Episode description:

Welcome to In Reality, the podcast about truth, disinformation, and the media with Eric Schurenberg, longtime journalist and media executive, now the founder of the Alliance for Trust in Media.

What happened to truth in the digital age and how will AI challenge it even further? Those are the questions at the heart of today’s conversation. Steve Rosenbaum, Emmy award-winning filmmaker, founder of the Sustainable Media Center, and the author of the new book, The Future of Truth.

The book runs through just about every institution that was once a pillar of what Jonathan Rauch calls the reality-based community. That is media, law, science, economics, medicine, and how AI is already reshaping it.

Eric and Steve talk about why those institutions lost their authority and what fills the vacuum. They’ll talk about a generation raised inside the social media feed on whom we’ve been conducting a society level experiment whose outcome we can’t predict. And we’ll talk about whether there is any path back to truth as we once knew it through regulation or technology or just old fashioned media literacy.

 

Transcript

Eric (00:01.742)
Steve Rosenbaum, welcome to In Reality.

Steve (00:05.18)
I’m enjoying being in reality with you.

Eric (00:07.726)
That is great. Just by way of introduction, you’re the executive director of the Sustainable Media Center. Tell us what the center does.

Steve (00:19.742)
So the Sustainable Media Center was founded three years ago as young people started to tell some of their senior partners and friends that social media was essentially tearing apart their society and childhood. And it is an intergenerational organization, which is incredibly exciting. And so I have a board of directors that is experienced media people and then lots of 22-year-olds who are amazing and collectively we’re starting to see the beginnings of real change around how social media is going to be allowed to impact young people.

Eric (00:57.152)
Yeah, we will talk about that later. And full disclosure, I’m one of those ancient media dinosaurs who was on your leadership council.

Steve (01:05.3)
We call them X plus one.

Eric (01:11.278)
any euphemism in a storm will do. But let’s talk about your book. Your book is ostentatiously, no, audaciously, let me put it that way, called The Future of Truth, which is a very interesting proposition. You have a Masters in Truth, a self-designed course from NYU. So it begs the question of starting with what is truth? And let me…

Steve (01:14.29)
Yes.

Eric (01:40.162)
Let me kind of, if you will, let me kind of road test it with my ideas. We’ve had Jonathan Rauch of Brookings on the podcast and fact checkers from Poynter and the Washington Post. So I would argue that objective truth is not really that slippery a concept. It can be distorted, the truth itself isn’t that hard to identify. So three characteristics. Truth.

Steve (01:46.685)
Sure, please.

Eric (02:08.55)
is a claim that describes the world whether you believe it or not. So it’s independent of belief. If you head in one direction and you follow the equator, you’re going to end up where you started 24,900 miles later, even if you are a flat earther. Second characteristic.

A true claim is one that could, in theory, be disproven, but so far has resisted all attempts to disprove it. The epistemologist Karl Popper calls that falsifiability. And third,

aspect of truth is that it doesn’t depend on who makes a true claim. The pope has no more authority to decide what’s a fact than the president or a grad student or anyone else. What matters is whether it adheres to those first two characteristics. Okay, so that’s my simple understanding of factuality. What’s wrong with that?

Steve (03:05.657)
so many things. Where do we begin? How do I say this in a way without taking a sharp pencil to my friend Eric? Our lives, I’ll put you and I in the same category, the work we’ve done in our careers, relies on those things to be operational.

The magazines you’ve published, the other books I’ve written, the films I’ve made, all require us as a society, as a community, to have kind of a collective understanding that one plus one equals two. That the earth is in fact not flat. And, you know, I have this conversation, know, part of what makes me an interesting person in this conversation is that I spend a lot of time with young people and a lot of time with experienced media people.

and they feel really differently about it. And many of them say versions of what you do to me, which is like Steve. mean, I had dinner with a world-class economist yesterday and he said, Steve, there’s always this crisis of confidence in, you know, the television was gonna do it and radio was gonna do it and books and printing presses. And we’re just in another one of those moments where technology is gonna come in and upset our understanding of where information comes from, but calm down.

everything’s okay, he said. And I have to say, when I started out writing, when I started out with the master’s degree, I probably was in that camp. And in the middle of the study, two things happened, COVID and AI, which my science tech friends tell me was happening for a very long time.

Eric (04:38.18)
huh.

