← New Books Network
New Books Network
Cory Doctorow on Enshittification: Why Everything Suddenly Got Worse and What to Do About It
01:36:34
Loading summary
Transcript137 lines
- [00:02]
A
This episode is brought to you by Netflix from the creator of Homeland. Claire Danes and Matthew Rhys star in the new Netflix series the Beast in Me as ruthless rivals whose shared darkness will set them on a collision course with fatal consequences. The Beast in Me is a riveting psychological cat and mouse story about guilt, justice and doubt. You will not want to miss this. The Beast in Me is now playing only on Netflix. This episode is brought to you by indeed. Stop waiting around for the perfect candidate. Instead, use Indeed sponsored jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C According to indeed data, sponsored jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit at Indeed.com podcast terms and conditions apply.
- [00:56]
B
Extra value meals are back. That means 10 tender juicy McNuggets and medium fries and a drink are just $8 only at McDonald's for a limited time only.
- [01:07]
C
Prices and participation may vary. Prices may be higher in Hawaii, Alaska and California. And for delivery welcome to the New Books Network.
- [01:16]
B
Welcome to Peoples and Things, where we explore human life with technology. I'm Lee Vincel, Foreign hi everybody. Welcome to a special live stream edition of Peoples and Things, a podcast about human life with technology. I'm your host, Lee Vincel, an associate professor of science, technology and society at Virginia Tech. I'm coming to you today from the Athenaeum at Virginia Tech, and as always, this show is made possible by generous support from Virginia Tech Libraries and Virginia Tech Publishing, and I am especially grateful to the hard work of my friends, producer Joe Fort and editor Mandy Lam. If you're interested in programming on technology and society, you can subscribe to our channel here on YouTube or on whatever podcasting platform form you prefer. We're here today to talk about decline and its causes. And as you know, our culture is full of stories of decline, what are sometimes called declension narratives. It's a genre that goes way back in history, in both the east and the west, and there are a lot of ideas out there that tie decline in our culture and society to technological change. You can think of people like today, like Jonathan Haidt or Nicholas Carr, or back in the day, Neil Postman. And in general, I think we should be skeptical leery of stories of decline so often anything they don't contain much of anything like thinking. It's more like our prejudices coming out, our unhappiness, even our mental illness at times. But I am interested in specific, describable, when possible, measurable forms of decline. Regular consumers of the show will know that I've been working on a history of economic hardship in the US from the seventies to the present called A Good History of Shit Jobs, and in such a work you just have to wrestle with the complex and at times inaccurate notion of de industrialization, including how certain regions lost population, production, wealth, you name it. That shit's real. So I'm interested in phenomena like ghost towns, technological obsolescence, and plant closures, and I wrote about these things in a recent essay you can find below in the show description called Disinvestment and Decline in Infrastructure Studies. But let's talk about another form of decline, Digital platform services degradation. I think you've probably encountered this before because it's let's be real. Twitter sucks. Facebook sucks. Google search has gotten worse. Or, if I'm being honest with you, the one that's bummed me out the most recently is just how bad Google Books has gotten. Now, if you're listening to the audio version of this podcast, then you can't see that right now I'm wearing a helmet. And that is to protect myself from my librarian friends who become enraged when they learn that I typically use Google Books instead of logging into the very expensive academic databases that my university pays for to do my work. But in the past year or so, if you put an author's name into Google Books, their goddamn books don't even show up in the results you get. Like, you might get books about them, but you don't get their books. Though I should be clear that in the last couple weeks this seems to have gotten better again. But what's going on here? Why are digital platforms getting worse? It's almost like they're. I don't know, they're getting inside. I suspect many of you know that we are all in luck, because we have with us the writer of science fiction, nonfiction, and much else, Cory Doctorow, whose new book, why Everything Suddenly Got Worse and what to Do about it, lays out a theory of digital platform decline. To make this conversation even better, I've brought in my friend Dana Boyd, who worked at Microsoft Research for years and years but has recently switched over to a job at corporate Cornell University. Dana is the author of the classic book It's Complicated the Social Lives of Networked Teens, and she just turned in the manuscript for a new book, Data Are Made, Not a story of politics, power, and the civil servants who saved the U.S. census, which will be out next year. She also hosted a Totally fascinating and amazing workshop at Microsoft Research on the theme of degradation. And she's been my greatest thinking partner in thinking about decline, among other things we'll talk about in her forthcoming book and in an essay I've pasted below called what Game Are We Playing? She writes about what she calls Jenga politics, which is how after experiencing wave after wave of budget cuts and assaults, federal agencies become more and more precarious. So I'm very jazzed to have Corey and Dana here, and I hope you're jazzed, too, or as we like to say around here, hey, get excited. Dana, thank you so much for guest hosting with me. Can you tell the people a little bit about your work? And, you know, what's this idea about Jenga politics? And where are you at with decline these days?
- [06:50]
A
Yeah, no, it's such an honor to be here with two dear friends who I'm very glad to spend the evening with. You know, I ended up studying the census in the weirdest way possible. I didn't even know what really how the census worked or what was going on. And then, you know, a series of, you know, accidental moments ended up with me talking to a bunch of government officials and saying, like, hey, there's this big AI fight happening, and I really want to understand what make data legitimate. And this, you know, the senior executive at the Census Bureau was like, you mean the problem we've been studying for 230 years? And I was like, point. And so I dove in to try to understand as an ethnographer what is involved in making these data. And, you know, there's beautiful stories that, you know, all your STS folks would love, and hopefully they will read the book next year. But one of the things that I kept coming back to was how critical it was for, you know, organizational infrastructure to be, you know, viable and resilient in order to produce these data that, you know, our governments produce in order to anchor, you know, our democracy. And I started to realize that the system was really brittle in ways that I really didn't understand. And this was even before the pandemic. And so I ended up coming back to the metaphor of Jenga. And if you think about the wooden tower, the whole game of Jenga politics is that some or of Jenga, sorry, is that somebody takes out pieces of the tower and puts more pressure on top, and it's all set up so that the tower will eventually fail, right? The game ends, and the person who pulls out the last piece that makes it fall loses. But actually, if you think about what it means to think about our infrastructure in society. Everybody loses. And so I kept coming back to this metaphor as I was thinking about how our societal structures are set up so that we're participating in different ways and putting pressure and taking pieces out, hoping that somebody else will repair it. And in the case of the census, it's actually civil servants who run around trying to be like, I can keep the wobbly tower together. I can keep the wobbly tower together. Somehow the wobbly tower won't fall. But of course, they're in. They're in deep shit right now. And so this was. You know, this. This book, in many ways, is a. Is a love letter to civil servants of. Just like, here's. You know, I recognize. I see you. I know what you've done to make this work on our behalf. But it also does connect to my broader interests, like both of you, in how things fall apart. Because we can look back historically and be like, oh, they fell apart. And we try to come up with these explanations. And of course, as STS scholars, we're like, the future is not preordained. There is not a guaranteed thing at the other end. But there's a lot of choices and actions that we make that create the conditions for normal accidents, for things to fall apart if we're not smart about it. And there's things we can do to reduce those future possibilities or reduce the likelihood of them. And I think part of what I'm so excited about with Corey's work is this reminder, like, there were these preconditions that existed long before our search engine experiences were so miserable. And it's not that they were inevitable per se. People could have stepped in, policies could have been different, engineering could have been different. But why wasn't it? And why did we see this trend over and over and over again? And I think it's this reminder, like, when we're in the middle of things going to shit, we need to look at how the actions and practices and the configuration of that situation is playing out. And so that's what I think that draws a lot of us together in this sort of passion for degradation.
- [10:32]
B
Totally. That's great, Dana. I think we'll be returning to a lot of these themes throughout the conversation, and we'll probably. We'll save some time for Q and A at the end, but if folks want to put questions for us in the chat, we'll get to them along the way. Corey, why don't you. I mean, why don't we just start. Why don't you tell us a bit about the insurification book and, you know, like, just give us a brief picture and then we'll hop in then from there.
