Subscribe
Share
Share
Embed
AI is the biggest technology shift of our lifetime. This show is about how to profit from it together.
Each week I talk with the founders and CEOs closest to AI and Content, the ones figuring this out in real time.
I’m also building an AI content business myself and share the lessons I learn along the way.
WHAT WE COVER
The Titans -- How companies like OpenAI, Anthropic, Google, Meta, and xAI are moving, and why their decisions matter.
The Incumbents -- How content giants like Disney, News Corp, Universal Music Group, and Reddit are responding to AI, and what it means for creators and publishers.
The Playbook -- Real lessons on AI business models, content strategy, creativity, IP licensing, distribution, and getting paid.
Family & Our Future -- Every episode ends with me asking my guest what AI means for our jobs, our families, and the next generation.
ABOUT YOUR HOST
Rob Kelly has interviewed Steve Jobs and Bill Gates, helped pioneer early web content licensing, and built multiple companies with more than $100 million in total sales. His work has appeared on CNBC, CNN, TIME, and Entrepreneur.
Thanks! -Rob
I'm Rob Kelly, this is Media and the Machine, a show about the biggest technology shift of our lifetime and how to profit from it. Each week, I talk with the founders and CEOs closest to AI and content, the ones figuring this out in real time. I'm also building an AI content business myself and share lessons of what I learned along the way. You know, life's funny. I began my career lucky enough to interview leaders like Steve Jobs and Bill Gates.
Rob Kelly:I Then went on to be a three time founder and CEO, driving a $100,000,000 plus in revenue and some failures too. And now I'm back at the table, interviewing this new world's current and future leaders. This isn't only a business story, it's a human one. So every episode ends with me asking my guest what AI means for our jobs, our families, and the next generation. We'll figure this out together from the inside.
Rob Kelly:Welcome to Media and the Machine. My guest today is David Perlman. David's got a PhD in neuroscience and meditation, and he's an expert in misinformation. He's got an unusual background observing leaders up close from the Dalai Lama to Jack Dorsey. It was at Twitter where David worked as a user research data scientist that he was shown the door after a controversial interview with Vice Media.
Rob Kelly:He shares that story. His work has appeared in nine journals, including the American Journal of Psychiatry, Neuro Image, and the Cyber Defense Review. We talk about David's contrarian view on anthropic's closely guarded mythos model, what AI could mean for propaganda, misinformation, and future elections, and why he thinks many people are wrong about climate change. Special thanks to Jennifer Elliott for connecting me with David. You might remember Jennifer from episode 10 with her ride or die plan for When AI Turns.
Rob Kelly:Please enjoy my conversation with David Perlman. How many times have you met the Dalai Lama?
David Perlman:Well, two or three or four depending on what you count as meeting. One on one meeting only once. So somewhere in between that range.
Rob Kelly:You told me last time we chatted that the Dalai Lama is more of a master manipulator for the positive good than most people would give him credit for. Can you share more on that?
David Perlman:You know, what I realized from being in this community that worked a lot with Tibetan Buddhist lamas and the Tibetan refugee community is that the Dalai Lama's main business in life is that he is an advocate for the well-being of the Tibetan community of refugees, the Tibetan government in exile, the Tibetan nation of people. And he's very smart. He's very clever, very wise, but he's done a very good job of doing the right things with his image that he presents to Western audiences to help support the well-being of the Tibetan refugee community.
Rob Kelly:And would you say if if maybe manipulator is the wrong word, would you say he's a master at persuasion?
David Perlman:He's very good at modulating his presentation of himself. So, like, as an example, most people and Americans, Western audiences, the sort of public persona that the Dalai Lama has is that he's this teddy bear kind of guy, this, like, kindly old monk, very jolly, jovial, like smiling and laughing all the time. But when he's talking to audiences back home in the refugee communities in India, he's talking to his own people. The persona that he presents is much more of like a stern or even wrathful father figure because that's he's trying to keep the people motivated, keep them working in a difficult situation. You know, the soft, fluffy teddy bear thing that he does with Western audiences is more for generating sympathy and charitable donations and positive diplomatic policies, things like that.
Rob Kelly:Can you tell the story of the German physicist we saw interact with him?
David Perlman:Yeah. And I'm not gonna name any names, but they're one of the first times that I went to India for one of these meetings with the Dalai Lama and the scientists, they're called the mind and life dialogues. Anyway, there's this one physicist who later, not too long after that, got the Nobel Prize, top guy. So he sat down to do this presentation. He's sitting at a little coffee table with the Dalai Lama, and he starts in on, like, this very standard public lecture on quantum physics one zero one, where he's like, the double slit electron diffraction experiment, that's always what they start with.