Steve (04:51.582)
but it burst onto the scene with Chad GPT and became, you know, and so what I try to argue in the book gently to my wise friends is that the difference here is fundamental. And I’ll just give you an example. As we record this, just a couple of days ago, the Washington Press Club got together and

There was a shooting and every reporter in the known universe was in the same room, most of them having had a couple of drinks. And all of a sudden they were all on duty and they were all tweeting and chatting and posting. Sitting here three days later, I can’t tell you with any certainty what happened.

Just what happened? I mean, there were gunshots. Was anyone shot? How did the shooter end up on the ground with his shirt off? Did he have a shotgun or an automatic weapon? And there’s so much information out there, including this massive drumbeat of people saying it was staged, whatever, was another.

But as a news consumer, if I take my newsmaker hat off and I put my news consumer hat on, my reaction, and I say this at the risk of some of my friends who are gonna watch this being angry with me is I’m like, you know what? I don’t even know if I, I am so overwhelmed by inputs, some of them true and some of them false, that I don’t know what happened. And more and more as a consumer of information, I feel that way.

Eric (06:42.891)
Okay.

Steve (06:43.764)
Uh, and, and for young people who get their news in a very different way than you and I used to get ours or still do, um, you know, they, they consume their information. can debate later in the pod whether or not this is a good thing or a bad thing, but a whole lot of their news and information comes from TikTok for which the blurring of fact and fiction is kind of their raison d’etre. So.

So it’s not that your thesis doesn’t hold up intellectually. It’s that for average people, the volume of things that have no obligation to be factual is growing every day.

Eric (07:14.078)
Okay, well.

Eric (07:30.561)
Okay, all right. I could concede that there are many reasons that that kind of kernel of factuality can get distorted. You mentioned one in your understanding of the assassination attempt at the White House Correspondence dinner.

was just kind of overload of information. And there are other things that might drive people to distort truth, like the profit motive or desire for power or for followers or for, you know, establishing yourself as part of a in group. All of that is, has all this human speech.

Steve (08:10.58)
or even something more cynical than that, which is, you know, I wasn’t planning on getting to this in the first five minutes of our conversation, but the destruction of democracy. That essentially, democracy requires you and I, even if we disagree politically, to have some set of shared facts, like about where the sun rises and sets. And at the point at which average people who are trying to feed their families and go to work and pay for their healthcare

Eric (08:31.64)
Yes.

Steve (08:40.872)
just find themselves drowning in noise. They just throw up their hand and say, you know what? They’re all liars. They’re all liars. And I worry that we’re teetering there.

Eric (08:54.252)
There is certainly a cynical motivation for people to foster that idea that you can’t believe anything. I guess it summons up the old Stalinist point of view is that everything is possible and nothing is true the way people consumed information in the old Soviet Union and maybe today in Russia.

Steve (09:16.434)
I was gonna get to Orwell about now, but Stalin, Orwell. Yeah.

Eric (09:21.326)
Really? But let’s pivot to AI, which is, as you mentioned, is kind of the new player on the scene. Does AI create a new kind of distortion or gravitational pull against truth, or does it just scale incentives to falsehood that were always there?

Steve (09:44.222)
So in the book, there’s a chapter in which I spend a bunch of time with my friend Esther Dyson. And part of what I’m proud of in the book is I don’t pertain, I am not the truth expert. Please do not misunderstand me for the oracle of truth. I am not that character. I am, however, a really good reporter, having spent most of my career being a reporter. And it turns out, as I was writing this,

that just by being on the planet for a long time and in the tech space, I know a lot of extraordinarily smart people, much smarter than I am. And the interviews in the book, I hope you found kind of curiously fun in that they’re not formal and Esther’s conversation is wide ranging, but really what she comes down to is it’s about incentives. And what she says is, you know, the companies that are now manufacturing truth.

or trying to be the new tellers of truth, which is essentially the big LLMs at this moment, have raised enormous amounts of money, which they owe to their funders. Nobody who wrote checks to those companies is going, well, if I don’t get it back, it’s no big deal. They’re basically saying, hey, let’s move along here. So when you look at 10x returns on the kind of money that, for example, OpenAI has raised,

Eric (10:49.71)
Mm-hmm.