- [11:00]
C
Sorry, it's not a zoom conference until someone talks with their mic off. I've now that I've gotten that out of the way. Um, so, you know, I'm very excited to be on the air with you two tonight because I respect your work so much and, and it has been so informative in my own work, I should say. I'm not an academic, I'm an activist. And I think that gives me a slightly different outlook. But learning from academics is very important. I think as an activist, it. It really informs my work and, and helps me be better at trying to make the world different. And, you know, I've worked with the Electronic Frontier foundation, which is a nonprofit digital rights group for a quarter of a century now. I started there 24 years ago, most of my adult life. I'm 54. And that is a job of trying to increase the salience and urgency of questions of digital policy at a time in which, for most people, they are very abstract and distant and complicated and technical, and they will not become immediate and at hand until things are so bad that it's probably too late to do anything about them. And so, in service to avoiding that point of no return, I have spent most of my adult life now coming up with metaphors and similes and framing devices and parables and so on, and insidification is one of them. And it's a way of talking about platform decay. Platforms are the endemic form of commerce or business or service on the Internet. A platform is an intermediary. And it's kind of funny that that's the Internet's primary form of business, because, of course, the big promise of the Internet in its early years was disintermediation. And here we are relying on intermediaries. But in retrospect, I don't think we were ever angry about intermediaries. I don't think anyone was like, well, the best thing about the Internet coming along is that each of us will be able to gnaw our own web server out of a whole log using our own teeth. Right? I think we all wanted. I think we all wanted to have at least some of the heavy lifting done for us by someone else. And to the point of your friends, I bet if your most Unix bearded friend and I sat down and I ran down all the different aspects of how they communicated with the rest of the world, that we would find something where they used an intermediary. And they would say, yeah, that's not the part that I think is interesting or important or sensitive. And you know, in the same way that privacy is the right to decide which of your data is disclosed and not to have none of your data disclosed. So to is disintermediation, I think just a shibboleth for I would like to choose which intermediaries I use. I would like to be able to change which intermediaries I use, and I would like those intermediaries to serve as helpmates and not to usurp the relationship between me and the other people that I'm trying to do something with. And I think that's why we were anxious about disintermediation, because I think the intermediaries of the world in the 90s had become gatekeepers and I think that's where we've landed now. And I think that when an intermediary goes from being a helper to a gatekeeper, it is irresistibly tempted to harvest value for itself at the expense of the parties that it is meant to serve. And in the case of our modern digital platforms, the inshidification thesis identifies a three stage process of platform decay which starts with platforms first being good to their end users, but finding a way to lock those end users in. And once those users are locked in, you move on to stage two, which is making things worse for those end users, secure in the knowledge that the lock in means they're not going anywhere. And you make things better for business customers. And in stage three, those business customers which have flocked in to be, you know, in service to those users and to make money from them, they are also locked in and have all the value withdrawn from them as well. And what you end up with is a pile of shit. You end up with a platform where all the value, except for a kind of mingy homeopathic residue has been taken away from the users. The users are locked in to each other, the businesses are locked into those users. All the remaining value is delivered to shareholders and executives. But I think the most important thing about insidification is not this descriptive part, but it's theoretical component. And the theoretical component is a thesis about why this is happening now and why it's hard for us to leave anyway. Right? Why? Why these platforms are so bad, but we seem stuck to them. I think it is a critical response to other theories that. Lee, I know you've written about this idea that, you know, the platforms were founded by evil dopamine hacking wizards who perfected a mind control ray using big data, and that keeps us from leaving. And I have a far more parsimonious explanation, which is that the platforms are monopolies and they are able to hold prisoner things that matter to us and that are more valuable to us than the things they extract from us. That we love our friends more than we hate Mark Zuckerberg, and that keeps us from going. And to what Dana was saying earlier, the book is also an attempt to recover, I think, a lost history of the policy choices that led to this and the individuals responsible for them. Because I think that the neoliberal minds app encourages you to think that the reason we have these bad platforms is because you, the almighty consumer who is the secret arbiter of all activity in the world, failed to shop with sufficient care and having voted incorrectly with your wallet, you have ushered in this world. If only you had shopped harder, we wouldn't have this problem. And I don't buy that. I think that's wrong. And I think it's a council of despair. I think it makes people miserable and also makes them into freakish consumption fetishists who spend endless hours worrying about whether they should be on Blue sky or Twitter instead of doing something meaningful. Yeah, and I think that there's like a second order version of this, which is that, oh, well, the problem is the cupidity and wickedness and greed of the tech bosses. But we did not invent greed in the middle of the last decade. And while these like Zucker, Muskie and Ketamine adult failures are definitely awful people, it doesn't explain why they ended up on top. And my thesis finally is that we have this policy, these series of policy choices that created an enshytogenic environment that ensured that these bad people would always be on top. That, you know, if Elon Musk overdoses on ketamine tonight, that after the succession battle between 10 horrible big balls, whoever ends up filling his seat will be indistinguishable from him. And that ultimately we have to recover these. This history of the policies, not merely so that we can hold the people responsible for these policies to account, although we should. They are sitting around polishing their fake Nobel prizes in economics and consulting for six figures for blue chip companies, and they should in fact be pariahs, but also so that we can reverse the damage they did. We don't have to be prisoners of the terrible policy decisions of the first two decades of this century.
- [18:39]
A
Well, part of it. You know, I was thinking about your arguments specifically around Monopoly. And I think that's what Lee is sort of edging me on around, because one of the things I've struggled with with the monopoly argument is that I think that it's missing another backstory. Right, and which is the financialization backstory that created the conditions for monopolies in many ways to be inevitable. And I keep thinking about all of the other policies. You know, you focus a lot on these tech policies, but I'm thinking about the financial policies and how we failed to learn lessons from different parts of the 20th century, early parts of the 20th century. And so I'm curious, you know, I don't think this is not something like you're going to massively disagree with me on, but I'm curious if you could help everybody unpack, you know, weird like looking behind monopoly, looking into venture capital, looking into, you know, the investment structures. What are the things that you see are critical in setting this stage up in addition to the monopoly component piece.
- [19:42]
C
So I think if you want to fuse the monopoly thesis with the financialization thesis, you do it by mentioning a university, which is Chicago, and the Chicago School of Economics advanced a package of theories that included not just a tolerance for monopolies, but a celebration of them. So theories like the consumer welfare theory and antitrust held that monopolies were evidence of excellence. That if we all go out and we buy all the same thing from the same company, it's because that company is so good that we all voluntarily choose them and not because they've cheated in some way. And they also militated for things like shareholder supremacy. And the idea that firms sole duty is to maximize profits at all expense, which is, you know, one of these things that's, it's a, it's a, it's a wonderfully flexible ideology because it can mean anything, right? Oh, I fired all those workers to free up capital in order to return a dividend to my shareholders. Oh, I hired a ton of workers in order to create, create an expansion that will give more capital to my investors. Oh, I like raised prices because that increased the margin or I cut prices because I thought that increased volume, right? Like there is literally like the idea that shareholder supremacy somehow produces an empirical standard that managers can be held to. It requires like a willful act of brain damage. Like you have to stick a drill up your nose to really believe that shareholder supremacy is somehow going to cut through these policy questions. It's outrageous. So, you know, venture capitalists, right, they have lots of structural advantages that they get to, that they get to engage in, and some of them are down to the concentration in capital as a means of preventing alterations in the policy environment. So every election cycle, there's a lot of stuff about something called the carried interest tax loophole. And I think a lot of people are like, oh, carried interest. That has to do with, like carrying the interest on a loan over from one year to another or something. Carried interest is a law created for 16th century sea captains. It relates to the interest that they have in the cargo that they carry. It has nothing to do with interest. And somehow financiers have successfully lobbied to have this applied to the majority of the income that they make, such that this income is taxed as capital gain at a half of the rate of the rate at which they're their. Their ordinary income will be taxed at. So you have these structural advantages. You have the structural advantage that you can offshore aspects of your business. So you have every tech firm pretending that they're headquartered in Ireland. And then Ireland has these special tax treatment of intellectual property. And so you can do this thing where you take your trademarks, the copyrights in your software, anything else that you can colorably call intellectual property, and you vest it in the hands of a holding company in Dublin. And then the American operation remits a license fee to the Irish company that is equal to 100% of its American profits. And they arrive in Ireland tax free or taxed at an extremely preferential rate. And then the company declares no profit in America and pays no tax, right? Whenever you hear Amazon made $98 billion last year and paid no tax, they've. Well, Amazon's based in Luxembourg, but it's the same deal.
- [23:06]
A
They've just one of the things about these companies, right, the tech companies that we're talking about is that they can move across borders, which of course, in the early 90s, we got to dream that this was a good thing. And we're also seeing the slippage of capital across these different kinds of borders as part of what's at stake here as well. The other thing that I keep thinking about that's unique about these tech companies, when it comes to the economic conditions, I grew up hearing about dreams of cooperatives, right? Like, wouldn't it be great if we were all part of, you know, the businesses that we, you know, were committed to and we all benefited when the companies benefited and this would be great. And the irony to me about what happened in the tech sector is that a class of workers, right, the elite, the tech, you know, engineers, etc. The product people, they in many ways are very much financially tethered to the return of on investment of these companies. And so we end up with this really perverse incentive around this, which is not just that the, you know, you know, it's not just these external investors that are, you know, the shareholders we're paying attention to, but so are most of the employees who feel as though that stock has to keep going up and keep going up. And of course the executives are also worried about losing that talent. Right, because they're, if the stock goes down, they're going to run off to their competitors. And so you end up with this perverse incentives that's happening in my mind in two directions simultaneously. It's coming from inside the house and from outside the house. And I don't think that that's, I don't think there are other industries as explicitly, you know, configured that way. And I think a lot of the configuration that way that occurred in the late 90s was in response to the financialization of private equity and hedge funds in the 80s and early 90s. So I think this is the weird thing for me to watch this structure, you know, unfold and why I think that these companies in some ways are unique compared to our normal frames of Monopoly. It's also.
- [25:06]
C
Yeah, go ahead.
- [25:08]
A
Well, I think the final point here is that, you know, as this configuring work happened, we saw another piece of it and I think we saw it this week with, you know, Nvidia going past 5 trillion, which should give us all heart palpitations because we have a small number of companies holding up the, you know, S&P 500 in the United States for which every endowment from universities and philanthropy is tethered to it. Everybody's, you know, you know, retirement plans, you know, is it tethered to it? And so then we're all getting enrolled in this project in a way that actually I find even more terrifying than thinking of it as a monopoly over there because I find that that enrollment en masse is something that is much harder to figure out how to break through. Meet the computer you can talk to.
- [26:03]
C
With Copilot on Windows working, creating and.
- [26:06]
A
Collaborating is as easy as talking.
- [26:08]
C
Got writer's block.
- [26:09]
A
Share your screen with Copilot Vision to help spark inspiration and use Copilot voice.
- [26:13]
C
To have a conversation and brainstorm ideas.
- [26:15]
A
Or maybe you need some tech help with Copilot Vision.
- [26:19]
C
Copilot sees what you see. Let co Python Copilot talk you through.
- [26:21]
A
Step by step guidance so you can master new apps, games and skills faster.
- [26:26]
C
Try now@windows.com copilot.
- [26:32]
B
Hello friends. Guess who? That's right, it is I, the replacer.
- [26:37]
A
Once again I've been called on so.
- [26:39]
B
You can play the new Call of Duty Black Ops 7 with three expansive modes, 18 multiplayer maps, and the tastiest.
- [26:49]
A
Zombie gameplay you've ever freaking seen.
- [26:52]
B
Call of Duty Black Ops 7 available now.
- [26:56]
A
Rated M for mature.
- [27:00]
C
So let me complicate that a little because I think that when it is true that a lot of tech workers thought of themselves as temporarily embarrassed entrepreneurs and you know, as co equal with the founders, that we're all just engineers here, we all wear hoodies. Mark Zuckerberg comes to the monthly town hall and lets me ask impertinent questions about our software architecture and business strategy. I don't think their bosses ever saw them that way. I do agree, right, that the creation of employee stock option plans and the enlistment of employees into these market based compensations rather than salary based compensations did change the dynamic rather a lot of. But you know that other thing that you were saying about the tech workers being like important and scarce and valuable also is a factor unto itself. That's freestanding in relation to that stock option compensation. So the National Bureau of Economic Research estimates that the average Silicon Valley tech worker adds a million dollars a year to their boss's bottom line. Right? This is why you get these incredible like playground workplaces and also the surgeon who'll freeze your eggs so you can work through your fertile years, right? They don't like you. They're just worried that you'll walk out the door and a million bucks will walk out with you. And I think a lot of those tech workers came into the field because of transformative experiences with tech. Personal experiences, right? Communities discovering the vocabulary to discuss who you are, access to ideas that let you transcend the milieu you were born into. And a lot of them, even a lot of the reactionary ones, got into tech with the hope of trying to spread that to other people. I call them Tron pilled. They want to fight for the user and they have this incredible market power, this incredible labor power that doesn't come from unionization, just from scarcity. And their bosses are terrified of them leaving. And so when their bosses ordered them to insitify things, they were able to say no for a long time. And the problem with getting your power from scarcity and not unions is that when the scarcity goes away, the power goes away with it. And we've had half a million layoffs in the tech sector in the last three years. And so of course, tech workers are either being less mouthy or being more unemployed because there's like five workers will take your job if, if you won't do what the boss says. So I think that's really important. I also think that, like, one reason to focus on monopoly here is that to the extent that all these other policy outcomes, intellectual property, you know, favorable treatment for intellectual property assets, other tax favored treatment, favor treatment for employment relations. Where you see like a lot of these companies now, the majority of their workers are gig workers who effectively have no boss. They're workers without a boss. They can still be fired somehow, but no one is their boss. Which means that there's no one to sue if they get maimed on the job or have their wages stolen. Their wages are stolen by an emergent phenomenon and not by a boss.
- [30:22]
B
Right.
- [30:22]
C
And all of these outcomes are downstream from the ability of firms to collaborate and to collaborate effectively. So, Dana, when you and I first started hanging out, it was in the middle of the Napster wars. And what was interesting.
- [30:34]
B
Yeah, what year was it? Because you guys go way.
- [30:36]
C
2003. 2002, yeah. Yeah.
- [30:41]
B
Wow.
- [30:41]
C
And so in those years, you know, you had 100 tech companies that were in aggregate one to two orders of magnitude larger than seven entertainment companies. And they had their asses kicked by the entertainment companies. And the reason was the seven companies always said the same thing. They always had one message. And not only that, they had a lot of surplus capital, right, because they weren't competing. They had divided up the market. And in fact, whenever their markets got too close, they just merged. And the tech companies were a rabble. And so there were always Vichy nerds who would defect from the consensus and sell out their users. And policymakers just gave all the focus to the entertainment companies, not the tech companies. And the lesson we learned from that was not we need to shatter the concentrated power of the entertainment companies. It was that we need to turn the tech companies into something that's indistinguishable from them. And that concentration is so foundational to all the other horribles in tech policy and in economic policy. Economic policy is a flywheel that reinforces itself, right? It's like you make rich people richer, they corrupt your policy and they get richer again, you know, but I think.
- [31:57]
A
Part of it, you know, as you're laying it out like. No, I agree with you. We've, we've obsessively, you know, fantasized about ideas of entrepreneurship. Right. And I think that that was brought to us. You know, I think about Lee, and I know this woman, Caroline Jack, has written beautifully about the 1970s and sort of getting everybody obsessed with dreaming of entrepreneurship. And I think that you're right that you have a lot of engineers that came into this as to be entrepreneurs. But the thing that I thought was different from when you and I started out in this world versus what I see now is that now you have everybody who wants to be a wannabe investor instead of a wannabe entrepreneur. And in fact, often becoming a CEO is the stepping stool to becoming a venture capitalist or becoming, you know, some sort of, you know, partner in a particular kind of arrangement of this. And I think. I mean, I hear it so much also from the students, right, where they're more interested in the financial structures than in the imagination of what they could build. And so I think this is where I keep thinking also there's another colleague that Lee and I have, Ben Chestakovsky, who talks about, you know, this work. He was sort of embedded inside a startup. And, you know, the founders of the startup did have pretty, you know, nice dreams. And we can discuss, like, the problems with it, etc. But as the venture capitalist really shaped and reshaped their vision, they made the monsters. And that's where I think to your point about why Elon Musk will be replaced by the next monster is because it's not Elon Musk that's actually configuring that reality. It's people behind him whose names we don't know except in insane investment charts that to me are an entangled hairball, are far gnarlier than what we could separate out under monopoly because their investment map doesn't split down like, where's a search engine? Or where is E Commerce? Or all of these other things. All one, which I find much uglier than even just thinking about the names and the brands and the people who we look up and go, yeah, so.
- [34:00]
C
There'S a line in a Terry Pratchett novel, she wants best. She says she wants best brandy, but she'll settle for beer. And I think a lot of these venture capitalists, right, so the venture capitalists demand these abuses not out of sadism, although I think a lot of them are sadists. But they demand these abuses out of a belief that they. And probably a correct belief that they are profitable courses of action. And the profitability of those courses of action relates to things like, is there a competitor who's going to poach all your workers and your Customers, if you do this, is there a regulator that's going to fine you more than you're going to make from it if you do this right? If you. What's a recent example? I'm sure everyone. Well, no, I have an extremely localized feed of social media, which is to say 40,000 people have sent me the story about the guy who figured out that his smart vacuum was sending a bunch of data to the company, so he blocked that port and they bricked his smart vacuum. So, like, that decision, right, is grounded in the idea that if you brick someone's smart vacuum, that their only course of action is to think hard about what they've done and regret it. And if the. If the actual next course of action is to go and buy a new firmware from someone else who ends up selling them their next smart vacuum, they. That's just a thing you don't do. Or if you do do it, you know, maybe you lose out as a result, or you think hard about doing it. You know, when I counsel, reluctantly, counsel entrepreneurs about what they should do in these circumstances, you know, they say, well, I want to start a company and I don't want to insertify. What should I do? I say, you need to take courses of action off the table now, because in the future, you will be prone to temptation and you will be prone to rationalization. And you will say things like, well, the venture capitalists say, I need to destroy the lives of 5% of my users or they're going to cut off the capital. And I've lured 150 of my best friends to come work for me and put their kids more, you know, college funds and their mortgage on the line. Plus those other 95% of users whose lives I won't ruin, they're going to continue to benefit from this. And if I don't do it, they'll fire me and replace me with someone who'll do it cheerfully. Whereas I actually have everyone's goodwill and good health and thriving in mind. So I need to live to fight another day. What you need to do to prevent that from happening is to ensure, for example, that as soon as possible, 51% of your users are not accessing your service on your own. Infra, they're all using the API with another service. So if you ever cut off the API, you lose more than half your users. I think that, like, if we want Blue sky to not become totally inside, at least half of Blue Sky's users have to be not Blue sky users, right? They have to be users on Other servers, you know, open source things. Give your workers a union, right? Do all this stuff where, like, your future self, your future rationalizing self gets into more trouble than they get out of when they sell out.
- [37:01]
A
I mean, I agree with you, and I think this is the thing. I mean, Catherine Bracey, in her book on venture capital, one of the things she argued is that, you know, when we were growing up in the inner Internet, there were many diversified ways in which you could actually, you know, get finance to run your business. Some of those was frankly borrowing from your friend's dad, right? Like, they were very local, they were very small, you could get loans. But the available realm of capital sort of, you know, really narrowed over the years to the point where, you know, you want to start a restaurant, and all of a sudden you're trying to, you know, go to, you know, a venture capitalist, where you're like, this is insane. Because the thing about VC is that it requires that hockey, strict growth. And it means that, you know, I. I look at Facebook and I completely agree with you. Like, all of your diagnoses are completely right in terms of, like, how every aspect of it has become grosser and grosser. But I also look at the arrangement, economic arrangement of it, and I sit there like, what are their options, right? Find more users. Good luck, right? You know, figure out how to actually diversify their profit loss structures by, you know, creating new open markets. Yeah, Metaverse went real well. People or just get sketchier and sketchier and shittier and shittier.
- [38:18]
C
But remember, they want. They want Brandy if they want best brandy, but they'll settle for beer, right? If extraction costs them more than they get. So, you know, when. When Zuck kicked off Facebook in 2006, opened up to the general public and not just Edu kids, he had to contend with the fact that everyone already had an account on. On MySpace. And he did try and pitch those people. He said, like, you don't want to use MySpace because social media, when it's owned by an evil billionaire is bad. And he also said MySpace is really bad because it spies on you. No one wants to use social media that spies on you. I'll never spy on you. But that wasn't enough, right? People love their friends, and they love their friends more than anything.
- [38:57]
A
He also encouraged everybody to think that there were sexual predators everywhere. Let's not forget that piece.
- [39:01]
C
Sure. That's right. That's right. But he gave them a bot, right? And the bot was a scraper. You gave it your login and your password for MySpace. It would go and get all the messages that have piled up for you on MySpace now that you've gone to Facebook and you could reply to them there. So he gave you a mechanism to shift from one to the other. Now there are people who've tried to do similar things to Mark Zuckerberg. Most recently last year, there was an app called OG App that was made by a couple of teenagers. It would log into Instagram as you and it would grab everything in your Instagram feed and throw away all the suggestions and throw away all the boosts and throw away all the ads and throw away all the things from three months ago that are suddenly getting a lot of clicks. And it would just show you stuff from the people you followed who had posted recently in reverse chronological order. And it became in the top 10 of the two app stores within a day. And then that night, Meta sent takedowns to Google and Apple. And because there is honor among thieves, that app went away forever. And, you know, the fact is that if people could leave Instagram really easily, if they could like zero out the ad load on Instagram, like we can zero out the ad load on web browsers, right? You just put an ad blocker in. More than 50% of Internet users have installed an ad blocker in their browser. It's the largest boycott in human history. No one can do that for an app because it's a felony to reverse engineer an app. No wonder the ad load in apps sucks, right? It's. It's like they want best brandy, but they'll settle for beer if they had to every time they put another ad in the app. Have to worry about you installing an ad blocker and never seeing another ad again, ever, forever. Because no one ever uninstalls their ad blocker, then they would just be very careful about putting ads in, right? So, you know, like, yes, they are greedy as fuck and they're awful people and they should be pariahs, but also they are greedy and want more money, not greedy, and want less money. And if they got less money from being greedy, which is a thing that you get by altering the broader policy environment as well as the financial stuff. But remember, being richer helps you screw with the financial regulation too. So we can do some redistribution by taxing them and by breaking them up and so on. But we can do a lot of pre distribution by denying them these rent extraction vehicles so that they cannot accumulate this wealth.
- [41:25]
B
So why don't we, Corey, just so we have the kind of solutions on the table too, just so that, you know, we can kind of play with those tools. Why don't you outline like what you think. How do we change the policy environment so we shift the incentives so we see different behavior out of these folks?
- [41:44]
C
Well, it depends on which policy. Like I, I think that this is all like super contingent and related to each policy. So for example, in the case of privacy, right, we haven't had a new Privacy Law since 1988 when Ronald Reagan made it illegal for people video stores to release your VHS rental history. That's the last consumer privacy law we've had. When we ask like, why does Facebook spy on us so much, the answer is the same as why your dog licks his balls. Because they can, right? You know, we could just stop them. We could just make it illegal and then they, they wouldn't, right? Or, you know, they wouldn't get the capital that they have now to do it because they would always be under threat of litigation and under threat of, of severe penalties and so on. So how do you get to a privacy law? Well, like there's a big potential coalition for privacy, right? People who think that the reason Grampy is a QAnon is because of Facebook. People who think the reason their teenager is anorexic is because of Insta. People who think that all the millennials in their life are quoting Osama bin laden because of TikTok. People who are worried about deep fake porn. People who are worried about January six rioters or anti ice demonstrators being tracked down with their Google location data. People who are worried about, you know, discriminatory lending, discriminatory housing, discriminatory employment. Those are all downstream of the privacy vacuum. So you build that coalition, you can get that done on anti circumvention, right? The law that bans reverse engineering and modification in the US it's pretty hard. There is a path to it which is like state by state doing things like right to repair laws that erode this until it's sort of discredited. But it's a very long run project. But what's very interesting is what's happening internationally in this moment because every country in the world was made to sign up to ban reverse engineering and modification by the US Trade representative under threat of tariffs in order to preserve the rent seeking of American tech companies so they could extract like billions and billions of dollars from those countries. And the thinking was basically like it is better for our economy overall to pay $10,000 a gallon for printer ink and send 30 cents out of every app dollar to California rather than going into business selling third party app stores and generic ink enabling software to everyone in the world, provided that we can continue to export goods to America without paying tariffs. And so happy Liberation Day, right? Donald Trump has imposed tariffs on anyone, everyone anyway, even the ones who did what he demanded. Except for China, who it turns out, are not paying any new tariffs. There you go. And they, and you know, it sets up this moment where like, okay, someone threatened to burn your house down unless you did what they said. And then you do what they said and they burn your house down. If you keep doing what they said, you were a sucker. And in the meantime, Donald Trump has made it really clear that America no longer has trading partners or allies, it has rivals. And that it will use tech companies to gut those rivals in the great game of geopolitics. So, like the International Criminal Court's chief prosecutor swears out a complaint against Benjamin Netanyahu for genocide. Donald Trump denounces it, and Microsoft takes away his Outlook account and he loses all of his email, all of his contacts, all of his calendar, and his access to all the working files of the International Criminal Court. Right. Like anyone who works for any government ministry anywhere in the world should right now be asking themselves, how quickly can I get off of these platforms? And the answer is not very quickly. Because even if eurostack happens, which I think it might, they're putting a lot of money into replicating American tech services. They're not gonna be able to get their data onto eurostack. No one's gonna copy and paste a million documents out of their government Ministry's repository on Office 365 and into Eurostack. For that, you're gonna need to like virtualize a phone in the cloud and do like iteration through the documents and move them over, but automatically or you're going to have to have alt clients and on device bridging or scraping or these other things that you've just basically promised you wouldn't do. Because it would make America sad. Time to make America sad. And to make this dynamic most perfect of all and most heartening of all. This is a one time offer. The first country that takes it, that grabs it, is going to be able to consume all the billions in monopoly rent that American tech companies have amassed and use them as fuel for a rocket to boost their tech sector into a stable orbit. Right, and export these tools for auto repair, ventilator repair, tractor repair, independent app stores, independent game stores for consoles, just, just getting, bypassing these like 500, $600 billion worth of annual rent, right? Just, just getting out of that and selling that to everyone else in the world. Because everyone else in the world, including Americans, wants these tools. And they will get the consumer surplus. They'll buy ink for $5 a gallon instead of $10,000 a gallon. But you will get the industrial policy, whoever is the first country. So this is the moment, right? And I think we're closer to this than we've been in my entire life of making a post American Internet just like Putin created a post Russian gas Europe. And Europe is like 15 years ahead of where they were trending to be on solarization, electrification. Trump is building a post American Internet right now.
- [47:07]
B
In that answer. One thing I heard is we have a question from aria2122 who's asking about open source and free software models as whether you see these as viable alternatives. I mean, I wonder about kind of the role of voluntaristic, you know, consumer choices where you see that play out in this whole picture.
- [47:32]
C
Well, I think they're necessary but not sufficient. Right. So back to Ulysses pacts and taking things off the table before you get started, right? So take the Eurostack. If Europe asks like Siemens and I don't know, Olivetti to make an Office 365 compatible suite, Canada's not going to install it, right? Like Canada wants to run its own infrastructure on its own metal. They have learned the lesson, right? However, if it's free and open source software, Canadian firms and the Canadian state will definitely contribute to it and host it and run it and it will become an infrastructure in a commons. And you know, I think that like this is a fusion of the two promises of open source and free software. So it's a fusion of the open source promise of instrumentalizing transparency to produce higher quality code, right? So you can just like see the code and fix the code. And the free software promise of enshrining the idea of self determination, freedom and sovereignty in the production model of the code, it's not necessarily about voluntarism, right? This, this is also an extremely good model for inter firm and inter government and government and firm and private sector collaboration. You know, we've had some good examples of this like when, when the White House adopted Drupal under Obama, they just threw a shit ton of money at securing Drupal and adding features to it. They just like paid for the kind of security audit that Drupal couldn't do before and everyone around Drupal benefited from it. But it was also cheaper for the White House than Asking, you know, Booze Allen to build them a website.
- [49:21]
B
I have a question here for Dana, actually, as well. So Cultural SF is wondering for me and you, the question is, how are undergrads today treating the insidifying tech careers? Is there a disdain for those courses and jobs, or is the increasing amount of money making it more people popular? And I have to say, on my end, I mean, I think really the biggest worry I hear from my CS undergrads is just about the overproduction of CS degrees at this point is really what they're mostly focusing on. But do you. Do you see any kind of response in your students to, like, what they see around them?
- [50:00]
A
Sort of. So, which is to say, so when I was coming up in college, when I was. When I was an undergrad, the majority of people around me, you know, wanted to go into finance because that was where the money was, right? The fact that you went to computer science, like, oh, like, you're such, like, weird geek. What's wrong with you? Like, and then that changed, but that didn't change because people suddenly woke up and, like, computer science, right? Like, those of us who were started out that way, it changed because that was where finance was, right? Like, that was the shift towards finance. And so what I'm seeing among, you know, students today is they're just looking for, where is the finance industry?
- [50:41]
B
And I think, where's the new hotspot, right?
- [50:44]
A
Where is the money? Where is the. And, and I think for me, that's actually just not different, right? Like, and so that's. I think that's the mistake is we thought it was different because, like, people suddenly liked computer science. They didn't like computer science. They were looking for how to make a lot of money. And I think this is one of the reasons why, you know, you know, my snark to Corey of just, like, people want to be investors, it's not that they want to be entrepreneurs, they just want to be investors, right? They're not product people. That wasn't their passion. And so I think that. And, you know, it's not surprising to me that people are now looking for other places about where the money is at, and they're looking for other ways of making bets. It's the same reason we see students who are into crypto, right? They're not into crypto because, like, the calculations are fascinating, because the math is fascinating, but that's not what they're in for. So I think that there's this. This interesting question about why, for so many decades have so Many of the smartest, you know, young people in the United States seen going towards wherever the finance is as their top goal because this, this has been decades now. And you know, I often think about that in terms of, you know, gently, some of our, you know, colleagues. Like, I think about, you know, Karen Ho's beautiful book on why people went into, you know, investment banking, right. And you know, how they all got socialized into it because it also turned into their social network works, Right. The relationships were so part of it. And I think this is the other thing I think about with regard to the tech industry, Right. Like, I remember when, you know, the tech industry went from being like, interested in product and geeky and problematic and yes, interested in making money to, to the like, wave of Lehman Brother Bros. Like, coming and overtaking all of it, right? And they had different relationship structures, they were friends, fundamentally differently networked than those in Silicon Valley at the time. Today's Silicon Valley looks like the Lehman Brothers of the early aughts. And so I think this is also an interesting Mona, like, you know, where are alternate networks forming and how do we build alternate networks that are more socially constructive, you know, and what does it mean that we have actually devalued nearly every social industry that does good in this world? We've devalued them for decades in different ways. We can talk about education and the devaluation of that, you know, due to gender politics and logics there or nursing. We can talk about the devaluation of civil service. So what do we do when we've devalued everything but finance? Because I think that's also what we're living in right now. And, you know, we're now, of course, kicking out in the United States anybody who, you know, came here to earn a living, you know, because if you. And, and we're still training our students that their, you know, pathway is to knowledge work. So I'm worried about this whole configuration because it's not stable. And I think this is one of the things, you know, when Corey's talking about Canada and Europe, why not? They have a different value system around work. And I actually think that that's also going to be interesting, you know, as the U.S. you know, isolates itself.
- [53:56]
B
Someone named Sierra Dave, who I think might be Dave Karpf, I'm not sure, said, you should write that up, Dana, and turn it into an essay. So what I liked about what you just said, Dana, actually, I hadn't thought of. I mean, there's a theory in certain areas of U.S. history writing, including kind of Marxist writers who think about the asset bubbles that we've had recurrently since the 70s and 80s. One theory is that the economy is largely stagnant in a lot of ways and it's money chasing looking for that growth area. But I'd never really thought about how degree seeking is so tied to that same thing. So I think that's really interesting and in this data.
- [54:47]
A
Hi, I'm here to pick up my son, Milo. There's no Milo here who picked up my son from school. Streaming only on Peacock. I'm gonna need the name of everyone that could have a connection. You don't understand.
- [55:00]
C
It was just the five of us.
- [55:02]
A
So this was all planned. What are you gonna do?
- [55:04]
C
I will do whatever it takes to.
- [55:06]
A
Get my son back.
- [55:07]
B
I honestly didn't see this coming. These nice people killing each other.
- [55:12]
C
All her fault. A new series, streaming now only on Peacock.
- [55:18]
A
Ford BlueCruise Hands Free highway driving takes the work out of being behind the wheel, allowing you to relax and reconnect while also staying in control. Enjoy the drive in blue cruise enabled vehicles like the F150 Explorer and Mustang. Machine available feature on equipped vehicles. Terms apply. Does not replace safe driving. See Ford.com BlueCruise for more details.
- [55:47]
C
Doug Rushkoff has this great little riff about what he calls going meta which is like don't, don't drive a taxi, own a medallion. Don't own the medallion. Start a ride hailing company. Don't start a ride hailing company. Invest in a ride stealing ride hailing company. Don't invest in a ride hailing company. Buy futures, buy options in a ride hailing company stock, right? And one way to think about this is increasing layers of insulation from consequence and increasing layers of insulation from risk. And de risking is an enormous piece of the role of capital markets and you can rarely de risk. Truly mostly what we call de risking is changing who bears the risk. So you know, the gig worker who gets paid only when there's a passenger in the Uber is assuming the risk that a taxi company would have to otherwise assume, which is paying the wages even when there's no rider. And you know, there is a pretty good case to be made that if the taxi company can't convince people to pay for a taxi, right? If they haven't advertised, if they haven't marketed, if they haven't made the case that they haven't priced it right, that's the taxi company's problem, right? It's not the driver's problem. What is the driver going to do to increase the number of riders in their cab? It is literally a thing only the taxi company can do. And so if they do that successfully, well, then the upside is theirs to take. And if they don't, the downside is the drivers to bear.
- [57:16]
A
Well, this makes me think of trustee McMillan Cottam wrote this beautiful book early in her career called Lowered, and it was all about why people went to for profit colleges in the hope of getting ahead. Right. And how this became its own form of like vicious extraction. Because one of the things is that this example you're giving is part of how we've promised all these opportunities to get ahead for people. If you just hustle, if you just get there, you'll get ahead. And people are recognizing that this is an, this is falling down, this is flawed logic. Right. And like, this is one of the things, I think we're in trouble around universities because we, many of us, like, you know, I can't speak for Lee, but certainly I, you know, have this like, nice, wonderful, like, hey, you can become more learned. Isn't this great critical thinking, all of these like, lovely things. And many people are entering into these schools. Like, you have just charged me a gazillion dollars and my only reason to do this is to know that my return on investment is worth it. Right. That is a very different set of logics, you know, than the thing that many of the academics who are teaching in these universities, you know, think about these as. And so this is where, like, where are all of these disconnects? Like, this is an example of these disconnects. And so to your point on de risking, I also think that, you know, there's this sort of weird shift of views also of like, what is the project that we're involved in, whether we're talking about the university, you know, you know, K to 12 education, whether we're talking about health care.
- [58:50]
C
Right.
- [58:50]
A
Like we have a problem where health care is like, you know, we have people going to doctors and nurses because they want to help people, only to realize they're in a financialized project that they despise.
- [59:00]
B
Yeah, yeah, right.
- [59:01]
A
And for the same de risk equation that you're dealing with. And so like, how do we also just generally reclaim dignity out of work that does good for the world rather than work that is just purely extracted.
- [59:15]
C
So, you know, when I am also visiting profit Cornell, that was the last time Dana and I saw each other. And when I was up there, I was talking to a Prof. Who'd been involved in student Research and had done a bunch of surveys of incoming freshmen and students of various years about their views of the university. And what they found was that students really wanted to learn, they really wanted to engage with education as an end unto itself. And they were also terrified that if they didn't get good grades and if they didn't pursue a degree that would produce a return, that they would be economic roadkill, that they had no slack, they had no second chances. And that in fact, going to university was one of these things where you like, I don't know, it's like jumping off a ledge and hoping you catch yourself on the other ledge. And if you don't, you go splat. And that splat is like half a million dollars in student debt with no degree, right? And if you catch yourself on the other side, you got half a million dollars of the debt, but maybe you got a degree to go with it. And you know, it does like kind of answer the question you were just raising, which is, you know, we just have to abolish precarity and have a more fair income distribution. And people will go to university for good reasons and enjoy themselves and when they get there, they will devote themselves to learning. You know, I also think that like secondarily in the university, there is a bit of a crisis in that the freshman year, the first few years are these very large courses often involved with learning how to be an academic, learning how to do academic work. So freshman comp being a really good example. And you're like, with a thousand people in freshman comp writing things that no one, including you, will ever want to read. And you're just doing it so you can be a better writer, but not so that you can write things that are good. And you have to hit a bunch of marks on the, on the assessment rubric to pass the course. And you have to pass the course to progress. No Wonder people use ChatGPT, right? Where if that was like a.
- [61:22]
B
And this is. It's been like that for a long time, folks. It's not chatgpt.