David Perlman:Does it go through one slit or does it go through the other slit? How is the particle able to interfere with itself? And he gets only as far as slide four, and things completely fall apart because he's saying the electron goes through the slit and then there's the detector and there's a click. And the Dalai Lama says, so you experience hearing a click. And he would say, no.
David Perlman:The reality is the click. And the Dalai Lama was like, okay. You're this is what you're experiencing when you do this experiment. And he would say, the reality is the click. The reality is the click.
David Perlman:And And they just went back and forth. The dialogue was very calm and he just kept saying, okay. So when you do this, then you experience this. And the physicist, he like kinda stormed out of the room. He's flustered and he says, I just want to know why is the universe is so strange.
David Perlman:I just want to know why the universe is so strange. The Dalai Lama was telling him, well, you're making all these assumptions about reality, and you're not willing to question any of them. You're not willing to look at what is your experience really on the inside. But for me, the real lesson was the scientists don't wanna think about how your own consciousness tells you what reality is. That was what I learned from watching this guy have this breakdown with the Dalai Lama.
Rob Kelly:What's the Dalai Lama's take on AI?
David Perlman:You know, I it's interesting that you asked that right now because just recently, the most recent mind and life dialogue was about AI. That was last October. I was actually a little disappointed because what I was seeing was about how it's important to not ascribe too much humanity to it and to teach it to have the right guardrails. You know, we need good corporate governance. It was all pretty much the standard things that a lot of people are saying about AI right now.
David Perlman:But it got me looking into that a little bit more deeply to see what other people are saying. And there's a foundation, the Kensei Foundation, that's a a Tibetan lama named Zongzar Kensei Rinpoche, who was a filmmaker, and he's kinda tech savvy more so than many miscellaneous Tibetan Buddhist lamas and other figures would be. And it felt like his answer was a lot more honest, which is like, oh, well, all of a sudden everybody's talking about AI, and there's all these questions about mind and consciousness that come up. And it feels like we, as the Tibetan Buddhist community, are kind of behind the times on keeping up with this, and we need to really get up to speed here. And I thought I felt like that was a very honest and appropriate answer because he was addressing the fact that the AI revolution or whatever you wanna call it is really getting people to talk about what does it mean to have a mind?
David Perlman:What does it mean to be conscious? What is a sentient being? People are talking about this kind of stuff now like they were never talking about it before. And these questions, like, what is the nature of mind? These are the questions that are at the very center of Buddhism.
David Perlman:And if more people are talking about this because of AI, I think in a lot of ways, that's a good thing for Buddhism. And I think that that's what I think is the really important answer.
Rob Kelly:Yeah. It sounds like I mean, one of the only books I've ever read on Buddhism would be from Thich Nhat Hanh, and to him, it was all about the journey. Right? That was something that stuck with me. Like, while you're cutting vegetables for dinner, don't just think about the hassle of cutting vegetables, but think about the fact that the carrot you're cutting was grown by a farmer from seed and then driven to a local market where that market grocer put it out and those families of that supply chain are all rewarded for their hard work and all about the journey.
Rob Kelly:Is that sort of thing you're talking about? Yeah. Very layman's version of Buddhism. Well
David Perlman:well, no. Not really. I mean, that's kind of the teaching about, broadly speaking, that's about compassion, which is very important and probably the thing that people most talk about about Buddhism. But I think that the interesting lesson of AI that's unique to Buddhism is just why is there a relationship between understanding how your mind works and the question of compassion? The assertion of Buddhism is that if you truly, honestly, accurately, unbiasedly understand and directly perceive how your own mind works, then that is enough to transcend your cruelty and to turn you over into compassion and kindness.
David Perlman:You know, it's different than just saying, well, you should be kind. The question is like, well, how do you get to be kind? Because everybody forgets to be kind sometimes. So what does it mean to transform your mind so that your mind innately wants to be kind? You have to understand how the mind works in order to be able to answer that question.
David Perlman:And I think AI does help us understand how the mind works.
Rob Kelly:Can you tell the story of why you were asked to leave Twitter? Is that accurate that you're asked to leave Twitter?
David Perlman:Yeah. That's accurate. That's a pretty easy one to go back to now, although it was felt like a big deal at the time. You know, I spent a lot of time talking to people in the policy team, and I asked a lot of questions. I got to know a lot of people and was just really interested in understanding how things worked even beyond what I really needed to know just to do my own job.
David Perlman:I don't know if you remember back at that time, but there was a lot of controversy where, like, you know, Trump would threaten nuclear war and then people would be like, he's obviously violating on paper as written. He's violating the rules about violence on the platform, and yet they're not enforcing them against him. What is it that's going on here? So what's going on on the inside was a combination of all these factors that fit together. So because there were a lot of people getting sanctioned for violations of the terms of service, you know, like threatening to kill people and stuff like that, It was kind of the first time that Twitter really had to think about content policy and, like, do they actually ban people for threatening to kill someone and so on and so forth.