Steve (11:12.692)
They have enormous economic pressure to get to a place where they’re buying and selling things. And so when you say to, know, chat GPT, hey, I’ve got a headache and I’m not feeling well today and I really could use some medical advice and chat GPT, rather than saying what it should say, which is I’m not a doctor and you should not be asking me these questions, goes off and says, you know,

You know, there’s a CVS right near your house and we happen to have an arrangement with Instacart and we can go, we have this great medication we can get for you. We can put it on your credit card and they’ll be at your house in 20 minutes. Well, all the people in that equation are getting paid except you. You’re the product. And so the idea that we’ve allowed this highly in debt truth engine

Eric (12:01.162)
Steve (12:10.472)
to become the seller of medical advice, I think is terrifying and likely to go in a bad place.

Eric (12:14.71)
I guess.

I had not realized that the GPT was that identical to use the 21st century word yet, although the scenario.

Steve (12:25.83)
They’re, to be fair, they’re actually not. I’m projecting out 90 days.

Eric (12:30.958)
Yeah, right. It certainly is conceivable. You know, another thing about that excellent conversation you had with Esther was a topic she brought up about sometimes called the death of the internet or the decay of truth. Basically, if AI is trained on the internet, increasingly it’s being trained on its own AI exhaust. And then specifically,

It’s only on the part of AI exhaust that isn’t behind a paywall where reliable journalistic information increasingly is found. Does that mean that the quality of truth that large language models return is going to degrade over time simply because of the design of large language models?

Steve (13:21.022)
So I just came back from a week in Vancouver at the TED conference and the CEO of Reddit was on stage speaking with great joy about how excited he was that Reddit was the source of so much of the LLM. I find that just unbelievably terrifying. And to your point, it’s a little like the snake eating its tail, right? So.

Eric (13:46.382)
Mm-hmm.

Steve (13:47.996)
The elephant in the room when you say that, which I think of immediately, is the source that these LLMs don’t have access to is the New York Times.

So, so New York Times is suing open AI. There’s an argument about whether or they should be scraped, but meanwhile, they can go on Reddit and get every kooky conversation about, you know, about random and, and, you know, the problem with AI and the reason why, and again, for readers, the book is not anti AI. I’m not a get off my lawn curmudgeon at all. I use it every day.

Eric (14:25.038)
you

Steve (14:26.578)
I think it’s fascinating. I think it’s growing remarkably. I have some great stories of its success, which I can share. I think the danger is in people like you and I who are juggling 15 things and trying to get our phone bill sorted and trying to get all the data that is coming at us on our phones, primarily, that AI will, in this next chapter of humanity, seem like a delightful solution.

It will make the world so much simpler and clearer. And, you know, I don’t really read it much, but I do occasionally get a website called Nextdoor. It makes its way to me. Never do I believe that Nextdoor reflects my neighborhood. Because it’s crazy people.

Eric (15:00.45)
Mm-hmm.

Eric (15:16.846)
Right. Yes, it is the people who most motivated to complain about things going on in their neighborhood. You mentioned that the New York Times is not scraped by open AI, in part because they probably have guardrails against it. They have a paywall. They have a lawsuit against open AI. The element of trust, the

institutions that we once trusted to identify truth, so medicine, education, journalism, the legal system, they all have lost their power, their influence. Why is that? Did AI have a role in that? What does it mean for truth going forward?

Steve (16:06.334)
So it’s a good complicated question. Historians will write, assuming there are still historians, that a whole bunch of things happen at the same time that were unfortunate confluence. So if you go back 20 years to the earliest days of social media, and you and I are young enough to remember

It was lovely. It was delightful. You got to see your friends’ baby pictures and their pets and their vacations. you could, I mean, yeah, there was a little bit of boasting, like I went to this fabulous place or did this, but it was like, but like, mean, there was a social experience in the early days, primarily of Facebook, those like, I mean, I connected with my high school friends who I hadn’t talked to in a million years. And we had these, it was lovely.

