what the run-up to the Iraq war would have been like if we'd had the modern internet media ecosystem. If we had today's internet in 2001, with the 9/11 attacks — what do you think that would have looked like, and how would it have manifested in consensus reality?
John West
I am glad we did not have the distribution method for conspiracy theories that we have now under 9/11. The 9/11 truthers were a thing — I remember them, and I'm sure they're still a thing. 'Jet fuel can't melt steel beams' was the refrain. I was young enough that I was like, oh, that's kind of weird. But now I think it would be a lot more dangerous. I think we'd see it spread farther and faster and stranger than before.
Something I thought about a lot in the book, which I didn't pin down as much as I wanted to, is this idea that one of the reasons technologies are not as destructive as they're often advertised to be — whether it's the novel, which was going to ruin society, or television, or writing, where Plato famously said we're all going to forget everything, which we kind of did, but it didn't matter — is that we have time to develop social norms around their use and deployment. The internet and AI and social media all move much faster than our norms are capable of keeping up with. That's one of the things that makes them different from television, where we had more time, or definitely the novel, or definitely writing, which took a lot longer to be deployed.
Demetri Kofinas
You also did a good job of highlighting why people are susceptible to conspiracies — because while they may get the facts wrong, they get the feelings right, to quote Naomi Klein in *Doppelganger*. And there's an institutional factor: when people are regularly gaslit by the official purveyors of truth, or when long-standing beliefs they held to be unshakably true are suddenly revealed to be untrue — as in the 2008 financial crisis, or Covid — they become highly susceptible to being reprogrammed into believing an entirely new reality, just as absurd, simply because they need something to hold on to. People cannot live in a perpetual state of uncertainty.
John West
I think about how hard it is at the Journal to say only true things, and not even imply false ones. How much work goes into every article I write, and every article I'm even touching, where it's: I want to say only true things in this article, I don't want to imply anything false.
It is so hard. It's an underappreciated fact that knowledge production is not easy — that it is thorny, and that there are often judgement calls about how far we can get. Can we say that thing we think is true but we're not positive about yet? Knowledge construction is a tenuous human thing that we have to do.
Every so often I see something like: we're going to solve journalism with AI, which supposedly doesn't have these biases, by being more truthful. And I think you have to be an embodied person with other embodied people at some level, and you have to be in the muck trying your best to produce real knowledge from that. That is a really, really hard task, and it's underappreciated just how hard it is. Something I underappreciated until I got into journalism was just how challenging it is to say true things.
You're right that people need something they can hold on to. For me it is much more about process and less about outcome. I try to trust not that the people saying the thing are right, but that they are running a process that is more often right than not.
Demetri Kofinas
That's the key insight. That's why you can develop a protocol for assessing credibility that isn't contingent on your having prior knowledge about a given topic. Of course prior knowledge is helpful, but the process is what matters. The better you are at assessing that process, the better you are at working out who is a good-faith thinker and who isn't.
Demetri Kofinas
In the years leading up to and during the pandemic, part of the American left became epistemically tyrannical — heavily invested in proselytising certain canonical truths that everyone had to accept, and if you were sceptical you could be slandered or cancelled. In response, a big portion of what we now think of as the American right has gone in the opposite direction: if something is officially true, it must be false. There's a catalysing, unravelling energy at the centre of society. I don't know how much of that is really technological, or just made worse by technology.
John West
There is a certain nihilism at the heart of a lot of digital communities right now. I don't think it's exclusive to actors on the right or the left — this dead-eyed view of the world that is very sad. It is often right-coded, but I think it can be an apolitical phenomenon in a lot of ways.
I think about reporting some colleagues did, which I helped with a little, about pro-anorexia online communities that encourage young girls especially to lean into their anorexia. In some ways I was more horrified by that than by any political actor I'd seen before. It's the same impulse: lean into the things society has for so long told us are good, and say no — it is a virtue not to be healthy.