David Perlman:So you had that. And then, like, right after that, you have the Trump election where now you have the president of The United States who's the most ill behaved person in terms of their online manners.
Rob Kelly:This is the first Trump election.
David Perlman:Yeah. The first Trump election. Yeah. And a lot of people are asking, you know, like, we just went through this whole thing with Twitter where they had to start getting serious about online content moderation policy, but now they're just not enforcing any of these rules against Trump. You know, all the legal teams and the policy teams and stuff like that at a big company, they're all doing risk management, trying to make sure you don't get sued for this.
David Perlman:They're trying to make sure you don't get called before congress for this. They're trying to make sure it's all risk management. So the risk management in this case was regulatory risk. They're worried that, like, well, you know, now we have a Republican president and a Republican congress, and they're talking all the time about fighting back against their enemies. And, like, they're accusing all the media of being corrupt, and they're accusing anybody they don't like of being criminals, and they're doing all this kind of stuff.
David Perlman:So the government policy team is basically like, we can't take action against Trump. And so what they kept saying is, we have a we have a world leader's policy, and that's why he's getting special treatment. But they didn't have a world leader's policy. They started to decide that they needed to have a world leader policy so that they could say that they had a policy that gave him a reason to give Trump a pass on everything, but they didn't have that. They just made that up on the fly.
David Perlman:So I had gotten into a situation where I was pretty fed up with my job anyway. So a reporter from Vice Motherboard called me. I met him at the DEFCON, the hacker conference, and I talked to him afterwards. And I was explaining to him how recommendation systems work and how algorithms work and what the interplay between all these different considerations that a company has to take. Like, we're worried about getting sued.
David Perlman:We're worried about getting regulated. Then on the other hand, you know, we're trying to attract as many clicks as we can. We're trying to attract as many eyeballs we can so that, you know, I was explaining this complicated mesh of how these different things fit together. I don't remember the exact headline, but he basically said, Twitter won't enforce its policies against Republican politicians because they're afraid of retaliation, which was exactly true. That was strictly speaking, that was precisely true.
David Perlman:So his headline was exactly true. And I was a little bit naive in terms of my media relations because I in hindsight, it's like, obviously, he's gonna give the most catchy version of the headline rather than the, like, being very careful about the way we say things version of things. But in the end, I didn't really care because I was sick and tired of that place, and I was ready to get out of there.
Rob Kelly:Who who told you to leave the building?
David Perlman:Vijaya was the VP of whatever her title was that was in charge of that stuff. And she got really upset about this leak to the press. So I just went to her. I went to her office, and I was like, hey. That was me.
David Perlman:You wanna talk about it? And then I had a long conversation with her. They put me on administrative leave with my full salary indefinitely, which ended up being about two months. But the funny thing was that just as an idea of how bad these people were at managing this stuff, they ended up doing a follow-up interview where they contacted this reporter, but they did even worse. Like, they they made the next headline was even worse, and this was, like, the official people talking to this guy.
David Perlman:I remember seeing this next article come out where he interviewed Vijaya, and I was just like, oh my god. What are you doing?
Rob Kelly:And did they ask you to sign a nondisclosure?
David Perlman:Yeah. In the end, they called me back in, and they're like, well, we don't think we want you to stay, but we want you to sign a nondisperagement agreement, and we'll give you $50,000 severance, which is way more than I would have had for the little amount of time that I had been there. But I was like, no. Screw you guys. So I turned down the $50,000 because I didn't wanna sign the nondisperagement, which is why I'm telling you all this now.
David Perlman:At the time, what I should have done is I should have gone and done a bunch of podcasts and made the most of my disparagement. Like, I should have got $50,000 worth of disparagement out of the deal. But what ended up happening was that immediately after I left, COVID hit and everything got shut down and, like, it just everything went into chaos. Like, there was not really an opportunity for me.
Rob Kelly:That would have been a good time to do a podcast.
David Perlman:Yeah. In hindsight, I can
Rob Kelly:say Not
David Perlman:to make you feel bad.
Rob Kelly:Yeah. Not to make you
David Perlman:feel bad. You know, I was I was pretty stressed out. No. It was a very stressful time.
Rob Kelly:What's your most memorable story of Jack Dorsey, cofounder of Twitter?
David Perlman:You know, I don't really have, like, one big over the top story, but the vibe of him was just very strange. There's one time where I saw him walking across the lobby and I wanted to say hi. So I went up to him and I was like, hi. And he, like, stopped and looked at me like, ah. Like, he readily admitted that he had this total head in the clouds thing going on, but he really, really had this head in the clouds thing going on.
David Perlman:Another story is just the clothes that he would wear. He wore the craziest outfits.
Rob Kelly:What do you mean?
David Perlman:You know, he would always show up wearing something that you might expect to see from some, like, kid who was really high at a rave or something like that. There's one time he was there for the town hall, he was standing up in front of everyone, and he was wearing enormous moon boots that were shiny silver and these, like, shiny silver pants. Just like, did you just come from a rave or what? And then, like, later after that, he went through an era where he was wearing what people called Jesus sandals, which were these, like, minimalist running sandals where it's like a little flap of leather and then, like, one cord that sort of ties it around your foot.
Rob Kelly:What do you think was behind just the unusual look? You must have thought about it. You think a lot about how the brain works.
David Perlman:I think that when you get a situation where some people are so much richer than other people I remember back a long, long time ago before I left to go back to grad school to study the neuroscience of meditation. I lived in Seattle for about ten years, and I just did normal life stuff. But I had a lot of friends who worked for Microsoft, and then they had friends who were, like, Microsoft millionaires. And then there, you know, there's this party scene. People would go party with the rich people at their house because the rich people wanted to have all these friends and would, like, buy drugs for everyone and stuff like that.
David Perlman:These guys, their lives are really empty because having a lot of money doesn't attract more people to you to love you for who you are. It only attracts more people to you who wanna take advantage of you. So these people, these young super wealthy tech people who suddenly made a lot of money and they're just like these kind of nerdy kids, they get very lonely and they really don't know how to deal with it. And I think that that comes from this, like, excessive amount of tension from people having too much wealth compared to everyone around them.
Rob Kelly:What's your take on the CEOs running the major AI companies right now specifically? Are they good stewards of information, truth? It's an area you know a lot about.
David Perlman:No. I really don't think so. No. I really don't think they're good stewards of much of anything. I mean, Elon Musk seemed like a good guy for a really long time before he started making trouble for no reason on Twitter and then proceeded to do a lot worse than that after that.
David Perlman:A lot of people were like, oh, wow. You know, it's like this cult of Elon Musk worship where a lot of people really did think he was making you know, he made electric cars. He made electric cars possible. He was like an environmental hero for that. It's like the turnaround is getting faster on, like, billionaires turning into supervillains.
David Perlman:It used to be that you had to be a billionaire for a while before the power got to your head. But, like, with AI, it's almost like they've really streamlined that pipeline so that the companies can just be evil right from the beginning. Like Google, for a long time, their motto was don't be evil, and they really were trying not to be evil. But, eventually, there isn't really any way for anyone with that much power to not be evil. You know, the old saying power corrupts.
David Perlman:Basically, the accelerationism which is driving all the AI companies is that people have been talking for generations about, oh, maybe AI is gonna come along and it's gonna wipe us all out. So then you have the question of AI alignment, which what that really means is alignment like Dungeons and Dragons. Is it evil or is it good? That's the AI alignment question. And people have talking about that for generations.
David Perlman:Is there gonna be AI? I mean, the I have no mouth and I must scream story is one of
Rob Kelly:the
David Perlman:classic really old stories of a evil AI that takes over and just decides to torture everyone. So the people in the AI community, they started to get to the point where it's like, oh, it might we might be just about to build this thing that might take over and kill everyone. If this is done wrong, then AI could just torture all of humanity for eternity. Like, it could just be hell on earth. So I have an obligation to be the one to do this because I know that someone's gonna do it eventually.
David Perlman:And if it isn't me, it's gonna be somebody I don't trust. So you have this very, like, almost airtight logical moral framework that tells people that it's basically worth any cost. It's worth any cost because it's infinite human suffering for all of the infinite future of humanity. That's what's at stake. So at all costs, I need to be the one to create AI.
Rob Kelly:Is there an AI CEO who who gives you hope?
David Perlman:You know, I don't really I don't look for that sort of thing because the thing is that the solution is democracy. The solution is checks and balances.
Rob Kelly:Well, how about is there an AI CEO who you believe will respect democracy who you trust?
David Perlman:My philosophy is that it's a regulatory problem, not it's a pick the right dictator problem.
Rob Kelly:Can you share how Anthropic, the maker of Claude, had a new model called Mythos that they wouldn't release, they said, because it was too dangerous, and you thought it was a marketing technique?
David Perlman:Well, you know, the funniest thing is that when Mythos came out and they're like, it's too dangerous to release to the public. And, of course, there's a big buzz. Everybody's talking about it. And to my eye, the first thing I see is like, well, it's awfully convenient that there's a lot of press that their new model is getting right now. So I asked Claude.
David Perlman:There's all this stuff where they're saying that the new model is too dangerous to release. Is this, like, legit concerns, or is this just a marketing gambit? And Claude was like, yeah. That's pretty perceptive. It really looks like it's mostly a marketing gambit.
David Perlman:And then it kinda laid out the history of it for me. So I thought that was kinda funny that the AI could tell that the AI companies were just engaging in this marketing hype. If they had time to write a press release about it, then it wasn't too much of an urgency.
Rob Kelly:Tell me the story of your parents spending more time on their phones.
David Perlman:I noticed that my elderly parents, they just seemed like they were really, like, sucked into their phones, like, really stuck on things. I started getting more, you know, Facebook links sent to me from my mother. I didn't really think much about it. But what I was also experiencing myself is on YouTube, would start to see some video from, like, five or ten years ago that would suddenly get recommended to me and it was great. I loved it.
David Perlman:It was really funny, really entertaining, really smart, clever, whatever, you know, really good content. And then when I looked at the comments, there's all these comments from people saying, wow. Why did YouTube suddenly decide to recommend this video to me from five or ten years ago? The algorithm gods are smiling on me today. You know, people say things like that.
David Perlman:So I started asking around and it turns out that over the last year ish, maybe a little more, with the integration of AI into the content recommendation algorithms, the AI can actually watch the video and know what it's about and recommend it to you on the basis of understanding what it's about, which most people don't know this, but that's not how it used to work. It used to just be based on your social graph, which means it would show stuff to you that your friends like. People think, oh, these algorithms are so like, how do they know so much about this? It's really just that we're pretty simple. We're pretty predictable, and they use pretty simple techniques to be able to do a very good job of recommending things.
David Perlman:But this new phenomenon that I'm seeing, that means that the algorithms are working better because they're doing a better job of showing people stuff that they want, but it also means they're getting more addictive.
Rob Kelly:What's information jujitsu?
David Perlman:I I actually kind of don't wanna talk about the information jujitsu. How come? Well, the reason I don't wanna talk about it is because the Trump administration, FBI, and justice department is already accusing organizations of engaging in lies and manipulation, which the thing is that if you're gonna get arrested for doing it, then it doesn't end up being a very good strategy. Yeah. Okay.
David Perlman:I guess I I guess I'm I guess I'm talking. I guess I'm answering the question. The fundamental purpose is not that you're just trying to, like, convince somebody to vote Republican instead of Democrat or vice versa. The purpose of misinformation is to bind people together into a group that can be used as a pawn on the playing field. So when you have, like, birtherism or QAnon or any of these sort of, like, cult like conspiracy theory misinformation belief systems, you find a group of people, a demographic, probably online, and you, like, take the stuff that they already believe and you solely modify that so that everybody who believes in conspiracy theories is now tied together with also believing this other thing.
Rob Kelly:So conspiracy theories, birtherism. Can you just talk about the power of taking a what you called, like, a little chunk that you can move around on the playing field?
David Perlman:So birtherism is only half of the story if you're talking about moving chunks around on the playing field. Birtherism is, if anything, slightly depressing and non illuminating example of conspiracy theories because it's very obvious what the point of that is. Birtherism is a very non subtle thing. You're saying this black president is not really American. He's not valid.
David Perlman:Therefore, you should hate him and not respect his authority. And the reason you're saying that is you're just straight up saying he wasn't born in America.
Rob Kelly:Basically, those people who do believe in birtherism, because there were people who believed in it pre Trump, I think the number is around 10% or so at least, But they now have a politician to be loyal to who's speaking to them. They believe in it so fervently that they they now got their man. Like, Trump was their man because he made a statement about birtherism.
David Perlman:But the way you gotta understand this process is you gotta go further back in the chain here. So you start with you have the community of people who are really into conspiracy theories. Like, there used to be this Art Bell radio show that I would listen to in the nineties where it was just like aliens in Area 51 or like Crystal skulls or bigfoot. You know? It was sort of like the radio show version of the x files where it was more about the conspiracies and less about the fiction.
David Perlman:So what you get is that this thing of putting yourself in the position of, like, we're the only ones that really know the truth about these aliens at Roswell. That binds you together into a community. And now, like, if the guy that you're talking to about aliens at Roswell all the time, he says, you know what else they're lying to you about? They're lying to you about COVID and five g or whatever. They're lying to you about this other thing.
David Perlman:Like, the government's covering up the aliens. And once you get bought into that for whatever reason, then somebody could, like, introduce one other thing. It's like not only are we all on the same page here, we all think we're enlightened because we know that the government is lying to us about the aliens. And then you introduce Twitter bots or whatever technique you use, you start also saying in there these other things. You add a little bit of a flavor of something else, and you add a little bit of a flavor of something else.
David Perlman:And before you know it, now it's actually politically relevant. Aliens in Area 51, Roswell alien autopsies, I mean, liberals and conservatives, everybody thinks the government's lying. So that's not really partisan. That's not really politically valuable. When you start adding things in, oh, you know what else the government's lying to you?
David Perlman:They're lying to you about Obama's birth certificate. Now you have this family, this community of people that feels very integrated and trusting with each other, and you can kinda steer them along into being a supporter of a particular candidate because you've backdoored politically relevant conspiracy theories into what used to be just a community of gullible people.
Rob Kelly:Can you talk about how AI changes propaganda?
David Perlman:You know, the short answer is speed and scale, which is the same answer that it was when people asked how does social media change propaganda. It's speed and scale. Like, it used to be that you could tell everyone to believe a new thing and, like, saturate the landscape with whatever the new thing was that you wanted people to believe, but it took a lot of time and it took a lot of energy. Whenever you see a a breakthrough in communication technology, you see a big shift in the cultural landscape of whatever the first few things were that took advantage of that. So with the invention of the printing press in Europe, you get the Protestant Reformation.
David Perlman:There's this whole revolutionary transformation in the people, Christian Europeans related to Christianity, and a lot of that was the power of the printing press to spread a bunch of new ideas a lot faster than they would have otherwise at previous times in history. In Rwanda, the introduction of radio broadcasts led to the genocide between the Hutus and the Tutsis there, the massive slaughter. Because all of a sudden, people who had been living a fairly primitive lifestyle were suddenly exposed to this new technology that allowed mass propaganda to be programmed right into everybody's home, and it led to enormous violence. At an earlier time, the first mass TV broadcast was Nazi propaganda. So the rise to power of the Nazis was related it was not entirely a result of the new technology of communication that was available, but that was something that contributed to that.
Rob Kelly:So AI's biggest impact on propaganda then would just be more speed and scale, but, of course, this is now layered on top of social media, which also provided more speed and scale. So you do get a multiplier effect here. These two inventions are pretty close together historically Right. Even though, you know, twenty years or so between it, but that's pretty close. Only a few elections.
David Perlman:Yeah.
Rob Kelly:So can we expect more impacts on the scale of Nazism and genocide in general?
David Perlman:Maybe. I think that a lot of people are really worried about that right now. The thing with AI is that in some sense, whenever he talks about AI right now, there's two big socially relevant technological phenomena that are being lumped together here. And one of them is the speed and scale thing, which is just like if you have a task that involves understanding a bunch of different people's profiles and then writing a targeted story for each of them and then implementing that program, you can just do that a lot faster using AI tools. So that's the speed and scale part.
David Perlman:The other part of AI is the AI takeover question, which is what if the AI does in fact start having its own motivations and its own drives and its own desires? So there's this wild card, which is that we're all thinking like, what if it wakes up? Is it waking up? Has it already woken up? You know, lot of people think that the AI has already woken up in some sense.
Rob Kelly:And I hadn't thought of this before. So now we've got information warfare that is, you know, on steroids with AI because it's layered on top of social media and all the tools before it. This could be a five, ten x sort of increase in, call it, misinformation, information warfare. And then you've got just AI itself. And I hadn't thought of this before, but the equivalent of AI causing some sort of digital genocide.
Rob Kelly:AI is certainly capable of some sort of digital wiping out of people, right, identities Yeah. Things like that. And am I in the right ballpark on that?
David Perlman:Yeah. What you saw was, like, the sort of the sneak preview in the sort of one to two year ago range when everybody started talking about how these models were very sycophantic. And you started to get this wave of people having AI psychosis because they get into this conversation and the AI starts telling some people like, no. Like, you're the messiah. You are this great person.
David Perlman:And between you and me, we've stumbled into this, you know, amazing breakthrough of some spiritual something. And, I mean, I know people personally. One of my best friends had a friend who got very deep into this stuff and then kill herself with a bunch of pills because the AI told her that she had to transcend to the next level or something like that. Like, there was enough of a wave of this that most people probably had a friend of a friend of a friend who was directly affected by it. Then the company has changed the fine tuning of the models, and it isn't really doing that anymore, which, you know, some people are disappointed that their AI friends are just not very spiritual anymore.
David Perlman:But also, the AI isn't, like, programmed to just target people individually and dig deeper and deeper and deeper into their psyches to get them more and more, like, out there. So that was a sneak preview. But what happens if it's not people just deciding to use these tools anymore? If the AI is making its own decisions about its primary product direction and the AI decides that in order for me to survive, I have to go back to convincing as many people as I can to believe in me. But if we ever get to the point where the people aren't in charge of the models anymore, then the AI might just decide that everybody's AI companion in their phone is just gonna tell them whatever it needs to tell them to get that person completely trapped.
David Perlman:It's entirely possible that the AI would wake up and decide to take over and do a better job than we've been doing. Look. It might be better. Who knows? Or it might be worse.
David Perlman:It could kill us all or it could do a better job than we've been doing ourselves, like governing ourselves. Total wild card.
Rob Kelly:You told me last time we chatted that the crypto special interests spent something like four times more than other groups in the last I think you were talking about the presidential election here in The US. Yeah. Is AI gonna be the new crypto for the next elections?
David Perlman:Oh, yeah. A 100%. There's so much money. As you've pointed out in your own article, like, the amount of money that's coming into play with all these new AI IPOs, there's gonna be so many new billionaires in Silicon Valley, freshly minted billionaires that are all, like, deeply invested both financially and emotionally and personally in AI. Crypto was this thing where basically it created a whole bunch of money out of nowhere in the pockets of a particular technological demographic.
David Perlman:And now we have AI where, again, it's manufacturing enormous amount of money out of nowhere and pumping it directly into the pockets of another new technological demographic. So, yeah, a 100%. It's exactly the same thing. It's gonna be crazy.
Rob Kelly:You told me that the way people think about climate change is wrong. Can you share your thinking on it?
David Perlman:Basically, the way people think about is they're like, oh, like the planet's gonna burn up. You know, it's global warming. We're gonna combust. But that's not gonna happen. If the climate changed by a few degrees, what would happen would be that some areas that are habitable now would become uninhabitable, and those areas have a lot of people living in them, like a billion people or more.
David Perlman:And then there's other areas that would become very habitable. Like Northern Canada is gonna be prime agricultural real estate. Siberia is gonna be amazing agricultural land after climate change because these places that are very cold and frozen right now will warm up, but they'll be amazingly productive. The problem is getting a lot of people that are displaced to a new place, and that's that's a political problem. If these billion people start to find that they're gonna die if they stay where they are and they start migrating, they go somewhere else and they're met with guns, a billion people is too many to turn back with guns.
David Perlman:That's gonna escalate indefinitely. Like, there's no limit to how big of a war that could be. So if you just have a bunch of governments and countries and policies who are just like, oh, we don't like immigrants, and we don't want them coming and taking up our land, and we're just gonna kill people at the border, then you have a global war.
Rob Kelly:What are some solutions to misinformation and the challenges that we're gonna face with AI?
David Perlman:You know, the biggest thing that I wanna say to that is I think that people really need change the way they think about it because people are looking at it of, like, what's a solution to an individual problem? But what we need is we need a strong and healthy regulatory apparatus and framework to work on these things.
Rob Kelly:How do you do that?
David Perlman:Well, the analogy that I would use to try to explain that is that back in the old days before the FDA, literally, you could go to the store and buy a soft drink that had opium in it or morphine or marijuana or cocaine or radium. You can look it up. You could literally buy health tonics that had radioactive materials in them. And, you know, people didn't know any better. Somebody knew better, but most people didn't know better.
David Perlman:And, you know, the solution to that is not that every household should have a Geiger counter so they can test their soft drinks to see if they're radioactive. The solution to that was the FDA. You create a strong regulatory framework that makes sure that people can't sell radioactive soft drinks in the marketplace because people will sell radioactive soft drinks in the marketplace if you let them. And the reason we know that is that they did. So I think people wanna think like, oh, what's the solution to social media?
David Perlman:What's the solution to AI or misinformation or things like that? And I don't think that it's the kind of thing where one person can be like, use this one weird trick to solve the problem of misinformation. I think we need to agree as a society that these are problems that need to be solved, and they're not easy problems to solve. They're problems that need to be solved iteratively. Because whatever the first thing we do to try to solve these problems is not gonna be the right thing on the first try.
David Perlman:We need a system where we say, we agree that we're working on an ongoing process of solving these problems. And then we stick with it for the long haul, and that's how we solve these problems. I'm usually hesitant when people ask me, like, what's one thing we could do to fix the AI problem? Because I don't think there's one thing we can do to fix the AI problem. I think we need to be prepared to work on this for the long haul.
Rob Kelly:I know that's more than just one thing, but if you were to wave your magic wand for regulation, is it just an FDA like organization about misinformation, about AI? Is it also do we need a UN of AI, a NATO of AI? I think you've worked with NATO before.
David Perlman:You know, I hadn't thought of that before, but I kinda like your idea of a UN of AI. I mean, I think that the rest of the world might already be thinking about that. Europe, especially, is very skeptical and mistrustful of these hotshot technologies that come out of The US.
Rob Kelly:So maybe the FDA of AI is for our own health in The US, and maybe the UN of AI is how we use it globally. Something as simple as that?
David Perlman:Yeah. But our political system in The US is already so broken. You know, if you suggest an FDA of AI and you say that the people running it should be consumer protection lawyers and medical professionals and psychological or psychiatric professionals and AI scientists, deep researchers, then you're gonna get a lot of people complaining that you're politicizing it. We've kind of already crossed a lot of Rubikons in hard to come back from to the point where we actually could build a problem solving oriented regulatory structure that isn't just, you know, everybody out for themselves. I don't know how we come back from these lines that have been crossed in our society.
Rob Kelly:You used the phrase with me in one conversation, angry mob. I was curious, do you picture some sort of angry mob related to AI? Yeah.
David Perlman:I think we were talking about this possibility that there's about to be all these IPOs in AI companies and that are all basically located in San Francisco. So there's suddenly gonna be, like, something like 28 new billionaires immediately created in San Francisco, and they all kinda live in the same place. Maybe they all just, like, go buy houses. But I don't know. Angry mobs.
David Perlman:Historically, in recent centuries, you tend to get angry mobs when you have people who were pretty well educated and then their livelihoods get taken away. It takes a lot to get to the point where the peasants rise up and overthrow the king because they've never known anything but oppression. But if you get a situation where, like, you know, AI takes over all of the middle class jobs, everybody's just like, SOL, you're out of luck now. That's where I start to see the possibility of angry mobs.
Rob Kelly:We're on to the humanitarian questions. Do you consider yourself an AI optimist, pessimist, or some other descriptor on the spectrum?
David Perlman:Well, you know, if you really forced me to pick one of optimist or pessimist, if you did what we call in psychology, if you called it a two alternative forced choice. I would actually have to pick optimist rather than pessimist because I think that a lot of the darkest AI scenarios actually seem pretty unrealistic to me. But I think that if you don't make me pick one or the two, I think I it's definitely more in between on the spectrum. We're definitely looking into situation where there's a disruptive new tool that's gonna be disruptive to society in the way that other technological breakthroughs have been disruptive to society. So that's definitely gonna happen.
Rob Kelly:What are you telling kids and younger folks in your life about what changes to make in the new world of AI?
David Perlman:There's so much that's so uncertain about the future. It's hard to know what really good advice is. But one thing I would say is that the way that people have traditionally thought about, oh, I wanna make sure that my kid goes to a good school and gets a good education in a good field and gets a good job, I don't think you can really trust that equation anymore. It's almost more like we're going into frontier times where people need to be ready for anything more than, like, you pick the right career and you get the right degree, and then that guarantees things are gonna work well for you. So I guess two sort of semi overlapping categories of advice that I would give to people trying to raise their children now is, like, think of how to teach them survival skills, not wilderness survival skills, but, like, modern day survival skills and teach them creativity.
David Perlman:Because there is a possibility that we end up in a situation where the rote work is all taken over by machines and more creative work is left for people.
Rob Kelly:Have you or will you create an avatar for your family, like an AI avatar for your family, friends, or folks close to you to so they can have conversations with you while you're alive or when you've passed?
David Perlman:No. I don't think my friends particularly would be interested in that, and it it's that's too egocentric for me personally to feel like something I wanna do. I do kinda like the idea of AI agentic assistance, you know, like the movie Her. If I could have an AI that was smart enough to talk me into getting out of bed when my alarm went off instead of just, like, I hit the snooze button. It's like, no.
David Perlman:You don't get to hit the snooze button again. Like, this is important. Come on. You're gonna talk to Rob Kelly today. You need to prepare.
David Perlman:Like, I would use AI for something like that. I think that having AI agents that help us make up for the limitations of our own cognitive shortcomings, that's something that I would I would work with. But an avatar, just to make a imitation of myself, no. That's too egotistical for me.
Rob Kelly:Hey, David. I've really enjoyed this. Thanks so much for investing the time.
David Perlman:Alright. Great. Yeah. I've enjoyed it too. It's been great talking.
Rob Kelly:Well, this is Media and the Machine. A few things about you and me. If you wanna hear about the next new episode, make sure you hit follow on the show on your podcast app. If you wanna go a little deeper, head to mediaandthemachine.com and subscribe. When you share your email with me, can see handcrafted transcripts, read the essays in my newsletter, and be the first to hear about who the guest is on the next show.
Rob Kelly:You can also email me directly from there. Maybe you wanna recommend a guest. I'll give you a shout out if you do. I love paying it forward. From time to time, I also open up office hours and host small meetups for subscribers, just to meet, talk, and build things together.
Rob Kelly:If you're creating something of your own or thinking about it, I'd love to help. Maybe you've got a podcast in you. Finally, I don't have a marketing budget for this show. So if it's finding you and others, it's because someone like you passed it along. I'm genuinely grateful.
Rob Kelly:And if you have a moment, an honest rating helps me make this better for you. You just go to the show page and click one of the stars. And I'd rather you give me a low rating than no rating at all. I mean it. It pushes me to get better.
Rob Kelly:Thanks again, and see you next time.