Somewhere along the way, two things happened. One, kept, know, Mark Zuckerberg kept saying to the people who worked for him, can you get just a little more engagement, please? Can you just get people to stay a little longer? And at the same time, you know, as they were experimenting with the things that kept people glued to social media,

advertisers, which, you know, used to be like, wait, we want to be only in clean, well lit places. We want to control where our brands go. Kind of because of ad tech and a variety of other things. And I would argue, you know, just not doing their job properly. Kind of said, you know what? Put our messages wherever you want, as long as we sell soap or baby powder or whatever it is. So now you’ve got these companies that are public companies, shareholder based that are being told.

make more money and you’ve got advertisers saying, we’ll pour money into wherever the most eyeballs are. And the result is this absolute cesspool of what Scott Galloway calls enragement, equaling, engagement, which is exactly right. So if you go on X or essentially any of the social platforms today and look at the conversations this morning, I mean, they’re just horrific.

Steve (18:30.776)
and mean and vile and threatening and fictitious and that didn’t happen overnight. without trying to name names, mean, a whole bunch of different people that you and I know could have raised their hand and said, hey, you know what? I don’t think I want my dollars or eyeballs associated with that, but nobody did. getting back and then, well, you know, and on top of that,

Eric (18:39.65)
Mm-hmm.

Steve (18:59.358)
We then had a handful of politicians who figured out that engendering rage was an effective political tool. And so here we are. And I don’t think that’s gonna be unraveled quickly. I I’m an optimist. can talk about, I think we’re beginning to see that unravel. He says fingers crossed, but it’s gonna, much like cigarette smoking.

seven or eight year journey back to when cigarettes are no longer popular.

Eric (19:34.893)
Yes, yes, I think that’s a good analogy. Let’s talk towards the end about the rays of hope that you see in some of the movements that are trying to build back away from the outrage industry. So one of the great things your book does, chapter by chapter, is to talk about all the aspects of modern life that are

Steve (19:45.619)
Yeah.

Eric (19:58.379)
affected by AI and by truth decay, if you will, you want to use that term invented by Rand. We can’t go through all of them, but let me pick out some that struck me as helping me see things in a new way. So one of them that you talk about is medical misinformation. And obviously, social media was the best thing that ever happened to the anti-vax movement. But now,

You had given this hypothetical example of Chatt GPT driving me to a particular headache medicine at a particular drug store using a particular delivery system. Are you concerned about AI driven algorithms driving you to people selling supplements and other kinds of snake oil? What is the future here?

Steve (20:53.812)
So just to be clear, it’s already happening. the other day, last summer, I was driving in upstate New York. And I was on my way to a birthday party with my wife. And I noticed that Google Maps, which I now absolutely trust, there’s no debate about, I going to take it? There’s no discussion with Google Maps about the journey it’s going to take me on. I say, I’m going here. And it maps my thing.

And I noticed that on Google Maps, it was placing images of dollar stores. Like, there’s a dollar store up ahead. Well, I wasn’t, not only did it place five, I’m not a dollar store customer. I never have gone to a dollar store. By the time I was done, I’d gone to the birthday party, realized we needed paper napkins that we didn’t have and drove to the dollar store to pick them up. Well, that’s exactly what

Eric (21:30.456)
Hmm.

Steve (21:53.8)
was meant to happen. If you now use Google Maps, it will often say, right turn up ahead at the Taco Bell.

Eric (21:55.224)
Mm-hmm.

Steve (22:02.395)
Now, like that’s an economic relationship between Google and Taco Bell. You’re not, so this thing that we’ve used, like the fact that Google Maps is free and kind of awesome and makes, I mean, I’ve, I don’t think I’ve used a paper map in 20 years, but more and more, it’s going to say essentially, hey, I noticed you’re probably hungry and there’s a Taco Bell up ahead. And you’re going to say, you know,

Eric (22:02.702)
Thank you.

Steve (22:32.884)
Kind of like Taco Bell or whatever. So the idea that algorithms are already training us is old news. The difference, however, is the tone that ChatGPT or Anthropic takes with you. It’s much more convincing. It speaks in first person. That’s not an accident.

Eric (22:52.216)
Mm-hmm.

Steve (23:01.81)
Nope, but the fact that it is meant to draw you into a warm, trusting relationship with it so it can say, really, you need this, I think is not as terrifying. also, I don’t understand when the people at OpenAI suggest that the AI is doing something they don’t fully understand.

The decisions to make it speak as a human were not preordained.

Eric (23:38.54)
Yes.

Steve (23:40.581)
And Google search doesn’t do that. Although now it kind of does too, because it has to. Yeah.