I think about the term 'vice signalling', which I always thought was interesting. There was this idea on the left that people were virtue signalling, and now you have a weird inverse, where people gain credibility by saying increasingly distressing things that even they might not think are okay — that they don't believe, but then maybe they do end up believing at some point down the road. That is a thing I definitely see online a lot. It used to be that you'd say the right thing and maybe you wouldn't believe it, and I actually think that's the lesser of the two sins by quite a bit. To say something you don't believe but think is a good thing to say, versus something you don't believe and think is a bad thing to say — I would much rather have virtue than vice.
Demetri Kofinas
So much of the complaint that produced the current moment is a result of outrage over the hypocrisy displayed for decades by people in power. And as a result many people have been drawn to a strongman archetype who is open about his corruption, because at least we didn't have to put up with both the corruption and the lecturing. But the inability to arrive at consensus truths, and the role internet platforms and their business models have played in enabling this, are an important part of the answer for why we are here.
John West
I agree, and especially the idea that power is the coin of the realm and that there aren't other ways of thinking about how we should act. The destruction of societal norms around goodness and virtue has been driven in part by the internet.
I don't know whether the internet would always have been this way, or whether it's the particulars of the internet we have now. Is it the case that we're just too connected — that when we are this connected, and able to create sub-communities of sub-communities independent of location, we're going to get these extreme reactions? Is fracturing itself a driver of extremism, or is it the particulars? I don't know. I haven't done a rigorous analysis of this, but I see a lot more antisemitism on X now than I did on Twitter before. And I don't know how much of that is contingent — had Elon Musk not bought the platform and dismantled some of the content moderation, would I see that phenomenon?
Demetri Kofinas
And that's because it's good for engagement. These explosive topics are good for engagement, and that's what drives the modern internet.
John West
The engagement algorithm — I made this kind of pointless thing called Leaves of Grass Tok. I took *Leaves of Grass*, the famous poem by Walt Whitman, stanza by stanza, embedded each stanza with an AI model, and then made a little engagement algorithm: the longer I dwelled on each stanza, the more stanzas like that I would get. I thought it was a dumb exercise that wasn't going to work. And it actually kept me engaged. It was remarkable. We have this almost content-neutral way of understanding the world, but it has ramifications for the content we have. The power of it — the way it exercised control over my psyche, even on something that was a poem from the nineteenth century. It works. It's very powerful.
Demetri Kofinas
That's why the framing of free speech versus censorship was always such a bad framing. It never applied to the digital world — certainly not to one of platform economics and mediated conversations, because everything you say is ultimately amplified or suppressed based on the economics and other decisions of the platforms and their algorithms.
John West
The idea that there is a neutral algorithm that we are tweaking to produce results we want or don't want — it's crazy. It is not the case that there is a neutral algorithm that would equally amplify voices on the left and the right. That's a desire you might have. You might want to create an algorithm that does that. But that is not the default state of an algorithm. An algorithm is going to have biases one way or the other — whether towards engagement, or towards your friends, or towards something. There are going to be ways that the construction of an algorithm, which is a constructed thing, will by dint of how it's constructed impact some voices more than others. That is just the truth.
I think there are examples of free-speech infringement on the internet. I'm not trying to say that doesn't exist as a category. But platform-level manipulation of algorithms — it's not a censorship thing. It's something else, and I don't think we have good words or good frameworks for it yet.
Demetri Kofinas
It's about values. If we had the internet I would prefer, it would be one that elevates truth — that would be one of the core metrics of success. For some people that's not important, or not what they value. So it's really a reflection of values. The modern state of discourse on social media reflects the values of our society, embedded in the algorithms that promote or suppress the voices of the people who use the platforms.
Demetri Kofinas
John, this was a great conversation. I really enjoyed reading this book. As I said at the beginning, I was brought in under false pretences — I thought this was going to be a book about the internet. It wasn't exactly about the internet, but it also kind of was. The title is *The Internet Will Die and So Will You*. When does the book publish?
John West
September 10th.
Demetri Kofinas
So people can pre-order it when this episode comes out. And if they're interested in your previous book, that one is titled *Lessons and Carols: A Meditation on Recovery*. If people want to follow you and read your other work, how can they do that?
John West
I'm active on Bluesky and LinkedIn and Instagram, sort of. John West at all of those places.
Demetri Kofinas
Well, thank you so much for coming on, John.
John West
Thank you.