- [61:25]
C
If that was a 15 student seminar in first year though, right? With a teacher really cared about. I don't know how you do this, but if all the freshmen courses were 15 student seminars as opposed to the third year and fourth year classes, those students would at least have a fighting chance of understanding the rationale for actually working really hard on getting better at writing rather than passing the course.
- [61:49]
B
Man, sadly, at Virginia Tech we're headed in the exact opposite direction. So I will not stay on that Sad note for too long. Corey, I wanted to ask you about another solution, potential solution. This one is also from ARIA 2122, which has to do. We mentioned worker cooperatives earlier and so it has to do with, you know, do, do you think worker cooperatives offer an alternative organization structure to deal with some of these problems?
- [62:21]
C
Yeah, 100%. And obviously I think worker co ops, of all the cooperative forums, you know, is the most powerful, certainly far better than consumer co ops. Consumer co ops are so brittle and find it so easy to lose their way. In Canada, the largest consumer co op was an equivalent to REI called mec, the Mountain Equipment co op. And the engagement was so poor that the board was taken over by these venture capital people who would just put up a slate every year until they had like super majority on the board and then they flip the co op to a private equity fund and now it's just called MEC instead of Mountain Equipment Co op. And it's not a co op anymore. Consumer co ops, they just don't require any engagement and so they're really, really easy to capture and make terrible. Whereas worker co ops can actually do some good work in terms of, and there are lots of opportunity for worker co op for things like IT consulting and web hosting and all that other stuff. You know, the stuff that's at the corporate periphery, even security consulting. I think that, you know, it's a co op like another way to say like if you were three lawyers starting a firm together, it would be a worker co op, but you call a limited legal partnership, it's the same thing, right? And like, so there are people who are sort of elite professionals in tech who could do this in terms of like platform co ops, right? If you're going to start like an Uber alike, you really need some path out of the silo that people are in and into your co op. And so otherwise you have this like chicken and egg problem where no driver is going to install the app until there are riders who are using it. And no rider is going to install the app until their driver is using it. But if we could legalize reverse engineering and modification. If we could get rid of the dmca, if we could get rid of its international equivalents, you could totally imagine a thing where a driver puts an app on their phone. They that is the co op app, but it also has like a listener. And if it sees that the Uber app has set up a ride, it goes and it checks a server and that server is run by the co op. And if you are the rider and you have the co op app, you have a similar process. And when you set up the ride, you make a little hash of it and you stick it on the server. And if the driver's app finds that hash that matches, you tear down the Uber ride and you rebuild it as a co op ride. And it just happens automatically in the background and then there's no switching costs. Right. It always makes sense to do the co op app or to page with the Uber app because if you get a co op driver you will use the co op rate and if you don't, you'll get an Uber driver and you won't be standing there for an hour and a half. And so there's this kind of path to glory because you could make the rides cheaper and you could make the compensation better. Everyone would benefit from this because you'd be stripping out the rent extraction of the intermediary. And there are like so many ways we could do this. I don't know if you've ever seen the apps, the browser plugins that look for ISBNs in your Amazon window. And if you load anything that's got an ISBN in the window, it checks to see whether that ISBN is available at your local library and replaces the buy it now button with a reserve at your local library button. And you could easily imagine doing this with co ops where you have small businesses and they all normalize their stock keeping units, their UPCs to the ASINs, which is Amazon stock keeping unit, which is not a copyrightable element. It's functional. So anyone could, could do this. And then all of the inventory of all the shops in your neighborhood are on a server. And whenever you load a page in Amazon, it checks to see whether that good is for sale in your neighborhood. And if it is, it replaces the buy it on Amazon button with the buy it at the shop down the street button. And you know, again, like you could just, you could just treat them as a commodity infrastructure utility and just like destroy them, eat their lunches. You know, Jeff Bezos said your margin is my opportunity. Their margin could be our opportunity, but we need to.
- [66:17]
B
What do you think is it is. Do you think worker cooperatives work under different incentives? And that's why what keeps them from initiative is, I mean, is it, is it that they have different kinds of buy in and they have reputational stakes? I mean, what, what do you think the structures are that well, they treat.
- [66:37]
C
Their workers right just like as a baseline and to the extent that the workers care about their users, the fact that the workers care becomes dispositive in a way that it's not if the worker is just trying to convince a boss. And it doesn't mean that they're automatically great, it just means that they. Workplace democracy is just like any other kind of democracy. You can elect someone really bad, you know, but, you know, the problem with democracy or the problem with autocracy is not that the autocrat can never be right. It's the problem is that when they're wrong, you can't do anything about it. And you know, it's, it's actually interesting. I always bring up Jimmy Wales in this context because it's interesting to see it playing out, especially now because he's in trouble because he accused Wikipedians of being mean to Israel about Gaza and said our entry is not encyclopedic and we need to revisit it because it calls what's going on in Gaza a genocide. But one of the interesting things about Jimmy Wales is that uniquely among the people who started these benevolent dictator for life projects that are supposed to be about a public interest Internet, Jimmy is the one who said, okay, but this is too important for one person to run. It needs to be run by its workers. Right? We need to set up a council of the people who write and maintain Wikipedia. And I will not be first among equals. I will just be another Wikipedian, albeit one whose voice carries a lot of weight, but not always right? Not when he wanted to put AI in Wikipedia and possibly not now in this discussion about Gaza. And what's most interesting about that for me, because I like Jimmy a lot. He's an old pal. I've known him for as long as I've known Dana. But one of the interesting things for me about Jimmy is that unlike all the other founders, these benevolent dictator for life projects, who all call themselves leftists like I do, Jimmy is a fire breathing libertarian whose favorite author is Ayn Rand. And he was like, yeah, no, one person can't be in charge of this. That would be crazy. No one is smart enough to be in charge of this. It needs to be a worker run, cooperative, said the Ayn Rand reader. It is very strange.
- [68:53]
A
Well, now we have Grokopedia as the counterpart because that's just great.
- [68:57]
C
And here we go, right? It's an example of like, so I guess, you know, I, I am with the neoliberals to the extent that I think that discipline is a really important factor in the performance of a firm. So what happens when you misbehave is like a Very, very important factor. And this is one of, this is why neoliberals like competition in markets. I think that just like broadly, we all function that way, right? When you get too big for your britches, you. You go crazy and do bad things. You crawl up your own ass and die, right? Like Heinlein got too big for his editor to edit and his books got long and sloppy. Howard Hughes got too big for anyone to tell him to just stick to building good airplanes and he started wearing Kleenex boxes on his feet and saving his urine in jars, right? Like, it's just what happens when, when nothing bad happens to you when you screw up is you start to screw up and you never stop.
- [69:47]
B
Corey, I was wondering, you know, I've been, as I've been thinking about inside Ification, you know, since you introduced the term a while back, I've been thinking about how you think about the lip. Like what it's supposed to describe, the limits of it, right? And it's clearly about digital platforms. Is it about other things? I mean, that's one question. Why don't we, why don't we cover that?
- [70:10]
C
So I think formally, right, like, insertification speaks to the unique situation for tech, right? So on the one hand, you have like, the universality of digital computers and the fact that platforms that were contemplating inshidification also had to contend with the possibility that interoperable products would disenchantify things and cost them more than they stood to gain. But also that digital platforms, when they don't have to worry about that anymore because of IP law that bans reverse engineering, they have infinite flexibility to just change the rules from moment to moment, second to second, second, and just like screw you over in lots of ways also this uniquely constituted workforce. But you know, my first languages were English and Yiddish. And the thing about both of those languages is that they have no language academy and words just mean what people say they mean. Begging the question can mean two identical and opposite things at the same time, right? And I am totally fine with people using inshidification loosely and colloquially to just.
- [71:15]
B
Means, oh yeah, that's happening big time, right?
- [71:18]
C
Please do. You know, And I'm fine with it happening without people crediting me. I don't think that, like, I think that coining a word is the least important part of it. I think that other people using the word is the important part. And like, you know, as I wrote in an essay about this, the only way to ensure that the word maintains a precise technical meaning is to confine its usage to a Group of irrelevant insiders. And the wider it spreads, the further it's going to get.
- [71:43]
B
You're attacking sts right now, Corey, so be careful.
- [71:48]
C
You know, the wider it spreads, the fewer people there are going to be as a proportion of people who use the word who know that it's associated with me. And I'm fine with that. I spent the last 25 years trying to get people to care about tech policy, not trying to get people to think that I made nice words.
- [72:06]
A
You know, the funny thing is, if anybody actually decides to read the book, he actually lays this out in the weirdest of places in the book. You know, only academics read the acknowledgments. And the weirdest part is it's not just in the acknowledgments where he lays this out, it is in the last paragraph of the acknowledgments. So he's at the absolute end where he sort of talks about his essay, Dirty Words are Political Potent. And he sort of opens it up with the fact that a neo. Neologism. How do you say that word, Corey?
- [72:37]
C
Neologism, thank.
- [72:38]
A
That word is sometimes decoupled from its theoretical underpinnings and is used colloquially as a feature, not a bug. And then he continues on to sort of highlight this. So I, you know, it's a weird plug, but read the acknowledgment.
- [72:50]
B
Yeah, go there. You learn a lot from those pages, actually.
- [72:55]
C
I also, in my audio books, when Wil Wheaton reads them, I make him read PI to a thousand digits at the end of the audiobook. It's a long story. It's sort of an inside joke.
- [73:04]
B
But is he reading in shitification?
- [73:05]
C
No, no, I read it.
- [73:07]
B
Okay. Okay.
- [73:09]
C
I did not myself read PI to a thousand digits.
- [73:12]
B
That's good. I feel I would be really upset if having you two on and not talking a bit about the AI bubble and. And so I thought, you know, one road into this, Corey, is to ask if the insidification theory has predictive value and if we would kind of expect that generative AI would go through an inshidification process eventually. Is it that kind of thing at this point where you think these things are so big that this is just what happens?
- [73:47]
C
Yes. And so on the one hand, AI is like very well poised for injurification in as much as it's very opaque. It makes lots of mistakes in its natural state. And so you can never tell whether it made a mistake that was beneficial to the firm that produced it because it is an error prone chatbot or because it has been programmed to, for example, when it's doing reasoning to add like 11 extra steps to burn some more tokens because every time it does that it costs you money. Or if you're like a programmer using a copilot style thing, if it periodically one shots things to keep you using it by doing something more computationally expensive and then drops down into a less computationally expensive thing that won't one shot so that you just keep putting money in the roulette machine or in the slot machine and pulling the handle. But I also think that a lot of the stuff that we worry about AI doing is kind of beside the point because the bubble's going to burst and the number of foundation models we have might drop to zero. And so there'll still be a lot of open source AI that is barely optimized because these companies that have already done $600 billion in capex or $700 billion and are committing to another trillion and they're only making $60 billion a year in aggregate across the entire industry worldwide. And that, that number is like super cooked. It's like that includes the $10 billion in compute credits that OpenAI gets from Microsoft and gives back to Microsoft. That's not revenue, guys. That is just, that's just a liability that like how did that end up as an asset on your balance sheet? You gave away $10 billion with the compute. That didn't make you $10 billion.
- [75:38]
A
There's a precedent here in 1929, that's exactly how the bank banks.
- [75:41]
C
Yeah, yeah, exactly. A lot of shitty paper. And yeah, again, they're like passing around the same hundred billion dollar IOU and booking it as revenue. They're depreciating these assets over five years when they only last for two years. Like it's, it's, there's just not, it's just not. And also all the jobs that supposedly can do it can't do. Which doesn't mean your boss won't fire you and replace you with an AI that can't do your job because your boss is relentlessly horny never to pay you again. But like, you know, it does, like it won't work, right? And so like eventually, like Stein's Law, one of the best things to come out of finance, Stein's Law, anything that can't go on forever eventually stops. And you know people are going to stop.
- [76:19]
B
Yeah, well, I have a buddy who's a, I have a buddy who's a technical documentation guy at a software firm and he uses generative AI all the time, all day long, because it's good at doing the very highly standardized boilerplate kinds of documents that he works in. Right. He's paying Gemini, like, $20 a month right now, you know, to do his work. And he's, you know, counting tokens and stuff. He thinks on, like, the heavy days, he's costing them about $40 a day.
- [76:51]
C
Right.
- [76:51]
B
You know, and it's just like, how is. Tell me how this is a fucking business model, folks. Like, how do you.
- [76:56]
A
I mean, we have a business model, right? Like, Uber did the same thing. It's just a question of when it catches up and what it looks like. Looks like. And I think one of the hardest things about AI right now is that you can enjoy the actual technical products of AI and be utterly terrified about what this whole thing is setting up and setting in motion.
- [77:17]
B
Yeah.
- [77:18]
A
And I think this is one of the things that I think there's going to be this duality. And we unfortunately said these put these against each other in a whole set of tech clash. And this goes back to Corey's point of, like, oh, as individuals, we can fight it. And it's just like. And for me, that's, you know, he talks about it as a bubble. Like, for me, what terrifies me about it is how it is rearranging financial structures in a way that's creating ultimate precarity, not just precarity for workers, although that is a part of it. Right. As Corey's pointed out, like, you use. You use AI as a justification to, you know, lay people off, but you were just going to lay people off anyhow. It's just a nice justification. You also use it to, like, squeeze people and tell people like, you better work harder and do more, more, you know, or I'm going to let you off. It becomes a threat as well. But it also becomes, you know, to your. To your friend's point, like, the actual costs of this, we are not paying. And you're just talking about the revenue costs, you know, for the models. What do we want to think about? The environmental costs? What do we want to think about? The theft of, you know, data costs? What do we want to think about? All of the destabilizing costs of this that we are forcing into our children to be paying for this. So this is not just a matter of VC is going to, you know, solve it four years from now. And I think this is where I share Corey's view. Like, this is a terrifying bubble, and I think it's going to, you know, I actually There'll be plenty of foundation models at the other end. I'm worried about all the other things we're going to lose when this whole thing comes crashing down. And I don't think it's going to be the tech we lose. I think it's going to be lives. I think it's going to be, you know, it's going to be, you know, livelihoods. It's going to be human dignity. It's going to be all of these other things. And that to me, the more. And I, you know, especially coming from academia these days, where I'm like, we're enrolled in that project too. Like, every university is like, do more with AI. Think about AI. It's like, how do we collectively start to contribute to this inevitability rhetoric where that inevitability is somehow positive? And like, what does it mean to hold on to? Like, yeah, the tech is cool and interesting, but how much are we contributing to the, you know, the bottom line that justifies the regulation, that justifies all of the other things that are set up. And that's where, you know, I think the thing that's interesting to me about, you know, what is going to be in shitified is it's not going to be the tech. I think. I think it's going to be the experience of work that is going to be initiatified to this process.
- [79:52]
C
So the reason zero foundation models is important is we're going to fire people who do important things, like figure out whose kids get taken away by Child Protective Services, and we're going to replace them with AIs that do that job badly, and then they're going to switch the AI off and then that job won't get done at all. Right? That's why zero foundations models matter. This is why I call this the asbestos we're shoveling into the wall. Walls, right? Like it's, it's. We are going to end up with the only thing that could be worse than having these important jobs done badly by AI, which is to have no way to do the jobs because all the workers who are skilled to do them will be gone. The process knowledge will have been vaporized forever. You know, those people will have either retrained or retired or whatever. Like, it's just going to be well.
- [80:38]
A
But I think that we're already experiencing that without AI, which is like, look at air traffic control.
- [80:43]
C
Oh, yeah, 100%. 100%.
- [80:45]
A
And that doesn't even require AI.
- [80:46]
C
And I was going to give you a reference. So I don't know if you've ever read Hubert Horan. He's an aviation analyst who took one look at Uber, like, 15 years ago, was like, oh, my God, this is a scam, and has spent, like, 15 years just picking over their balance sheets and their disclosures and writing the most trenchant, detailed criticisms of their finances. So Hubert has just sent me a draft of an essay where he's trying to synthesize his work with Ed Zitron, who has been doing the same for AI, and he's showing how Uber created the playbook for AI. And it's. Actually, I should send this to him because I just thought of it. It's basically how tobacco invented climate. Climate denial, right? It's this. It's the same thing.
- [81:30]
B
Say how? In what sense?
- [81:31]
C
Well, the Tobacco Cancer Link denial playbook is literally, like, invented by some PR firms that were paid by the tobacco.
- [81:39]
B
Oh, I see. Okay.
- [81:40]
C
It's literally the thing they repurposed for climate denial.
- [81:43]
B
Mm.
- [81:44]
C
Mm.
- [81:45]
A
But I guess part of what I'm thinking about, you know, I think. I think there's a lot in the climate denial comparison, but I think that there's also in that aviation comparison. What does it mean when we pay costs into the future for decisions in the past, like the aviation situation? Right now we're paying for the 80s Reagan, you know, layoffs, you know, because of the union busting that occurred. We're still paying for that on top of all the other pieces. And I think, to your point, unlike, we're going to just dismantle the things and lose the knowledge, right? There's a. There's a concept in STS called agnotology, right? Which is the study of ignorance. And this is where it connects to your. To your climate. Because one of the things that I always find fascinating is that the people who are obsessed with agnology lay out three different kinds of ignorance. The first kind is that which is just not yet known, right? And it can be at the macro level of like, well, yet we science, or it can be at the micro level, which is why kids get an education. Then there's a second level of ignorance, which is the idea of that which is forgotten or lost. And again, at the macro level, we can think about indigenous knowledge lost due to genocide. And at the micro level, we can think about, like, the dude who knew how the copier worked, left the building, and now we're in deep trouble, right? And then the third category is that which is polluted or, you know, undermined. And that's, of course, where we get you Know, the tobacco companies all the way down to the, you know, everyday domestic violence, gaslighting. And so I say this because one of the things about constructing ignorance is that I actually am more worried when we're talking about AI of that actual second tier, we're mostly focused on the third tier where we're like, oh my gosh, fake, you know, like fake news or like all this. I actually think about all of the things that we're going to lose, you know, in terms of shared knowledge because there aren't people doing that because we're like, oh, you know, the AIs will do it, so we're not going to invest in training of people. And like we can produce an ignorant society through the loss of those knowledge, through those degradations of our systems. And I think that, you know, that gets to your point on like what happens with cps, right? Which is where do we lose that knowledge through, you know, our choices of how to position the AI systems in these equations.
- [83:59]
C
Yeah. Oh, I just had a thought and it left me. We're talking about loss. Oh yeah.
- [84:07]
B
I don't know.
- [84:07]
C
So those other applications, deep fake porn, AI illustration, you know, these, these, these things that are quite AI election disinformation, you know, whatever else we can say about them, and they're, I think they're quite horrible. They're not materially important to the future of AI, right? They're at best demos. So if you think about AI as a material phenomenon whose success rises and falls by the amount of wages it can displace, the total volume of wages available for election disinformation is nothing. It's basically zero, right? Porn is not a well compensated occupation illustration. Vaporize the entire income of every commercial illustrator in the world and you do not pay for the kombucha for the technicians overseeing one run of Mid journey, right? Like it's just, it, it's a demo, right? It's disgusting that they're doing it. It's gross, it's awful, it's not materially important. Ban. Banning election disinformation with AI is as consequential for the ability of AI companies to raise money as banning the imminentization of the eschaton by creating an AI. God, right? This is not in anyone's serious business plan. There isn't like a line item in their revenue projections for like the second quarter of 2028 for the money they get when the fancy autocomplete becomes sentient and turns us into paperclips. And so like one of the reasons it's great to talk about that stuff if you're an AI bro is it doesn't matter if it gets banned because it doesn't change your investment situation.
- [85:48]
A
And I think that's what makes these things scary isn't like it. Also, you know, you can take away Stack Overflow, right, And obliterate all of the kinds of communities, right? And you're not. It's in this question of whether we're dealing with just intellectual property or we're dealing with communities. The most beautiful part of Stack Overflow was the gifting economy that was part of it. That made people feel like they were part of something bigger and like, you know, dialing back to, even to our conversation about higher ed. For me, it's not even about the classes in terms of learning from the teacher. It's about the relational work that actually happens on the college campuses. So the other question for me around a lot of these AI pieces is how much destruction of, you know, social fabrics within society will we tolerate and what are the consequences of those constant destructions? We are very much paying for the destruction of COVID and AI is another one. We're just like, oh, let's just pile on and keep destroying all of the ties that we have that allow us to come together as an imagined community in a meaningful sense.
- [86:48]
C
There's a great section in Dan Wang's new book Breakneck about China and America, about the distinction between process knowledge and intellectual property and how process knowledge is really what you need to make stuff happen, right? Like if I give you the instructions for making a 4 nanometer or feature scale chip, that won't help you even if you have a chip fabrication facility, unless you have a lot of incredibly specialized process knowledge. And the reason bosses love intellectual property is you can alienate it from workers and sell it. But process knowledge is, you know, when you hear people talk about non competes, there's a shibboleth you hear sometimes which is the intellectual property in my company walks at the door every day at 5pm it doesn't. The intellectual property is in the filing cabinet. The process knowledge walks out the door at 5pm the person who knows how fast to put the feedstock in the machine so it doesn't jam, and if it does jam, knows who can recalibrate it and if that person is sick, knows the phone number of the person who retired who can recalibrate it and will do it for a hundred bucks and a six pack of beer, right? Like that is stuff that you just can't recover once it's gone. You have to rebuild. Can't be reconstituted. It has to be rebuilt from the ground up. And bosses don't value it because it's not a commodity. But you can't run a shop without it.
- [88:12]
A
Lee, is there something like somehow my brain is going lasers? Like there's some text that I'm thinking about in terms of rebuilding lasers. Do you know what I'm talking about? Is it Collins?
- [88:22]
B
Yeah, yeah. Harry Collins. Yeah, yeah. Tacit knowledge stuff.
- [88:27]
C
Tacit knowledge, yeah.
- [88:29]
A
And all of these different processes where it's like throughout these different technologies, how do these things get held within. Where was expertise within the structure? And we often think expertise is in the books, but it's not.
- [88:40]
C
And I think the distinction, tacit knowledge and process knowledge is that tacit knowledge is often things about how to do your job that you know. But process knowledge is information about how other people do their job. Like who you call like, you know.
- [88:55]
B
I'm about to ship you that like extended cognition stuff.
- [88:57]
C
Yeah. I'm about to ship you some RAM that is failing our qa. So you're going to have to underclock it. But I know that you have a line of lower powered devices. And so even though the serial number on this says that you should put it into your higher, you know, your higher spec devices, actually run it at low clock speeds and your low ones and I will invoice you for it as though it were lower speed chips is like a thing that you can do if you have intra firm process knowledge. Right. And it requires like trust and relations, but also knowledge and. Yeah, yeah.
- [89:33]
B
So we usually wrap up by talking about what's up next. Dana, you just turned in a book manuscript like you always have more than one thing going on. So what's next for you?
- [89:45]
A
I think the top of mind for me is that Jennifer Teshdi and I are sort of revisiting an old project that we had so years ago we sort of joked around that what happens when two computer scientists, two sociologists and a lawyer walk in the room. And that was, you know, Andrew Selbst, Suresh Venkan Submarine, Sorel Friedler, Jennifer Teschi and I. And we started to talk about how, you know, technologies get into these abstraction traps, right. Where you design to try to create fairness. But like, you keep getting these traps that mess, mess, mess you up. But Janet and I kept coming back to this old piece and trying to think through how we sort of comparatively think of it in terms of decoys. Right. What are the decoys in this AI conversation that are being set up where it's just like, look over there while we do this over here. And how much is that its own form of trapping, but a sort of a higher level of peace. So, you know, stay tuned. This is a project that the two of us are sort of going to do a deep dive on in a few weeks and I can't wait.
- [90:46]
C
Cool.
- [90:47]
B
I look forward to talking to you more about that for sure. And I have thoughts already, but I'll hold them. And Corey, you just turned in a second draft of a AI Book, right?
- [90:58]
C
Yeah, it's called the Reverse Centaur's Guide to Life after AI. It'll be out from Ferrar, Strauss and Giroux in June. They're the ones who published in Shidification. I am out for them on a tour as we speak. I'm at home for like two days to do laundry and change suitcases. But I leave for Miami tomorrow. I'll be at Books and Books and Coral Gables. I'm going to be back in Burbank for the Burbank Book Festival. I'll be in Lisbon for Web summit. I'll be in Cardiff for the Hay Summit. I'll be in Oxford with Tim Wu. I'll be in London for three consecutive nights doing a virtual event with Vass Bednar for the Vancouver Public Library. I'll be in San Diego, Seattle, Madison, Connecticut, basically, you know, Wesleyan. And I'll be in Denver. So that's all the stuff on the rest of this tour.
- [91:44]
B
Cool, man. And do you know what's. I mean, you're always doing stuff. So what's next after the AI book?
- [91:50]
C
What's next? Well, so I have these two graphic novels in the shoot. One is a young adult reader or young like not young adult, middle grades adaptation of my novella Unauthorized Bread. And that'll be next year. And then the year after that there'll be the graphic novel adaptation of Inshittification by Corin Chadney.
- [92:14]
B
Oh, is that for kids too? Or is that.
- [92:16]
C
It's more for adults. I tried unsuccessfully to pitch them on a young reader's edition. I think it'd be good focusing on, you know, super like of the moment platforms and trying to rebuild the argument. But they just, they thought that it would cannibalize the graphic noise novel. So we're not going to do that. And then I hope there'll be a graphic novel. They're considering a graphic novel of the AI book as well. And I have a great graphic novel publisher first, second at Macmillan and just they do really good non fiction and fiction. They did my fiction as well. And then I have to write one more little brother story and then I'm ready to do a little brother short story collection and so that's in my shoot as well. And then I have like three more Martin Hench novels on the drawing board. I don't know when I'm going to write them, but eventually my wife just took a job overseas and so she's going to be relocating for much of the next year and I'm spending much of the next year in London with her and I haven't figured that out yet, but I'm sort of in the middle of an overseas move which is going to cut into some of this.
- [93:15]
B
Right on. Well y', all, I knew this would be fun to get you two together, old friends getting you back in the getting you back together for a little chat and it was a lot of fun. So thank you so much, both of you.
- [93:27]
C
Thank you. Thank you.
- [93:29]
B
Thanks for those of you who joined us out there in Etherland in STS land.
- [93:37]
A
Thank you both so much.
- [93:39]
C
Yeah, thank you.
- [93:48]
B
I hope you enjoyed this episode of our podcast. You can reach us with questions, comments and suggestions@leevinselmail.com or by following me on Twitter tsnews or on YouTube peoplesthings. Our podcast is distributed by the New Books Network, the leading platform for academic podcasts. So that you can find us wherever you get your podcasts. Peoples and things, like most things in this world, depends on the work of many people. I want to thank my brother Jake Vincl for writing the music for the show. I want to thank my buddy Juliana Castro for designing the logos for the podcast. You can check out her work at julianacastro. Co. Joe Fort is the producer for the podcast and Mandy Lam is the production assistant. This podcast and other Peoples and Things programming are produced in affiliation with Virginia Tech Publishing and supported by the center for Humanities and the University Libraries at Virginia Tech. For information about other podcasts from Virginia Tech Publishing visitors, visit Publishing Vt Edu. For the entire Peoples and Things team, I am Lee Vincl and most importantly, I want to thank you for listening. Thanks.
- [95:33]
A
Sam.