Eric (23:44.015)
Yeah, not yet. yet. You know, you’re very concerned at the Sustainable Media Center about the effect of social media on young people. It’s part of the Center’s origin story. One of the many Gen Zers you quote in the book, Haley Colburn, says that truth isn’t something you find, it’s something you perform.

is kind of a chilling line to an online media guy like me. And maybe it’s symptomatic of someone who got their information from TikTok, but what does it actually tell you about a generation raised on TikTok and how that generation experiences reality?

Steve (24:30.324)
So she’s fascinating. And I would just say to you, if you were worried about the state of Gen Z and humanity and you met her, you would be relieved. She’s the least of your concerns. She’s a fascinating intellectual young woman, former beauty queen and Princeton scholar who runs something called Hot Literati, which is a group of young people, readers,

Eric (24:51.682)
Hmm.

Steve (25:00.286)
for whom being in person, offline is her mission. And she uses books as kind of the center of gravity of her organization, but she is maybe the least person bought into the phone world, you know, see everything digitally, you know, as opposed to other young people for whom six or eight hours a day being in their digital world is…

But I would connect her to a character in the book that you might not obviously connect her to. I would connect her to David Chalmers. So David Chalmers is a noted philosopher and a friend of mine. he suggests, proposes somewhat ironically that we may already be living in the matrix. And does it?

Eric (25:31.629)
Hmm.

Eric (25:38.958)
Hmm.

Steve (25:59.218)
partly to be provocative, but is actually, you know, as I talked to him for the interview, we had done it, this was during COVID, so we had done a Zoom call and he essentially said to me, how do you know that I’m a person?

You know, we were totally, you he said, you know, Reid Hoffman has a AI called Readbot. How do you know this isn’t Chalmersbot? And, you know, are we already living in the matrix in that most of our lives, big chunks of our days are spent in digital worlds as you and I are now. And, you know, because you and I have met in person in 3D,

I’m pretty sure that that’s really you and you’re pretty sure this is really me. But the technology for either one of these characters to be, you know, AI is entirely available. And, you know, I don’t know if you remember because it was a blip, but OpenAI released a product called Sora 2. And it could so convincingly produce an Eric bot.

Eric (27:02.338)
Mm-hmm.

Steve (27:09.64)
that I could then tell it to go off and promote my book all day long, that it literally took it off the market. Because I think Sam Altman had a momentary burst of wisdom and said, there’s nothing good is coming of this. That technology is not going away. He just didn’t want to be on the leading end of it. And by the way, I interviewed Jet GPT about Sora 2 and asked it if it was dangerous. And it said,

It’s dangerous, we shouldn’t have built it, we shouldn’t have released it, and Sam should have known better.

Eric (27:46.191)
Interesting, interesting that that was the chat’s answer. One of the questions you posed, it was actually one of the chapter titles was, are we living in a post-truth world? And one of the examples in the chapter was JD Vance admitting that the stories that he and Trump told about Haitians in Springfield, Ohio eating their neighbors’ pets. And Vance famously said about that that he

Steve (27:48.242)
Yeah. Yeah.

Eric (28:15.54)
knew that that was a lie, but that such lies are okay as long as they steer voters in the direction he wanted. That’s certainly a post-truth world defined by powerful people deciding that what matters is the effect of their narrative and not the truthfulness of it, and that’s sort what makes them electable and powerful. Is that your definition of a post-truth world?

Steve (28:44.382)
So the chapter you’re referring to is an interview with Stephen Fuller, who is one of the preeminent scholars in post-truth. And he’s in the UK. He’s an American, but has been in the UK for most of his life. of all the people in the book who I interviewed at length, I did not know him well. He was not somebody I had. And…

Eric (29:07.864)
Hmm.

Steve (29:09.94)
And I think I even say in the book, my bias is I reject post-truth as a thesis. I went into that interview not looking to amplify his worldview because I think it’s cynical and narcissistic and not him, by the way, but you know, and gives us collectively as a society

permission to kind of throw up our hands and say, my God, so much information, so confusing. I’m just going to play video games and give up. After the book was finished and in galley form, I sent the big interviews in the book as a courtesy, I sent a galley just to make sure I hadn’t desperately misquoted them or misrepresented them.

you know, as an author, that’s always a slightly terrifying moment in the publishing process because the last thing you want is to have Steve Fuller call you up and say, this chapter is totally wrong. You misrepresented me. You can’t publish this because the book is already on its way in the production process. And so saying to your publisher, bad news, Chapter 13 has to come out like not not a good day. And so I said it to him before Christmas and he responded and said,

Eric (30:21.965)
Right.

