In short
Adam Becker (journalist and astrophysicist) discusses his book More Everything Forever and the themes behind his new podcast Dreaming Against the Machine: imagining a realistic, good future; pushing back against tech-billionaire narratives; and critiquing “agentic” AI/LLM workflows that replace thinking with automation. He also debates energy transition (solar vs fossil subsidies) and attacks “vibe-based” AI hype, including orbital data centers.
Guest backgrounds
Adam Becker is a journalist and astrophysicist; Ed Zitron hosts Better Offline.
Key claims
AI “agents” and LLM-assisted writing can reduce genuine thinking and intellectual ownership; “writing is thinking” and “writing is rewriting.” Billionaires avoid introspection and consequences; AI hype cycles will likely crash, with compute/data-center power repurposed for the next hype. Orbital data centers are impractical due to heat dissipation and maintenance constraints.
Notable examples
Mark Andreessen’s “no introspection” remark; solar in Vegas; Claude Cowork drafting for Wired newsletter; “agent” workflows using tools like Figma; claims about orbital data centers needing huge radiator systems; Sam Altman/yudkowsky rhetoric and the Sam Altman Molotov incident (condemned).
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOIntroducing Adam Becker and His Podcast
1:56 to 2:55
Ed Zitron welcomes Adam Becker, discussing his new podcast and its themes.
“joined by the author of more Everything Forever.”
The Search for Positive Futures
2:55 to 4:52
Adam shares his thoughts on envisioning a better future and the challenges in finding positive narratives.
“I mean, that's that's sort of the podcast is me thinking out loud about these things with different guests each week.”
Technology and Solar Energy
4:52 to 6:16
Ed and Adam discuss solar technology, its benefits, and the current energy landscape.
“Wouldn't energy also be cheaper if we were mass solar?”
Introspection and Tech Leaders
6:16 to 7:35
The conversation shifts to the lack of introspection among tech CEOs, particularly Mark Andreessen.
“But yeah, no, he said that he doesn't introspect and that introspection was invented in Vienna a hundred years ago by Sigmund Freud.”
Critique of AI and Intellectual Labor
7:35 to 11:17
Adam critiques the concept of AI handling intellectual labor and its implications for human thought.
“It's just what I don't get is what I mean.”
The Naming of Tech Companies
11:17 to 14:01
A humorous discussion on the odd names of tech companies and their implications.
“And he posted this thing the other day where he was like, yeah, I was spending hours working on like a private wiki for all of his research.”
The Problem with Tech Names
14:01 to 15:01
Discussion on the challenges and absurdities of tech company names.
“Yeah, having to walk around saying Figma.”
AI's Impact on Writing and Thinking
15:02 to 15:50
Exploration of how AI tools affect the writing process and critical thinking.
“But nevertheless, it's, we were all stuck about this before.”
Tech Journalism and AI Writing
17:34 to 19:25
Analysis of the relationship between AI and modern tech journalism.
“This Wired piece is particularly bizarre.”
Intellectual Ownership in Writing
19:26 to 21:47
Discussion on the importance of personal engagement and ownership in writing.
“not just the the process but also the sort of like intellectual ownership i i don't mean like you know, in the, in the like intellectual property sense, but like, like the, I wrote this thing.”
Show all 20 chapters
Learning Through Friction
21:48 to 24:15
Exploration of how discomfort and challenges contribute to learning and understanding.
“don't write it, then you're not thinking your way through it.”
The Misconceptions of AI in Job Automation
24:16 to 28:00
Critique of the belief that AI can fully automate complex tasks like coding.
“I mean, it's also reminding me of one of my favorite tweets of all time, which I just found and I'm going to read it to you.”
The Illusion of AI Capabilities
28:00 to 29:40
Explore the misconceptions surrounding AI's ability to write code and its limitations.
“But they couldn't find the vulnerabilities in their own software.”
Marketing Gimmicks in AI
29:40 to 31:00
Discuss the use of marketing tactics in promoting AI technologies and their implications.
“And it's also, I'm kind of honestly a little bit scared of how well it worked on people.”
Challenges of Orbital Data Centers
31:00 to 32:40
Delve into the feasibility issues of orbital data centers and their power requirements.
“And so, because like space is a vacuum vacuum is an insulator.”
Future of AI and the Tech Bubble
34:50 to 37:10
Analyze the potential outcomes of the current AI bubble and its impending crash.
Consequences of AI Industry Practices
37:10 to 42:04
Examine the societal and ethical implications of the AI industry's actions and rhetoric.
“They want everyone to be using cloud computing and not doing any computing on their own actual computers.”
AI Alarmism and Public Perception
42:04 to 45:56
Explore the tensions around AI development and public fear.
“I also don't see this alarmism with the AI psychosis.”
The Irony of AI Industry Discourse
45:56 to 48:22
Discuss the contradictions within the AI industry and its discourse.
“He has invented a new way to fuck up his company.”
Adam Becker's New Podcast Launch
48:22 to 50:30
Learn about Adam Becker's upcoming podcast and its themes.
“We've got some great sci-fi people like Kim Stanley Robinson coming on the podcast.”
Transcript
Automatic transcript. May contain errors.0:00This is an iHeart Podcast. Guaranteed human. Spring showers bring epic off-road adventures. So bring on the mud and bring on the spring at the Honda Spring Event. Honda vehicles are built for exploring. With the rugged Honda Pilot, Ridgeline, CRV, and Passport, the road might stop, but you're not done. All from the 2025 Kelley Blue Book Best Value brand. For great deals, visit your local Chicagoland and Northwest Indiana Honda dealer Today. Based on 2025 consumer choice awards from Kelly Blue Book. Visit kbb.com for more information. Run a business and not thinking about podcasting? Think again.
0:37More Americans listen to podcasts than ad supported streaming music from Spotify and Pandora. And as the number one podcaster, iHeart's twice as large as the next two combined. Learn how podcasting can help your business. Call 844-844-IHEART. 2%. That's the number of people who take the stairs when there is also an escalator available. I'm Michael Easter, and on my podcast, 2%, I break down the science of mental toughness, fitness, and building resilience in our strange modern world. Put yourself through some hardships, and you will come out on the other side a happier, more fulfilled, healthier person.
1:12Listen to 2%, that's T-W-O-percent, on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts.
1:28Hello and welcome to Better Offline, I'm your host Ed Zitron.
1:41Better Offline. As ever, go to the episode notes to sign up to the newsletter, please do the premium, it's how I make my money, buy a fuck data centers t-shirt, or join one of the assorted communities if you ever want to fall further into the larger Zitronverse. But today we're joined by the author of more Everything Forever. That's right, journalist and astrophysicist Adam Becker. Adam, how you doing? I'm doing well, Ed. How are you? Got a new podcast, don't you? Yeah, I do. It's called Dreaming Against the Machine and it just came out pretty recently. What's about? Well, it's like a more optimistic version of this podcast.
2:17Okay, no, I like this. I want to hear about that. Tell me. Yeah. So, you know, I spent a lot of time when I was writing my book, More Everything Forever, thinking about terrible people and their bad ideas about the future. And that made me want to think more about good ideas about the future and people who I actually like. And so the podcast is about, you know, trying to imagine what a realistic good future could look like. Yeah. Can you give me some examples? Oh, no, I have no idea. Oh, OK. Yeah. Still working that one out. Yeah. Still figuring that out. I mean, that's that's sort of the podcast is me thinking out loud about these things with different guests each week.
3:02So, yeah, I mean, it's certainly a future where we found a way to wrench power away from these tech billionaires. uh and yeah it's not a future that involves like you know space colonization and super intelligent ai um but uh more than that i'm still figuring that out that's that's what the podcast is for yeah i do get a good amount of people who email and are like yeah well why don't you do you talk about more positive things on the show a it's tough to find them but b also sometimes when i do it people get mad at me yeah but people like i don't think that you mr shill i don't think you mind people being mad at you ed i yeah i don't but at the same time it's like when i'm like i mentioned this uh nebula x1 uh projector thing i really like it's quite expensive it's like this insane thing in like a box as a projector it's like got the speakers you can plug in they're wireless and it's beautiful that works really well i talked about that once and i have people email me saying ed you're a shill you're being a shill you can't and it's like oh my christ do i have to hate everything because there's plenty i do don't get me wrong yeah i feel like you get plenty yeah but i feel like there are some really good things like i feel like if america was not under the gastocracy right now we would be in a kind of golden era of solar like i have solar in my house in vegas and barely paying electricity bill.
4:39It's kind of magic. It's kind of amazing. Yeah. But yeah, no, I mean, there's a lot of great technology and solar is a really good example. Like the solar technology has gotten really great. Battery technology has gotten really great. I think that these are wonderful things. And yeah, if we weren't subsidizing fossil fuels and actually thinking clearly about both what is the cheapest and most effective sources of energy and what would be best for the climate and the planet, I think it would be pretty clear that we'd be going through an even faster energy transition than we are already. Wouldn't energy also be cheaper if we were mass solar?
5:20I'm sure someone's going to hate me for saying that. I think that's true. I'd have to look at the numbers, but solar is, I believe, the cheapest source of energy at the moment. It has been for a bit. What if instead we could just have natural gas everywhere? which is very cheap, I'm hearing at the moment. Yeah, I mean. There's lots of it. Nothing's happened to the supply. Of course. Everyone's fine. Yeah, everything's fine. Nothing is broken. And we're getting really tired of winning. Yeah. Talking of thinking, though, we're talking beyond, just before we got on, one of the most insane things I think I have heard a tech CEO say, and I say this as someone who's heard Alex Karp speak before, Mark Andreessen saying he doesn't have introspection.
6:05Yes. This was kind of my favorite thing that's happened recently. Right. For values of favorite, where I mean like horrible and unhinged. But yeah, no, he said that he doesn't introspect and that introspection was invented in Vienna a hundred years ago by Sigmund Freud. And like, okay, what do you think that, I don't know, Hamlet was about? What do you think? All right. What was Hamlet about? Hamlet was about a guy who cried a lot when his life was awesome. Same with Macbeth, same deal. Yeah, exactly. Like, what does he think? Midsummer Night's Dream, drugs. But, like, what does he think? Just ruining Shakespeare, life on air.
6:54Yeah, exactly. Like all art, like any, any, any form of art pretty much ever created by anyone at any point in history. Like, why does he believe this? I think, you know, it does seem like Mark Andreessen doesn't think he doesn't, he doesn't introspect. But what's going on in that magical brain all the time? That magical brain, the egg-like brain that he has. the fucking like the dr robot neck yes yeah it's so funny as well because i'm constantly introspective to the point where i'm like tearing up my own theories and things that i've done and thought about again and again and again it's one of the enjoyable things about being human yeah exactly one's ability to reflect yeah that's kind of makes sense when you look at all the ai stuff though yeah i mean i think that for someone like andreason he's gotten so wealthy and so powerful he's not used to hearing no and introspection from that perspective is just an opportunity to tell yourself no and why would he ever do that to himself but um you know llms definitely never introspect they're not really capable of it uh yeah and you know we we uh we've got this chain of reasoning stuff that's supposedly like introspection but it's not and i think that if you don't introspect that's going to make it so much easier to be fooled into thinking that these things are thinking yeah and also i like the chain of thought reasoning because if you think that's what thinking's like you must not think so hard yeah because it's just what is thinking well every time i hear literally anything i think what would they what do they mean by that what am i it's like every single thought you have is a tucker carlson monologue what do they mean yes what is that yes what do you mean how's my day how's my day today it's the afternoon is it like is it just the daytime do i care about nighttime now does this person care about me do they care about my family what's going on here just like in this constant state of like confusion and then using way too much energy to just go, yes, fine.
9:10Yeah. It's fine. I'm all right. You know, keep yourself busy. It's just what I don't get is what I mean. I take that back. Yeah, I actually totally get this. But if you're that rich, you have so much more time to do so many more interesting things and think about them, for example. Yeah. You know, just like a book or like a song or at that price, at the amount of money that he has, He could just be like, I wouldn't mind seeing, I don't know, Public Enemy. I bet I could just call all of them and pay them all to just do a concert in my living room. I don't think Marc Andreessen is a big Public Enemy fan, but just for the sake of example.
9:50Right, yeah. And it's very, it gets me back to this thought I've had this whole time with the LLM era, where it's like, this is just the digital dunce cap era. This is when we find out all the people who just have not been thinking the entire time, who just all they've been doing is just walk around going like money, money, money, growth. Is it like Facebook? Is it like Facebook? It reminds me of Mark Zuckerberg. Give him a million dollars. Wait, he needs$10 million. Give him 20 just in case someone else gives him 10. Just this constant state of like angry anxiety. Yeah. I mean, God, what was it?
10:28I saw this LLM startup advertising somewhere online that was saying that they were going to do your intellectual labor for you and they were going to do your homework for you and stuff like that in school. And I'm like, do you know what intellectual labor is saying that you're going to do it for me? Because they were saying then you just get to reap the benefits of the intellectual labor. And I'm like, okay, saying that you're going to do it for me and then I get to reap the benefits is sort of like saying you have a robot that's going to do pull-ups badly. And if I just watch it do pull-ups badly, then I will somehow get stronger.
11:08Like, that's just not how it works. That's not how thinking works. Well, it kind of reminds me. I saw this Andre Carpath. You remember Andre Carpath? He's one of the many co-founders of OpenAI. Yeah, yeah, yeah. And he posted this thing the other day where he was like, yeah, I was spending hours working on like a private wiki for all of his research. Not his research project, but stuff he was like looking into strongly. And it's like, yeah, and then you can use an LLM to ask it questions. It's like, why? Like, just why? Yeah. Like, I would get it if he was, like, to be clear, if he was saying, I, as a scientist, I'm collecting tons of data and I want to have it in a place I can poke at it.
11:50Fine. It didn't seem, he was suggesting it was very much, like, just researching the stuff he was looking at on, like, his personal knowledge base. It's like, what is going on in your brain, mate? What are you thinking about all day? It's just like, oh, wouldn't it be great if all these thoughts I was having were in a wiki format in Obsidian? i mean oh christ also the whole point of a wiki is that anyone can edit it why does he want anyone to be able to edit something that he's using well it's not anyone it's just him yeah it's just him it's just his wiki right but that's just but that's just a document that's just like a word document that's i don't understand no i don't either and like i there was a guy underneath there who i'm just gonna be in a newsletter i'm working on today that will actually be out before this whatever yeah um and there was a guy below but he said yeah i'm doing a less professional version of this using open claw it's like to catalog what great thoughts exactly that's what what grand missives from the depths of the brain minds do you fuckwits have what going on because every time i see my new game is to anytime i see someone just being like i i should a i'm I'm running 11 agents concurrently and they're doing this and that.
13:07I try and find what it is they're working on. I've never been able to. Never. There are images of like, there's like a photo of them with like 19 terminal windows. The CC usage saying they spent two and a half thousand dollars and a$100 a month subscription. But you can never just like look at them. You can just be like, oh, wow, you're cooking. They don't know what they're working on half the time, right? They've got one of those processes is like an agent where they asked it, oh, come up with ideas, right? Yeah, come up with ideas, work on an interface series to observe in Figma, and then have another agent look at that and say, this is the one.
13:49And then at the end, something will happen. I'm sorry, can we just pause for a second? The fact that there's something called Figma, like, it's just a Ligma joke. I don't understand. It is. Yeah, how did they do this? Yeah, but Figma's kind of useful. Sure, no, absolutely. I'm not saying it's not useful. I'm just saying the name. The name is a problem. Yeah, having to walk around saying Figma. Yeah. It's kind of like having to say Hadoop. Yes. Hadoop or Superbase. Yeah. Versal's fine, but there are a lot of, even Anthropic is a stupid name. Yeah. I think we need to return to an era where we only let Taiwan name our companies.
14:32because look you got hon high precision limited now there's a company name time one semiconductor manufacturing corporation yeah that's exactly what they do exactly i know i nothing fancy don't need to you got quanta i assume that's something to do with one of their founders maybe they named it if you named it before the 80s it's fine but like no more startup names you can't do dot lee and anything figma nope call it like i don't know art corporation dot biz like i've I don't care. Just make this easier for me. But nevertheless, it's, we were all stuck about this before. And there is another theory that I'm growing with LLM as well, which is, and I'm specifically leading to the journalists using AI one.
15:15It kind of feels like all of these agentic quote unquote processes are taking up more energy than just doing them normally. Yep. Yeah. I think that's exactly right. Way more. Yeah. Cause like, you know, all of this just feels like people trying to avoid the hard work of having to think, right? Especially if you're looking at people using this instead, you know, as a way of writing. The old saw is that writing is thinking. And I also think, you know, I mean, another old saw is that writing is rewriting, right? And so when people say, oh, I'm just using it for editing. And like, that's part of the writing process.
15:50If you're using it for editing, you're using it for writing. And that's like, that's not good.
16:062%. That is the number of people who take the stairs when there is also an escalator available. I'm Michael Easter. And on my podcast, 2%, I break down the science of mental toughness, fitness, and building resilience in our strange modern world. I'll be speaking with writers, researchers, and other health and fitness experts and more to look past the impractical and way too complex pseudoscience that dominates the wellness industry. We really believe that seed oils were inherently inflammatory. We got it wrong. Many of the problems that we are freaked out about in the world are the result of stress.
16:44Put yourself through some hardships and you will come out on the other side a happier, more fulfilled, healthier person. Listen to 2%. That's T-W-O percent on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts. Run a business and not thinking about podcasting? Think again. More Americans listen to podcasts than ad-supported streaming music from Spotify and Pandora. And as the number one podcaster, iHeart's twice as large as the next two combined. So whatever your customers listen to, they'll hear your message. Plus, only iHeart can extend your message to audiences across broadcast radio.
17:24Think podcasting can help your business? Think iHeart. Streaming, radio, and podcasting. Let us show you at iHeartAdvertising.com. That's iHeartAdvertising.com.
17:37This Wired piece is particularly bizarre. Like, I don't need to think about Kevin Roos too much. I don't think he certainly doesn't, and I won't. But Alex Heath, Alex is a weird one because he's been around tech journalism for 15-something years. Doesn't seem like a terrible bloke. But he describes this process of writing his newsletter, where he claims he feels like he is cheating. And I quote Max Zeph, the reporter of Wired.
18:37Then Claude Cowork then automates the drafting process that used to take place in his head. To your point, you were completely correct. After the agent finishes its draft, Heath goes back and forth with it for up to 30 minutes suggesting revisions. It's quite an involved process and he still writes some parts of the story himself. But Heath says the workflow saves him hours every week and that he now spends 30-40 % less time writing. Yeah, mate, how much time do you spend prompting? Yeah, and also 30-40%, even if that's accurate. it 30 to 40 percent less time in order to produce something that's writing in the smeared out voice of the internet and yes you're trying to get it to write more like you but come on and and also you know i i just i don't understand why they would be willing to relinquish not just the the process but also the sort of like intellectual ownership i i don't mean like you know, in the, in the like intellectual property sense, but like, like the, I wrote this thing.
19:40Yeah. And, and, and like the, the way that you know something when you wrote it that you, you can't know even if you were there for the editing. Like I know, I mean, I, you know, I'm, I'm getting older as we all do until we die. And as I get older, my memory is, you know, not as good as it used to be. I think it's still pretty good, but you know, I have trouble remembering all sorts of things that I didn't used to have trouble remembering but I know the stuff that I've written inside and out because I was in there you know in the trenches with it writing it yeah and and so I know it you know chapter and verse why would you give up having that kind of knowledge of your own work and your own thinking well that's the thing look for me with my writing as well it's not just like the writing and the reading all that it's i will be writing something and i will think and in my case it would be like didn't the information mentioned in june 2024 the open ai's annualized revenue like i will go through an insane chain of events in my brain but it's because of my interaction with the work that i remember that stuff it isn't just and then i linked to thingy and then think it was way that that reminds me of something I was working on in 2024, 2025.
20:59And I'm not sure that, I'm not sure that these people have had that joy of an experience. Yeah. I don't even mean this condescendingly. I just mean there is a certain joy to being like, wait, I remember this and digging something out that you painstakingly read and probably didn't even use at the time. Yep. Yeah. No, that's exactly it. It sucks. And also it seems like more work. yes i agree this seems like way more work to me like i am okay i connect all of these things and i make sure they work and i've written a long text document also he says that it's got all of his old articles in it wouldn't that fill up that context window anyway yeah putting all that technical stuff aside yeah yeah yeah no i mean i just um i don't understand because like if you don't write it, then you're not thinking your way through it.
21:54And like, yeah, okay. I go into writing something and I have sort of an idea of what it's going to be, but it's as I write it that I realize what I actually think about these things. And I get to a spot in the writing where I'm stuck and I realize, oh, I'm stuck because there was something that I thought I understood or two ideas that I thought were connected. And now I'm realizing the connection is less clear or my understanding is less good than I thought it was. And so I have to sit down and work it out or I have to go check an idea that I realized I was taking for granted and maybe it wasn't true.
22:26And then this stuff is out there in the world with my name on it. And I just can't imagine having something out in the world with my name on it that I wasn't dead sure of and felt that I had thought my own way through on. Otherwise, I'd die of shame. Same. Yeah. Like, I fuck stuff up. I make typos. Most people love to email me about them. And they sure do more than they say nice things, in fact. But it's the beauty of that fuck up that makes it mine. But on top of that, you have said this, and Cal Newport recently said it as well, where it's people are scared of thinking. And I extend that to, Cal also said, concentration.
23:14I extend it to friction. I think that that moment you just described when you're like wait what does what does that mean what does that mean wait what do you well no wait i don't get that yeah that moment is deeply uncomfortable that is introspection yeah like and it sucks and when you don't like i spent a lot of last night learning about the japanese economy because me curious little critter and it does tie into things in worrying ways but and there were moments where i'm like great okay oh the plaza record what's that oh christ i have to learn another thing oh what's this oh there's some sort of dilemma oh japanese economies are And there were constant points.
23:50I'm like, what does this, why am I fucking stupid? Why don't I understand this? But that's how you get smart. That's how you learn things. And you, I don't see how you do that. Having an LLM burp stuff up for you, even if it's based on things you have said. yeah no i think that's exactly right um yeah also if it's just reading stuff you've written before is that like at some point it's going to stop being you entirely because it's just eventually it's just going to be more ai than person very weird yeah no i i completely agree And you're also, I mean, this to me is bringing me back to something that I said, I think the first time I was on your podcast, which is that in a way this is taking the sort of billionaire experience and making it like the billionaire cognitive experience and making it available for everyone because billionaires never have to do anything uncomfortable like think.
24:56Yeah. I mean, it's also reminding me of one of my favorite tweets of all time, which I just found and I'm going to read it to you. Being a billionaire must be insane. You can buy new teeth, new skin. All your chairs cost$20 ,000 and weigh 2 ,000 pounds. Your life is just a series of your own preferences. In terms of cognitive impairment, it's probably like being kicked in the head by a horse every day. yeah i think i'd be fine but yeah yeah i mean i would be good i think a billion dollars i'd be just fine i feels like a skill issue to me i oh it's definitely a skill issue although you're also yeah yeah yeah the uh of you know that uh the arrested development meme of like no it never works for anyone else but it just might work for us yeah exactly yeah people fool themselves to think this would work for them yeah i and i also think that there is a busy box mentality to it as well when you start using these things and it's very clear from the moment you start using them that they will not just do anything on their own you can't just be like do this yep like just just check this out what do you like will you do this and it spits out lots of stories about things being one-shotted but very few actual examples that matter to me or really anyone I talk to it's like oh it fixed one problem but you can't just say do this entire task yeah so once you've done that I know my reaction to that is always the fuck is this I don't I'm I don't log on to a website have it break and then go that's okay I use another website but I think that there is something about the experience that because it's so convoluted and you have to contrive all of these little doohickeys like fucking Pee Wee Herman making breakfast yet when you get to the end of it you're like wow i'm so smart i made the i made it do this but also i'm going to give it credit for the thing that i ended up doing most of the work on if i'm honest right i mean but this is also this is part of why i think the people who are most convinced that this is going to take over every job and one shot everything are so often software engineers because yeah okay this thing is changing the way that people code and does seem to be able to you know write some code for some people but you know that's a limited human task right and and even for people whose job is you know to be a software engineer the actual writing of code is is just one part of of what the job is and the thing that's i think that the trap that these people fall into a lot of the time is they believe that their job software engineering writing code is like the the ultimate human intellectual achievement and so if you have something that can do that it must be able to do anything and that's just not true and it can also it can do that kind of yeah like it's like it can write code yep but no one also can seem to explain what that means in the end because all these companies it does sound like you've got companies where you've got a bunch of people just not writing any code yep and i hate to ask now what yeah like really they're not firing anyone and the largely across the entire software stack things seem to be getting worse so yeah good i mean it's like question mark yeah i mean anthropics uh claude code you know code base dropped and they're like yeah we wrote claude code with claude code and then there were a bunch of coders in there saying yeah i can tell that you did i also love that we're meant to be scared of claude mythos this thing that can find this obvious marketing trick yes that can find any number it can find every vulnerability and everything and that's other than the fact it required human human people to actually find the vulnerabilities and coaching and yep and indeed like one of the main but one of the main exploits of free bsd was not actually exploitable you could just make it crash like and regardless Yeah.
29:23But they couldn't find the vulnerabilities in their own software. Yeah. I mean, I'm sure, like, I'm very willing to believe that it's a helpful tool for finding certain kinds of vulnerabilities. But you're completely right. This is a marketing gimmick. Come on. It is. And it's also, I'm kind of honestly a little bit scared of how well it worked on people. Yeah. like how many people just like yeah this is the scariest thing ever i will because i couldn't see it it's because we've got this this narrative that comes out of these you know ai cults like the rationalists that uh that ai is going to take over the world and be able to you know make itself super intelligent and end humanity once it's able to write its own code and all of that's just a sci-fi fantasy but you know it makes it very easy for anthropic to do these sorts of marketing gimmicks or um oh man i don't remember who it was who pointed this out but someone i saw someone online say it's not just that it's a marketing gimmick it's also that this is a way of sort of soft launching it as an enterprise only product that's not available to the general consumer and you could you can tell they're kind of toying with that idea too but it's not really obvious what the plan is yeah because the way they sell this shit is having one million different guys with an ai avatar on twitter posting that this is what changes everything every time they do anything yeah like and i just i've also been playing a fun game which is called go and try and work out what these big integrations do it's not a very okay when i say it's fun i mean it's exhausting because i went on the goldman sachs and anthropic they're automating accounting and compliance roles and uh but don't worry the firm is in the early stages stages of developing agents based on anthropics model that will collapse the amount of time these essential functions take what are those functions who the fuck knows they don't say they just don't say they they like they don't because they probably don't have an answer i want to i want to see those contracts i want to see the actual outcomes but it's i mean this is kind of a vague point to say but i feel kind of insane that we're still talking about it on this level yeah it's still we're still talking about it like we're trying to establish whether the sasquatch exists yeah i mean i just the the amount of vibe-based crap around all of this like to me i mean because of my background as an astrophysicist to me like the the number one thing that is clearly just vibes is this whole push for uh orbital data centers which is just like what they do i feel like we might discuss this before but it's like isn't isn't the amount of power they create like wouldn't they isn't it an insane amount of power for like one megawatt i haven't looked too deep into the tech yet i don't I don't remember how much they like, I don't remember how, like what the, what the numbers are for power generation, but I know that bleeding off heat is a serious problem.
32:38And so, because like space is a vacuum vacuum is an insulator. And so even though it's cold, you can't actually get rid of the excess heat very well. And so you need these giant radiator veins that are bigger than the entire international space station just for one data center. Yeah. And it's just not going to happen. and that's just one issue. You've got the radiation, you've got the fact that you can't upgrade the damn thing without doing a spacewalk. Like, what the hell are they thinking?
33:182%. That is the number of people who take the stairs when there is also an escalator available. I'm Michael Easter And on my podcast, 2%, I break down the science of mental toughness, fitness, and building resilience in our strange modern world. I'll be speaking with writers, researchers, and other health and fitness experts and more to look past the impractical and way too complex pseudoscience that dominates the wellness industry. We really believe that seed oils were inherently inflammatory. We got it wrong. Many of the problems that we are freaked out about in the world are the result of stress.
33:57Put yourself through some hardships and you will come out on the other side a happier, more fulfilled, healthier person. Listen to 2%. That's T-W-O percent on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts. Run a business and not thinking about podcasting? Think again. More Americans listen to podcasts than ad-supported streaming music from Spotify and Pandora. And as the number one podcaster, iHeart's twice as large as the next two combined. So whatever your customers listen to, they'll hear your message. Plus, only iHeart can extend your message to audiences across broadcast radio.
34:36Think podcasting can help your business? Think iHeart. Streaming, radio, and podcasting. Call 844-844-iHeart to get started. That's 844-844-iHeart.
34:49so i just looked up real quick projects like star cloud aim for roughly eight kilowatt of power in early prototypes that's not enough uh well just to give you some context uh an mvl 72 of gp200 72 gpus is about 142 kilowatts yeah so that that's probably not gonna work no that's but you know what that's also insane i read about that i read a headline about that i think in the new york times read one in cnbc it it makes me think that if i was an evil person like a like a had a top hat and like a little curly mustache of some sort like i could just get so far just being like yeah you know i i used the model the other day and it actually is fully alive it's fully alive and it told me it's coming for your job specifically and then just go on being verified and like give them the job i'll be like yeah the ai bot came up with this mate they're coming for your wife i would be on the front page of new york times terrifying message from british traveler yeah i would just claim i was from the future like that would be in the kevin roos would be immediately intrigued it's funny i was i was about to say kevin roos would believe you yeah no i think i think he would hit me with a bean can like sila in heroes little reference for those listening remember heroes from decades ago but it's just it's frustrating as well because even though you got the big blinking warning signs and you especially like you've really gone deep into how nutty these billionaires are and how distant the things they're saying are from reality it is ours are still somewhat of a niche belief which is strange still yeah it's like i know i know it's like we're the reality-based community and somehow this is not the dominant narrative or even one of the dominant voices in the narrative instead it's like okay you know there's a few of us and the crazy people are just louder and have way more money how do you see any of this going like just you don't have to come up with a perfect dance here i realized somewhat on the spot but it's like how does this realistically go from here because even putting aside the economics it doesn't seem to be getting better in a way that matters yeah i mean we're in a bubble it's gonna crash at some point right that bubble's gonna pop and at that point i'm not completely sure what's going to happen it's going to depend on you know where we are when the bubble pops what happens um or what's happening when it pops but uh you know i think that we're gonna see a few things right we're gonna see
37:40i i would like to well there's a few different things that can happen yeah sorry i'm i'm stumbling over my words here but like i think okay i think that what the tech industry or the leaders of the tech industry would like to happen at that point is they would like to use the fact that by then you know that they will have and kind of already have um you know eaten all of the new computing hardware and you know created these giant data centers and try to repurpose them for something else because at this point, you know, like it's, it's, you know, hard to get a stick of Ram, uh, if you're not using it for a giant data center.
Read the full transcript
38:24Um, and I think they love that, right. They want everyone to be using cloud computing and not doing any computing on their own actual computers. Um, so they want to, you know, take that power that they've taken and keep it. And so they try to come up with, uh, with some other thing to be the next hype cycle that will incorporate, you know, the earlier hype cycles of AI and crypto and all the other stuff that came before and say, oh, no, no, no, this is the big thing and it's going to make all of the other things work and it'll be different this time. We promise, guys, we swear the same way that they did with this versus crypto.
39:00And it's going to be the same people shilling it both in the industry and the industry watchers, you know, the same way that Kevin Roos was a shill for crypto now he's a shill for ai um in any event uh that's what quantum yeah if you see kevin if kevin roos writes about quantum yep that's how you know the ai bubble is burst yeah i think that's probably right he's running for the exits yeah no i i think that's exactly right but that's what the industry wants right you know the other or another way that it could go is when that bubble pops, you're going to see the current AI backlash, which I think that outside of the tech industry, AI is not actually very popular at all.
39:45And I think people in the tech industry really underestimate that. And if there's a bubble that pops and that leads to a crash, that's going to supercharge the backlash against AI in a way that the tech industry is just not prepared for, I think. And I would like to see that lead to some kind of reckoning. So I want to be, I'm going to lead into a sentence that you're going to almost immediately guess what I'm talking about. I want to be clear that what happened to Sam Altman, the person who threw a Molotov cocktail, the person that threw that shot bullets, that is morally reprehensible. You cannot do that.
40:25It's disgusting. I'm against it. Yeah. I will say that I'm surprised that these companies are surprised because it's like for the best part of four years, they have said, we are taking your job. We need everything. As you, a regular person, struggle to get credit or thrive or pay your student loans. Everything's more expensive. Everything sucks. but we the ai industry get whatever whatever we want and our explicit goal is to take your job and um the thing we own will control everything and all sources of income and everything will go through us you should be scared because we're scared too i feel like they should have seen this i don't think it should happen to them i want to be really clear for legal reasons but it's like why are they surprised like this is exactly the like you you've been scaring mental you've been scaring everyone but mentally unstable people will also get scared yeah and it's horrible and it it's horrible because it's like i don't want more of this to happen but i don't see how it stops Yeah.
41:35No, I mean, they are not used to their actions having consequences. Yes. And again, you know, billionaire brain, right? When you're a billionaire, your actions pretty much don't have consequences. In a way, again, what happened with Sam Altman and the Molotov cocktail that is morally reprehensible, that stuff should not happen. But in a way, the reason that was a headline was that it was a billionaire who almost faced a consequence. yes and someone who has said and it was i i've recently checked this since i think it was april 2023 sam altman has been saying i'm scared of what we're building yep dario amade has been saying for years this will destroy all white collar labor yep they have been saying we always need more compute and as i think the regular people face this as well i don't think an average person feels like they are taken super seriously by the world yeah the world does not feel hospitable to the average person but every time an ai person says literally anything it's one page news and it's taken with the the complete seriousness the government has meetings about it yep so yeah i don't know why would people possibly feel angry yeah about any of it's just what the thing that i'm saying is saying newsletter as well it's like i think these people should tone down they've said that we need to tone down the rhetoric i agree i think that they should stop talking about the models the way they do.
42:59I also don't see this alarmism with the AI psychosis. Yeah. No, I mean, it's really striking the things that they're willing to sort of incite a moral panic about and the things that they want, right? They want to get people worried about this stuff taking all white collar jobs. They want to get people worried that it's powerful enough to destroy the world and that it's the most important thing that has ever happened in the history of humanity uh and they don't want to talk about the people right now who are you know going insane and dying because of the existing products that we already have yeah and i think also here's the thing that you you will did you see the the one of the blokes one of the blokes who the one who threw the molotov cocktail apparently he had posted yodkowsky's um if if if they build it will shit myself yeah whatever it's called yeah yeah like if anyone tells it everyone dies yeah just like oh i don't know are you surprised that people who have been that you've monetized scaring people uh-huh and people are scared yeah yeah no and i mean you know for all that altman says that he disagrees with yudkowski altman also said that he thinks yudkowski He probably deserves the Nobel Peace Prize, which I just...
44:25Which is one of the more... What for? Yeah. Like, what possibly for? Strangest man. Yeah. No, these guys are just nuts. Weirdest guy we have to hear from. Yudkowsky with the Nobel Peace Prize for Strange Boy. Weirdest, most unnecessary Harry Potter fanfic. Yeah, definitely the most. New categories in the Nobel. But it's just, no, but that's the thing. It's like, to your point, yeah, it's exactly, it is something that Altman claims to not endorse, but wholeheartedly does. Oh, yeah. And what's also interesting is, I haven't heard anything from Dario Amadei about him. Nope. You'd think within a situation like this, Dario Amadei would go and be like, I'm so sorry it's happening to Sam Altman.
45:11It's terrible. It's nothing. Yeah, no, that is a good point. I hadn't thought of it. Kind of respect it. Yeah. No, I kind of respect, because like we've really not had men of industry who just fucking hate each other in a while so i think it's nice to see nice to see like some classic robber baron shit but it's and now as of the morning of recording this um open ai that an internal memo with a clear purpose of being leaked saying that anthropic is overstating their revenues it's just it's just it's really we are getting into the funny part at least yeah the cool zone as it were yes exactly that's where we're on it it's so unfortunate it really like this on one level it's very funny on the other it's just like we could be doing a for now yep no i mean these guys love talking about opportunity costs and like what about the opportunity cost of like you fucking having your money and your position in society because i think that that's just enormous for like the opportunity cost of having a sam altman or having an elon musk is just so enormous and they never fucking talk about it because of course they don't have you seen did you hear about spacex's excellent economics um they look they lost five billion dollars last year oh god no i hadn't heard that i've been busy with other like with launching you've been busy like enjoying your life rather than reading about Elon Musk.
46:41I will give Elon Musk credit. He has invented a new way to fuck up his company. Just like attaching an AI company to SpaceX, which has useful products like Starlink that people seem to like. Yep. Yeah, no, I mean, if it works, then he has to fuck it up, right? The same way he did with Twitter. Talking of billionaires, have you been looking at Gary Tan? no actually i haven't what's he been up to so he's the head of y combinator and he's currently experiencing his own ai psychosis of sort because he's become he's become this weird like he has this thing called g stack which is his claude code setup and it's just it's literally just at claude.md i believe it's like and it has people really like it because people want to get into y combinator right so they say nice things but it's been very funny because apparently his website just has like tens of thousands of lines of code on it it's like it's it kind of feels like watching a kid at a cinema arcade machine and they just kind of like pull up they not got any money in it they just put it at the wheel like ridge racer or something or like time crisis they got the gun pointing at the screen no money in there it's just very weird because i keep getting back to this thing it's like the glasses from they live we're finally seeing who the dunces truly are all of these people who are allegedly smart are actually the literal opposite yeah yeah absolutely yeah it's unfortunate well why don't we wrap it up there yeah adam tell us about this new podcast where what sort of things have you got coming down the pike what what plans have you got let's end on a high note absolutely so um the podcast again is called dreaming against the machine and uh we're launching i mean as of this recording we're launching tomorrow was our first real episode uh and we have a bunch of great guests coming up uh you're one of them ed uh and uh we've also got uh who else do we have we've got a bunch of the original AI bias people like Sophia Noble and, and Kathy O 'Neill.
49:01We've got some great sci-fi people like Kim Stanley Robinson coming on the podcast. And you know, some, some really great scientists as well. Like John DePrescott Weinstein who we're talking with about Star Trek. And, and also, you know, with some amazing haters like you or Dave Karp. So yeah, it's going to be a lot of fun. The conversations have already, we've already got a few episodes in the can and those conversations have been great. And basically it's going to be a place to try to figure out what a good future might look like and try to look for some hope and be really cringingly sincere and earnest because that's kind of my entire deal.
49:54And I think that's the only way to fight against the darkness, really. Yeah. Like it's the only real way forward for the light to win is to actually be sincere about the things we like because as much of a hater as I am, I do love the people in this community and the people that go, you're awesome. Like we can have cool conversations on this show that are not happening elsewhere. I'm glad that someone is doing something positive. Thank you. Yeah, I mean, that was the idea. It's like there's enough darkness out there. if we want to win this thing it's not enough to say what we don't like we got to talk about what we want hell yeah well Adam thank you so much for joining me we will link to all of your various escapades in the episode notes flawless there just going to keep going I am of course Ed Zitron you can find me on a monologue this week I'm not going to surprise you with Ronan Farrow interview again but yes thank you all for listening of course you'll catch us on the monologue and we'll have some fun interviews coming up, everyone.
50:51Got the boys from This Machine Kills coming as well. Anyway, catch you on the monologue, everyone, and thank you, Adam, for joining us. Thanks, Ed, for having me.
51:07Thank you for listening to Better Offline. The editor and composer of the Better Offline theme song is Matt Ossowski. You can check out more of his music and audio projects at mattosowski.com. M-A-T-T-O-S-O-W-S-K-I.com You can email me at ez at betteroffline.com or visit betteroffline.com to find more podcast links and, of course, my newsletter. I also really recommend you go to chat.wheresyoured.at to visit the Discord and go to r slash betteroffline to check out our Reddit. Thank you so much for listening. Better Offline is a production of Cool Zone Media. For more from Cool Zone Media, visit our website, coolzonemedia.com or check us out on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts.
52:122%. That's the number of people who take the stairs when there is also an escalator available. I'm Michael Easter, and on my podcast, 2%, I break down the science of mental toughness, fitness, and building resilience in our strange modern world. Put yourself through some hardships, and you will come out on the other side a happier, more fulfilled, healthier person. Listen to 2%, that's T-W-O, percent, on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts. This is an iHeart Podcast. Guaranteed human.
From the publisher
In this week’s Better Offline, Ed Zitron is joined by journalist and astrophysicist Adam Becker to talk about introspection-free venture capital, the Pee-Wee’s Playhouse effect of LLMs, and trying to find a better future to focus on.
https://bsky.app/profile/adambecker.bsky.social
https://www.youtube.com/@DreamingAgainstTheMachine
https://www.hachettebookgroup.com/titles/adam-becker/more-everything-forever/9781541619593/
Save $10 off a year of my premium newsletter: https://edzitronswheresyouredatghostio.outpost.pub/public/promo-subscription/gzqwkv54e1 - I’d be so grateful!
YOU CAN NOW BUY BETTER OFFLINE MERCH! Go to https://cottonbureau.com/people/better-offline and use code FREE99 for free shipping on orders of $99 or more. Buy our new “FUCK DATA CENTERS” shirts today!
---
LINKS: https://www.tinyurl.com/betterofflinelinks
Newsletter: https://www.wheresyoured.at/
Reddit: https://www.reddit.com/r/BetterOffline/
Discord: chat.wheresyoured.at
Ed's Socials:
https://www.instagram.com/edzitron
https://bsky.app/profile/edzitron.com
https://www.threads.net/@edzitron
Email Me: ez@betteroffline.com
See omnystudio.com/listener for privacy information.