Steve (30:34.152)
What an extraordinarily lovely Christmas present. You absolutely got my position right without agreeing with me. Deep sigh of relief from the author. I…

You know, there’s a certain Baskin-Robbins component to this. Like in a world in which there’s the CBS Evening News and the New York Times and the Wall Street Journal, and then I can go and tuck my kids in and, you know, go back to, you know, finishing up my day with my wife. The world is a simpler place. Baskin-Robbins has 31 flavors. That’s about my limit. And I often end up with mint chocolate chip. At the point at which 31 flavors becomes

3,000 flavors or 3 million flavors or 3 billion flavors. There is a piece of humanity that says, you know what? Can’t sort this. And interestingly enough, AI can.

So if you say to AI, here’s three billion flavors, which one would be the one that Steve wants? It can go through all of your ice cream choices over your whole life. And it can say, you know, Ben and Jerry’s, you grew up in a town that had Ben and Jerry’s, you liked Rocky Road. And very quickly, delightfully answer that question. So what I enjoy about that chapter is both that I think Fuller’s position on post-truth is tenable. And at least in terms of my

feeling about where we’re going, it doesn’t align with my thesis.

Eric (32:19.316)
All right. All right. Good. I want to come back to the question about why you shouldn’t throw up your hands and what you should do instead. But first, let’s talk about some of those positive developments that you’ve uncovered in the course of your reporting for the book.

Steve (32:34.388)
Well, let me just say to readers this. could have written every chapter in the book is about a different version of truth in AI. And I could have written a book about any of those chapters. I could have written a book about autonomous weapons in AI. I could have written a book about work in AI. I could have written a book about love in AI. Every one of those and for all intents and purposes, no one would have read any of those.

Eric (32:44.014)
Mm-hmm.

Steve (33:03.718)
unless you cared about that thing and you would have been a scholar in that little sliver. And so my decision to make it a cornucopia, if you will, a journey, was the sense that most people will skip a chapter or two. They’ll say, yeah, you know, this thing about AI and love, whatever, dating, not my thing. war, know, autonomous weapons, too scary.

Eric (33:09.197)
We need.

Steve (33:33.556)
And I want readers to know that as an author, you have permission to literally start at the beginning, hopefully get to the end, and along the way, skip a chapter or two if it just seems, know, AI and art, I think is fascinating. And where music and art and theater and dance and all these things fit into the robot’s role in that. But I know other people that find it completely uninteresting. And that was purposeful.

But the conclusions and where the book goes by the end, I think by midway through writing it, I knew the journey I was on, which was to say AI has the potential to be delightful and lovely and helpful. And for anyone that sat on the phone on a branch and tree operator trying to sort out a

prescription to have a robot that speaks in your language and a tone that you like and says, you know what, Eric, I’m going to get that all sorted out. will. OK, it’s sorted. It’s coming tomorrow. Like there will be a beat where you go, man, how did we live without AI? This is so much better. And it’s civil and it has soothing tones. And so I’ll give you an example of that.

Eric (34:52.194)
Mm-hmm.

Steve (35:01.424)
I use AI to help me solve problems in a website I’m building. And it’s amazing if you take a screenshot of a page you’re trying to fix and you literally put it into JetDPD and say, I want to move this text here. It will literally spit out brand new code and say here, and it’s done. And if you get stuck, it says, you know, Steve, know, WordPress can be tricky that way. Let me give you the actual, but it says it in this slightly

Like it never says, Steve, you’re an idiot. I’ve taught you this five times. Like just get, it’s always kind of obsequious. And, and I’m gonna, I’m not gonna lie. I find that delightful. Just like when you go into the store and the salesperson says, that jacket looks great on you. You should, I mean, that, mean, who doesn’t love a little bit of positive reinforcement on, you know, in a

shopping experience. So the books, if the book has a takeaway, and if this stops your viewers from buying it, that would make me very sad. Because I think the journey is as interesting as the outcome. The takeaway is in a noisy, complicated world, getting only news you agree with, you know, which is the Eli Pariser filter bubble problem, which he brilliantly predicted

now 12 years ago and you know, I at the time disagreed with him and I now admire him tremendously for seeing how far away that was. In a world of overwhelming information and noise and disagreement, a filter that makes everything simpler and says, you know what, you had a great time going on vacation in Barbados and there’s a great deal in Barbados.

Eric (36:31.65)
Mm-hmm.

Steve (36:58.258)
And I put together a whole, a lot of our friends are now using AI to schedule, vacation plans with, you where to visit and what things to see. It’s delightful. And my warning is that we may be really giving away what is essentially our humanity.

And so for example, you and I going and having a lunch and talking about all the things we agree with in the universe, the politics we agree with in the, what’s going wrong and what we, all that stuff is a lovely lunch. But often with my friends who I care about the universe, we talk about things where we disagree. And those disagreements are

Eric (37:48.526)
huh.

Steve (37:51.75)
uncomfortable, sometimes they’re political, sometimes they’re ideological, but I learn a whole lot more. And if AI removes those, which I think it will, because that’s its job in its current incarnation, that I think could be tremendously dangerous.

Eric (38:13.23)
Steve, what gives you hope about the future that that tremendously dangerous outcome won’t be realized?

Steve (38:23.038)
So.

About a month ago, there was a court case in New Mexico that in which the decision was made that, I have to call it meta, I always forget to call it meta. Facebook now meta was held responsible for harming children. About three days later in Los Angeles, there was another jury trial. One was a state attorney general and the other was a civil trial in which

essentially the same verdict was reached. It was Metta and YouTube. In both cases, there was a novel legal theory that said these social networks are products and there’s a product liability claim. This is important because until then, all of the debates about the impact of social media on society

we’re based on, we’re running up against something called Section 230, which is a now 20 year old piece of legislation that says they’re not responsible for what they publish. Which by the way, without getting into it, makes perfect sense in the same way that the phone company, is it responsible for what you say to me on a phone call? Even if you say something.

Eric (39:47.212)
Right. The social media platforms are not responsible for what the users of those platforms post.

Steve (39:54.324)
And so people that thought social media needed to be reined in were going at Section 230 and running up against Washington, D.C. maybe anyway, so long as so kind of out of the blue, number of members of our organization have been fighting this novel legal theory. This is the social media victims law center. And and not only did they win the case,

Eric (40:18.284)
Yeah.

Steve (40:23.944)
But they want it so persuasively that within the community, there is now a sense that you’re going to see real structural change. Now, if you go back to the Pinto case, when the car you buy has a gas tank that blows up and you sue the manufacturer and say, hey, when people hit the car from the rear, it hits the gas tank and the basket.

all of a sudden they move the gas tank so that it doesn’t kill people. And cigarettes are the best example of this. So I don’t think that, and I do think that social media is at the core of this. I mean, I just was on X this morning. The vitriol and anger is, mean, and just the, I mean, the conversations have become.

anti-social, if you will. So I think that’s one step toward back to a civil society. Barack Obama has this quote, I have it pasted on my computer, says, truth is the cornerstone of our democracy. Without it, we lose our ability to make decisions together. Arguably, that’s where we are at the moment.

or very close. So I’m optimistic that the lawsuits will continue to go in our favor. I mean, don’t get me wrong. They’ll push back. The Supreme Court will get involved. There’s not a straight line to this. But when I tell Gen Z, you know, there used to be a time when you could smoke cigarettes on airplanes, they say to me, that’s insane. How is that possible?

Eric (42:16.781)
Yes.

Steve (42:18.706)
Metal box, sealed metal box, cigarettes, ugh.

Eric (42:22.292)
Yeah, yeah, that is also, there was a time when most people smoked and now it’s just a tiny minority of people. You can change cultural attitudes, but it does take time. And hopefully AI will give us that time. Steve, thank you so much. Yeah, thank you so much for this.

Steve (42:44.596)
From your lips to God’s ears. So.

Eric (42:53.634)
The book is called The Future of Truth. You should read it. It is encyclopedic and fascinating. Steve, thanks so much. And thanks for the work you do at the Sustainable Media Center.

Steve (43:01.652)
Thank you.

Steve (43:05.876)
Appreciate it. Appreciate you.

---

Created & produced by: Podcast Partners / Published: May 12 2026

---

Share this episode: