Zuck says AI is for everyone, His Rivals Ask Washington To Slow it Down

Summary

MORE OR LESS, July 31, 2026
Episode: Zuck says AI is for everyone, His Rivals Ask Washington To Slow it Down
Hosts: Dave Morin, Jessica Lessin, Brit Morin (absent for most of episode), Sam Lessin

---

Episode Overview

This lively episode digs into the recent narrative battles between top AI companies, especially Meta’s bid to brand AI as “for everyone” amid industry and political tensions. The group parses Meta’s new ad campaign, open debates about AI “light side” vs. “dark side,” China’s chipmaking progress, compute scarcity, the evolving role of X (Twitter) in tech, and workplace/industry drama. The hosts infuse conversational wit, relatable techie anecdotes, and pointed skepticism about the motivations and PR moves behind leading AI companies.

---

Key Topics & Insights

1. Meta’s “AI is for Everyone” Campaign

Context: [09:01–14:00]

• Meta (Facebook) launched a new ad campaign branding AI as a democratizing technology—“AI is for everyone.”
• Jess: “...there’s a narrative out there that it’s elitist, it’s all about profit ringing, that it’s evil. We reject that. We think it is the latest tool that will further the democratization of technology in the hands of billions of people.”
• Hosts discuss how this narrative is pushed across Meta’s products, PR, and Mark Zuckerberg’s op-eds.
• Dave applauds the optimistic angle: “I just couldn’t agree more with this perspective... I think it’s awesome that he put the muscle of not just pen, but the, you know, creating video version of that message and then it looks like they kind of deployed it out across the entire Facebook network, across every single surface.” [12:25]
• Sam compares campaign narratives to “light side vs. dark side” Star Wars analogies.
• Sam on Mark Zuckerberg: “Mark is one of the rare leaders that actually is in a position as a founder at a trillion dollar scale who already has all the people on his platform who can try to tell the light side of the story...” [13:13]

Takeaway:

Meta is firmly positioning AI as a public good and tool for individual empowerment, in contrast to ‘doomer’ or consolidation critiques. Hosts oscillate between healthy skepticism and interest in how much is real conviction versus PR strategy.

---

2. Star Wars/Lord of the Rings Analogy for AI Power Struggles

Context: [14:00–25:00]

• Extended riff comparing AI company rivalry to the “dark side” (OpenAI as the Empire, Nvidia as Palpatine/Nvidia, Anthropic as former Rebel Alliance, etc.)
• Sam: “OpenAI is like the Death Star... Nvidia is the Emperor...” [18:15]
• The analogy explores how rival companies position themselves as underdogs (“the Rebel Alliance”) but may become what they rebelled against.
• Dave challenges Meta’s supposed benevolence, arguing “Meta is still a centralized company... whether or not they will actually put the power in the hands of the people I think is yet to be seen.” [23:15]
• Discussion of how tech movements (personal computers/Apple) often start as decentralized but end up recentralized (Apple is now one of the most closed companies).

Notable Quotes:

• Sam: “It's a political movement for decentralization that became the most centralized company in the world.” [25:54]

---

3. Market/Macro Update: China’s Lithography Breakthrough

Context: [28:04–31:44]

• Jess highlights recent reporting on China achieving significant progress in chipmaking lithography (DUV), potentially breaking Western chokeholds on advanced chips.
• “This story obviously majorly moved the market for ASML... it is also significant, just more broadly to the point of the jittery nature of investors right now in terms of how this will all play out.” [29:39]
• Many tech execs publicly claim the US must “beat China at AI” while privately conceding both countries will be leaders—demonstrating opportunistic public narratives.
• Jess: “Having witnessed that, this has always been very top of mind for me. And, and so I think this, this story is really important in that regard.” [30:32]

---

4. Compute Scarcity: Will Cloud Resources Get More Expensive?

Context: [31:44–38:47]

• Hot investor debate: Is compute (GPUs, cloud resources) about to get much more expensive, or is the risk overblown?
• Dave: “Everyone is compute constrained... the entire supply chain is operating at complete capacity to fill the demand... I don’t see any signal that the demand for tokens is going down. I only see bottlenecks.” [33:42]
• The group shares war stories about cobbling together hardware and cloud for ML research, from buying Mac Studios for local engineering to discussing the limits of local vs. cloud workflows.
• Sam prefers everything cloud-based: “My laptops are truly dumb terminals... I don’t feel constrained by that at all.” [37:47]
• Dave: For running 50+ Codex agents locally, you need a beefy local Mac with lots of RAM. “You have to have memory and compute available locally on your machine in order to run that process.” [36:29]

---

5. X (Twitter) as Tech’s Town Square

Context: [38:54–45:07]

• Debate over X’s renewed importance as the locus for AI and tech conversation.
• Jess: “...X is not just the town square, but it's really like the AI town square. Do you guys feel that too?” [39:23]
• Sam: “I think X has never been stronger. I think it’s a great..." [41:03]
• Dave: “If you work in the tech business, if you’re a technologist, it’s where the conversation is... It’s not on Threads, it’s not on Reddit, like it’s definitely not in Facebook groups.” [42:13]
• The hosts note ideological clustering: X = right-leaning, Threads = left-leaning.

---

6. AI Talent Wars & “Slow It Down” Letters

Context: [47:01–54:42]

• Recap of drama involving Lillian Wang leaving Thinking Machines Lab for OpenAI, allegedly for “mental health,” then immediately starting at a high-pressure rival.
  Jess: “If you’re just leaving a company to go to a rival two days later, don’t pen an essay about mental health. That would be my advice.” [48:24]
• Dave & Sam mock the “slow down AI” open letters signed by thousands at Frontier Labs:
  • “If you don’t want to work on capitalist AI, you are totally welcome to come join the actual nonprofit AI Club…” [49:22]
  • “No, no, they want to get paid the same for doing less.” [49:19]
• Sam: “I think a lot of it comes down to people's sense of their unique talents and abilities, right? Where everyone has been told they're these special snowflakes...” [51:13]
• There's a parallel to Oppenheimer-like complexes—the myth of being the only one responsible enough to wield dangerous new powers:
  • “They believe they’re Oppenheimer. At least the leaders believe they’re Oppenheimers, right? And that they’re on this, like, existential mission where they will unleash hell, right, in the form of atomic weapons... But people really desperately want to be Oppenheimer in this age, or they want to feel the emotional importance of their lives like that. And it keeps playing out over and over.” [52:14]

---

7. Deeper Questions on Meaning, AI, and Isolation

Context: [53:39–54:38]

• Dave asks if working on AI is causing a new crisis of meaning, worse than the one social media triggered.
• Sam refers to his recent X post: “This is the most viral thing I’ve written... my whole thing about AI is a crisis of meaning.” [54:01]
• Dave wonders if AI workers' emotional fatigue is “coming out through X and through these types of letters... What's the cultural societal thing...?”

---

8. Miscellaneous: Conference Power Moves and Lifehacks

Context: [54:42–58:45]

• Discusses the savvy move of having an executive assistant proactively request dates for a conference before being invited—both a logistical tip and subtle power play.
• Sam: “That's a great way to get invited. Very, very smart.”
• Calendar invites and email hacks for busy people seeking access.
• Episode wraps with inside jokes about children running around, Breville machines, and tech world work-life chaos.

---

Notable Quotes & Timestamps

• “I think AI for everyone.” – Sam Lesin, [10:36]
• “So much of what Anthropic does is guided by anger and spite towards OpenAI and the people involved in OpenAI...” – Dave Morin, [05:05][19:25]
• “Caffeine runs Silicon Valley.” – Sam Lesin, [07:19]
• “You have to have memory and compute available locally on your machine...” – Dave, [36:29]
• “My laptops are truly dumb terminals...” – Sam, [37:47]
• “If you work in the tech business...it’s the only place you can go to have that conversation in the world right now.” – Dave, [42:13]
• “If you don’t want to work on capitalist AI, you are totally welcome to come join the actual nonprofit AI Club.” – Dave, [49:22]
• “Everyone has Oppenheimer complexes and think they're working on the Manhattan Project.” – Sam, [51:46]
• “This is the most viral thing I've written in, was like, my whole thing about AI is a crisis of meaning.” – Sam, [54:01]

---

Flow and Style

• Friendly, occasionally irreverent banter, heavy with Silicon Valley references (Star Wars, Lord of the Rings, Miatas, Breville espresso machines).
• Sarcastic and playful, with inside jokes balanced by deeper commentary.
• Occasional poignant pauses to question the trajectory of the tech industry and its impacts on mental health, meaning, and power.

---

Key Timestamps

• [09:01] – Discussion of Meta’s “AI is for everyone” campaign
• [13:13] – Star Wars analogy, Mark as the Jedi/founder messaging AI for good
• [23:15] – Critique of Meta’s centralized structure despite its “empowerment” claims
• [28:04] – China’s progress in chipmaking and its effect on markets
• [33:42] – Compute bottlenecks and pricing debate
• [39:23] – X as the “AI town square”
• [49:22] – Satire of “slow down” letters and AI worker activism
• [51:46] – Oppenheimer complexes among leading AI researchers
• [54:01] – AI’s impact on meaning and social dynamics

---

Conclusion

An episode blending sharp industry critique, techie life hacks, and playful cultural analogies. The hosts debate whether the “AI for everyone” narrative is real, deluded, or just PR, compare industry players with cinematic empires and rebels, dissect news impacting tech markets, and ask if the AI boom is spawning a new kind of existential crisis for technologists. The show’s blend of industry insight, sarcasm, and friendship offers a uniquely entertaining whirlwind through today’s AI wars.

---

For more, follow the podcast and hosts on X or at moreorlesspod.com

Transcript

[A]

Can we call Meta the Miata of AI companies?

[B]

It's the opposite of a Miata. You can't spend $100 billion on a Miata.

[C]

So much of what Anthropic does is guided by anger and spite towards OpenAI and the people involved in OpenAI. And because of that, I think what you end up with is a lot of opposing energy reaction to that.

[A]

Have you guys been following the saga of Lillian Wang?

[B]

I don't even know who that is.

[A]

So Lillian Wang was a co founder of Thinking Machines Lab, Mira Moradi's startup that is slowly but adapt, at least starting to release products so we know what they're actually doing. Maybe the end of last week she penned a lengthy missive, as one does in this age, about her mental health suffering by working so hard. So she was leaving and this was met with Mira cheering her on for taking care of herself and, you know, a whole wave of all that. Take the time you need. Yes. I read in the information today that she started at OpenAI.

[C]

That's funny.

[B]

I was a bastion of mental health. More or less.

[A]

Is he?

[B]

No.

[A]

Or is it? Yes. We'll debate the text. That's best when we get More or Less Dave and grip your salmon jest. Put it all right to the test. More or less. Why, hello. Welcome to More or Less Summer Storm Edition. Sam, you look like you're in the middle of a squall.

[B]

I was. I moved in. Now there's children running around everywhere. I'm trying to be outside so I can actually do more or less because

[A]

outside in a storm is the optimal audio condition for a podcast.

[B]

That's all I got, guys. That's all I got.

[C]

The weather is beautiful here in San Francisco.

[A]

Okay? I have a very good structured episode planned with lots of important topics, so I will just tune out the children in my ear.

[C]

I'm coming to you live from south park in San Francisco. AI ground zero.

[A]

So are you in a car? Am I the only person who managed to find a table? What's going on?

[C]

Yes, indeed. This is the first time I'm pulling a Sam during the More or Less pod. So I feel proud of myself for being a little bit less buttoned up this time around.

[A]

Okay. We're telling our audience exactly how much we care about them, which is a lot because we're willing to fit this podcast into extremely busy lives.

[C]

I was on a very important AI Compute related mission and, you know, had to get it done, so here I am.

[A]

Okay, well, that is one of our three topics today, so Buckle up folks, we've got Meta, its new ad campaign earnings, other things going on there. Mark has taken, taken up the pen and is writing op EDS like it's his second job. Big news out of the information about China catching up and some very important technology that wildly moved the stock market this week. So we have to talk about that. We do have to talk about this compute debate raging on X that seems to be very, very hot at the moment. Anything else you guys want to add to the list?

[C]

Let's go now.

[A]

Okay. Britt's working, so unfortunately we don't have her. I will try to channel Britt as much as possible. But first, guys, I have some big news this week. This has been a milestone week for me. Can I tell you about it?

[C]

Yes.

[B]

Is it about Benson Boone?

[A]

No, no, no Benson news at the moment. I learned to make a proper cappuccino. So if you'll indulge me for one second, I am probably. Sam and I are probably customer number one of the Breville cappuccino machine. This was an item featured on the informations holiday gift guide circa four years ago. And we won't say how many we have because it's embarrassing, but we have a number of them.

[B]

More than three.

[A]

Great.

[C]

Sam, wait, does this mean you're going to upgrade to a real espresso machine?

[A]

No, no, but I, I was, I was yesterday years old when I learned that the pressure dial on the front of the Breville is not a mere ornament, but rather a signal of whether you are getting the right amount of pressure through your cappuccino making process. And I've got to give credit where credit is due. The great Jessica Sittig, one of the most prolific espresso drinkers in the world,

[C]

has, she is one of the greats,

[A]

has tutored me in this and it's game changing now. Unfortunately though, it also led to an overuse of said machine that tripped all the breakers in our cottage. So it's not without its consequences. But. But it's been life changing.

[B]

What tripped all the breakers was the wire fraying to the cottages, which apparently is an emergency PG&E check or wherever it is, Constellation, whatever.

[A]

Well, it did only fray the ones adjacent to the cappuccino machine. So I still think there is a connection.

[B]

But no, it's like on the street, like literally several cabins don't have power properly.

[A]

Sam, are you letting facts get in the way of a good story?

[B]

Listen, I'm have to. I'm tethering on why I'm looking forward

[C]

to getting you guys into upgraded espresso gear.

[B]

It's never going to happen.

[A]

Well, I just want to say that I appreciate this because I feel like it has taught me to slow down and pay more attention in life in general.

[C]

So.

[A]

So because I've stared at that pressure gauge forever, all the time, and it never once occurred to me that I was trying to get the red Arrow into the espresso range.

[B]

I have a different interpretation.

[A]

Okay, Sam, you gotta turn.

[B]

Which is I listen, coffee's not cheap. And watching Jessica Sitzig sit there and twist the dial and pour out coffee grinds. And twist the dial and pour. This is like. Like pounds of coffee going through the machine and directly into the trash.

[C]

Oh, snap. I'm like, it's exactly how I'm spending tokens.

[A]

It's called a poll. It's called a poll. Like a pull request.

[C]

How is this any different than tokens?

[B]

I don't throw the tokens in the garbage.

[C]

Where do they go?

[A]

I think there was one or two spare polls. One or two spare polls. And honestly, the pursuit of perfect pressures.

[B]

Okay, for the pressure. I mean, I drink all the. I kept saying, let me have the ones that don't have the espresso mark because they're fine.

[A]

You got the slop, you got the AI slop.

[B]

I'm happy to drink slop all day long.

[C]

Fine. Drinking coffee slop.

[B]

It's. I don't understand people's like, I don't. Coffee tastes like coffee. It doesn't matter. But I'm glad that you have a new dial to look at.

[A]

It's so less bitter and it's so satisfying because you watch it and you're like, is it going to get there? Is it going to get to espresso rage?

[B]

I don't know. It sounds like we should introduce you to a Gap. Something called Calci. You can bet on it, Jess. We can have a Calci betting.

[C]

In my mind, getting the coffee to pull properly actually increases the amount of caffeine per shot. I don't think that that's the actual chemistry, but that's in my mind what's happening.

[A]

I just feel like I have a new artisanal routine for a gadget that I have a lot of Capex invested in, and I feel good about that. And Sam, can I have the slop? But I'm a new person.

[C]

Caffeine runs Silicon Valley.

[B]

So listen, Jess had a good insight, although, about these Breville machines, which is, you know, those, like, water bottle things with like 87 bottles of water saved this year because you use the water filter machine thing. This is a good insight. I really want that thing to have a counter and it would be even great if it was connected to the Internet as easy like you're like a rat hitting the button. And so I want like my. I want that data set and then I want a physical counter of the number of espressos that have gone through. Because that's kind of fun, right? Yeah, the amortization.

[A]

But we're seeing AI economics in our coffee habits.

[C]

Should we talk about tech?

[A]

Yeah, no, we're going to that, but I think this is important.

[B]

I will say, you know, the Breville's have also gotten cheaper. Have you noticed?

[A]

No, I.

[B]

This one was only like $500. The one before it was like 800. So the cost of. There's something going on with the mass production of Breville, Sam.

[C]

The Breville is like the Miata of coffee machines.

[B]

And I love Miatas and Boston Whalers. It's all the same category of thing.

[C]

Yeah, it's perfect for you guys.

[A]

Okay. Speaking of the Miata of AI companies, the last thing our friend. That people will never visit us again. If they just end up on the podcast, though we could usually, we could really use them. Can we call Meta the Miata of AI companies?

[B]

I don't know what that means.

[A]

Is that a pivot that we can make? I don't know.

[B]

I mean, they spend so much, but they have so much infrastructure. It's like the. It's like the. It's the opposite of a Miata. You can't spend $100 billion on a Miata. It's like a battle. It's like a aircraft carrier.

[A]

Yeah, fair enough. Okay, well, mark that. The shift to hard tech topics now,

[B]

Kimmy is like the Miata of AI.

[C]

Glm, maybe.

[A]

Sure. We're not. I actually don't have open source on the agenda at the moment, and let's keep it that way, so.

[C]

All right, let's talk Meta.

[A]

Did you guys see Meta's new ad campaign?

[C]

Yeah, I did.

[A]

Yes, Dave. Reactions. Two thumbs up, two thumbs down.

[C]

Are you talking about the. I mean, this is clearly like an integrated campaign where we've got letters, we've got interviews in the ft, we've got, you know, videos. Are we talking about the whole thing or just the ad?

[A]

Let's start with the ad and then go to the campaign.

[B]

I like innovation, not automation. That's like a good vibe. That was like kind of the one I took away from it. That I liked. Right. Was this whole like, AI is not for automation, it's for innovation. I'm not sure it's true or not, but I like the vibe of it.

[A]

Okay, well, let me recap this first. So meta comes out with this ad. And sharp listeners of the POD will note that I emphasize this point in the launch of their latest glasses, which was AI is for everyone.

[C]

Right.

[A]

That seems to be the new tagline. It's sort of the campaign tagline. It is a theme that runs through interviews and op eds Mark wrote this week or published this week. The idea being there's a narrative out there that it's elitist, that it's all about profit ringing that it's evil. We reject that. We think it is the latest tool that will further the democratization of technology in the hands of billions of people, fostering creativity, you know, entrepreneurial spirit, all that kind of stuff. Is that a decent summary of what you guys think the talking points of the campaign are?

[B]

Yeah, yeah, yeah. I think AI for everyone.

[C]

Okay. Of the ad campaign again, or the blog posts?

[A]

Well, both.

[C]

Right.

[A]

Well, Dave, what with that is a loose backdrop. What jumped out at you about it?

[C]

I mean, I think that this. I'm trying to pull it up right now, but I could not agree more.

[A]

But you have no Internet connection, so good luck.

[B]

We can tell you're trying to pull it up because you're crushing your Internet connection.

[C]

I could not agree more.

[B]

Either that or you're scrolling Instagram.

[A]

That's you, Sam. Sorry, you don't have enough bandwidth to shop while you record right now.

[B]

It's true. I don't even have a second device on me because my other phone's out of batteries.

[C]

So my understanding of this campaign is that Mark is trying to project a. An optimistic view of the future of technology and AI and that I believe it's in the blog post that he wrote that he just can't understand how someone who believes that AI is going to automate and take away everyone's jobs and sort of threaten the future of society would continue to work on building AI technology. Like that, to me, was the strongest statement that I think that he made in his piece. And I just couldn't agree more with this perspective, obviously, like, I'm speaking my own book here. You know, I'm spending an enormous amount of time working on OpenClaw and trying to give people a much higher agency version of AI that is much more personal and much more individually empowering. And I guess that's what I'm hearing from Mark, and it's the first thing that he's written in a long time that's kind of, I think, shining through and breaking through in a way that to me is very distinct and different for both Mark and meta. And I think it's really cool and pretty great. And I think it's awesome that he put the, you know, the muscle of not just pen, but the, you know, creating video version of that message and then it looks like they kind of deployed it out across the entire Facebook network, across every single surface. And so I do think people

[B]

need

[C]

to hear the optimistic side of this take. I mean, we've talked about the pessimist side and the doomer side of this across many, many episodes of this pod. So props to Mark for, you know, sending that message is my take.

[B]

I agree with that. I mean, look, it's just like Star wars is the dark side of the Force and the light side of the Force, right? And in some ways the dark side is easier, certainly easier message to sell and raise capital against and a bunch of other stuff, but ultimately the light side of the Force, which is this is like how you do it for good, is I think, hopefully ultimately victorious. And the longer and the better message, it's just hard for other people to pull off. I mean, I think it's very easy to sell the negative story, especially in a place of non incumbency. I think Mark is one of the rare leaders that actually is in a position as a founder at a trillion dollar scale who already has all the people on his platform who can try to tell the light side of the story, is what I would say. And look, they're probably both true. The Force does have two sides. It always does.

[A]

So do you believe it or you think it's part of the truth, or do we not know and we just need to put everything into a box right now? But no one really.

[B]

I mean, it's almost certainly all of the above. It's almost certainly all the above. I mean, like, look, I. I'm like case study 0 this because in a lot of ways, like, Jess knows this, like my relationship with my bot infrastructure has only grown exponentially. You. You told me. I was only allowed to tell you one more thing. The bot told me last night.

[A]

Guys, I'm worried.

[B]

You shouldn't be worried. But it's sick.

[A]

Last night, Sam is like, no matter what he's doing, if Sam has a smile on his face, it's because his bottom bot has told him something interesting about him. And on one hand, that's really cool.

[B]

It's so cool.

[A]

And on one hand that's like a little bit scary because like people and other people aren't gonna communicate that way to him. Slash me.

[B]

No.

[A]

And I just wanna make sure that like you have space and accommodation in your brain for. For human communication.

[B]

I. Look, I've kind of turned the corner with the infrastructure I built around AI personally where it's yielding just awesome stuff for me. Right. Like I'll give you an example. I. I fed it like four gigs, scanned of my dad's journals and work logs from like the late 80s through current day. And the stuff it pulled out of it like about history and it's just like flooring. That's cool, right? And so I was like kind of having a conversation back and forth with, with this infrastructure I built about like, you know, his personal life and history and business deals and career and different characters and like it's just like a level of depth in like you would never. It's like so cool like when you kind of. I also like have gotten really nerded out pretty deeply on like the health data side of it. So I now feeding it Garmin and Aura and Apple Health and like six other devices and integrated all the data sets.

[A]

Yeah, we talked about this.

[B]

But yes, well, but it's like. I mean I actually sent to someone involved in Aura. Like it basically was giving me like this morning. It was like, okay, like. Like they figured out my new thresholds for like zones of training, pushed it to my watch to update all the settings automatically without even like asking. Like it knew how to do it and what I wanted and then gave me the variance against Aura, which I then sent to someone who works at Aura on like where Aura is making mistakes in heart rate measurement and like it's just like getting so powerful. So the light side of the force is like I feel like I'm getting all this data about the world and my life and interactions and like making better decisions because of it. And like it's great. I think I'm like six clicks more aggressive and removed from how other people are using it. Like the way I have my stuff set up I think is beyond what almost anyone off the shelf can find with respect to OpenCL. You're just not gonna build it. You have to. Takes more than it's obvious, but it's great. So all for the light side on a personal level, I don't think that that is gonna change the world fundamentally. I just think it's good. Whereas I think the stuff. And by the way I could go on and on about four other use cases running for me that are amazing. Like, I'm like, there's so many things that I'm, like, thrilled about what it's doing for me personally. You know, on the professional side, you know, we know that the narratives run faster than reality. The narratives are super scary. The narratives are super distortionary. The power dynamics are super scary, like, blah, blah, blah, blah, blah. And we know that side. So I think they're both true. I think the question is, what story dominates? And, like, that's the unclear question. But if Mark wants to work on, like, taking the ridiculous level of infrastructure I've built out around AI and, like, trying to commercialize it and give it to the world, God bless. I mean, it is pretty cool.

[A]

Yeah. I think it's interesting to see now the real rival. I mean, everyone's just trying to say, like, we are not anthropic in some way because anthropic is winning. And anthropic also has some of the biggest problems in Washington. And so you're seeing all this sort of branding and messaging that sees, yeah,

[C]

the rebel alliance has formed.

[A]

Yeah, it's not. I don't think it's a bad time to be the empire either, Dave, but that's a different story.

[B]

I think this is a good. Let's keep going. Star wars dark side, light side, which is, like, if you had asked a year ago, who is the evil empire and like, who's the rebel alliance, you probably would have said that the evil empire is open AI and an alliance is everyone else. And, like, actually Nvidia is the emperor. Right. Like, it's like OpenAI is like the death Star Darth Vader, but, like, Palpatine is like Nvidia.

[A]

Right.

[B]

Like, there's some version of that, but

[A]

you're getting really into. Yes, okay, but.

[B]

But OpenAI would have been the evil empire, and Anthropic would have been part of the rebel Alliance. Now, like, things change so quickly that, like, maybe anthropic is. Although they're great. I mean, I use their stuff all the time. I love the empire, but it is interesting, like, to kind of how it's all playing out.

[A]

Your bot is an empire bot. Your bot is a dark side bot that you are using.

[B]

The core of it is like the core of fable is the core, like, decision maker. There's a bunch of other stuff it uses, but fable is like the ult ultimate decider in my infrastructure.

[C]

I mean, I think that this analogy is super interesting because the Star wars analogy. If you go to the scene, I believe it's in Empire Strikes Back. It might have been Return of the Jedi. Can't remember where Luke confronts the Emperor and Darth Vader. The scene plays out where Luke is having to decide whether or not to succumb to his anger. And the Emperor says to him, you know, give in, Luke. Give into your anger.

[B]

Let the power flow within you.

[C]

Let the power flow within you. And what's interesting, I think, about this analogy is that so much of what Anthropic does is guided by anger and spite towards OpenAI and the people involved in OpenAI. And because of that, I think what you end up with is a lot of the opposing energy reaction to that, Right? Like, you have this reaction happening in the valley where a lot of people are saying, like, why do you constantly, you know, talk to me as though you are better than me, as though you are angry towards this other entity, Right? And I think that that's, like, actually one of the more interesting things to talk about and look at here is there's a very human story playing out behind this.

[B]

Is it that? Or is it more Lord of the Rings where it's the ring of power? And, like, I don't know. The question is. And you. Basically, it's too tempting to use, right? Like, I. I don't know. It's an interesting.

[A]

Or is it mean girls?

[C]

It might just be mean girls.

[B]

Probably not mean girls, but say more.

[A]

You guys don't know that if Brit

[C]

was here, she would say it's mean girl.

[A]

Yeah, I think it's mean girls. But honestly. And. And I think it is also interesting to remember, though, that Anthropic came out swinging on the brand first by basically going right for the jugular of OpenAI when it was the evil empire with those, like, we won't run ads. I mean, do you guys remember. I think it was a Super bowl ad. Like, so this has been going on for a while.

[C]

Yeah. I mean, Jess, that's because that's the view, right? Like, it was literally born out of, we are leaving OpenAI. OpenAI is evil and is the empire. And so therefore, we are going to go create, you know, an opposing force. But then.

[B]

But just to push. Here's the other way to look at it, which is OpenAI is the original empire, right? And what they basically said is, there's a power vacuum on AI it's going to be so powerful that even if it's dangerous, we better be the ones to control the dangerous force ourselves, because otherwise Bad people will have it, and at least we're good people. Like, that was like, the original framing, right?

[C]

Like, yeah, I mean, Sam, that's the truth.

[A]

Well, actually, if we go way back, they were worried that Google was that. Now I'm just showing off how long I've been following this. But, like, it was called open AI because it wasn't Google. Yes, but finish your point. And then we're moving on things.

[B]

That's why it's like Lord of the Rings. Like, for we could talk about the Ring of Power. One ring to rule them all, one ring to guide them. Is everyone like, okay, this is a new power in the universe, right? And it could be the Force early on, whatever, but the Force is ancient, so it's a new power. And it's so powerful and dangerous that we have to be the ones to wield the power to save everyone else. Right. But then the power is ultimately corrupting. Like the Ring of Power, right?

[A]

Yes.

[B]

And so. And then basically, you have these generations of people being like, oh, my God, like, we have become the thing we feared and therefore we must leave to, like, fight back against the evil empire. But then, honestly, the power is corrupting, become. So the interesting thing about Marth, to take it back in the Meta. Meta is like the rare company that's powerful enough on its own that they're going to be like. They'd be very easy for them to go dark, right? And just be like, we will, you know, engage it. But they're like, no, we're powerful enough to, like, wield the light side of the Force, right? We can. We are like Tom Bombadil. We don't need to take the dark side of the Force, and we're going to take it on. I think that's an interesting. I don't know. There's something in here that's interesting, I

[C]

guess, though, that deserves a steel man on the other side.

[B]

Tom Bombadil is the best character in Lord of the Rings, for what it's worth. And I was really upset. I think Peter Thiel finally made a company, Bombadil, which pissed me off, because it is by far. He is by far the best character, even though he's missing from the movies.

[A]

Okay, Dave, last comment on this, and then we're moving on.

[C]

I mean, I think that there's sort of a. You have to steel man the other side a little bit, which is that you're right that Meta is one of the rare companies. I mean, it's one of the only companies with a founder still running it that serves consumers directly, that has enormous scale that can telegraph a message like this. But at the same time, Meta is still a centralized company. They run a massive centralized infrastructure that is dependent on advertising for the majority of their revenue. And you know, whether or not they will actually put the power in the hands of the people I think is yet to be seen. You know, one of the reasons I'm personally working so much on OpenClaw is that I really believe that we have to put the power in the hands of the people. Like you have to take the computer out of the cloud. Just like in the 70s, the computer, you know, the computer had been used to, you know, implement the Holocaust, right? IBM had used giant computers to, you know, do the Holocaust. And so a lot of what happened with the personal computer movement in the 80s was trying to get the personal computer as a tool into the hands of the people. And that's kind of the question that I have here is like, will Meta be a force for bringing this power truly to the people, or will it continue to centralize the resource and try to grab the ring itself? Like you're saying, Sam, and will it just be the next in the line? You know, will it be the next after?

[B]

But Dave, just to push you, when the decentralization is a great political movement in a lot of ways and like, there's a few examples like Bitcoin where it's kind of worked, but like in the Star wars universe, the Rebel alliance kind of sucks, right? Like they're not capable of winning. Like they're, they're totally losing. And it's only because you have a young Jedi come along that saves their ass, right? And so like, I guess the question for me is like, do we really believe that like a distributed open movement can win here against empires, or is it more that you need a Jedi?

[A]

We know Dave say your answer, but we all know what you think because you've been very consistent about it.

[C]

I don't think that it's necessarily about open, decentralized movement. I do think it's about, I think about it through the lens of the personal computer that like we must put in the hands of the people the most powerful possible version of AI so that individuals have true agency. Whether or not that's a for profit company that does that, like a future apple of some kind, or whether or not it's a, you know, open source project doesn't matter to me. It's matters to me more that it. That the directionality of the movement is going in that direction.

[B]

Hilariously. Though, Apple's a great story because, like, one. I mean, as a very smart friend pointed out to me recently, it was a political movement. It wasn't a technological one. Like, early Apple computers sucked, right? Like, it was completely a political thing of value that kind of drove it. But now, you know, years later, they're the ultimate centralized player. Right? Like, so it's like one of those things where, like, you become the thing you're set up to fight. Right? It's a political movement for decentralization that became the most centralized company in the world.

[C]

I don't understand that at all. I mean, I'm talking to. On a computer that I own, that I have all of my files. I do. You know, I have a phone. Like, I have truly powerful tools in my hands that I can use to do things, you know, I have.

[B]

But you can't use them however you want.

[C]

Yes, I can.

[B]

You can't put whatever you want in the App Store. Like, it's totally locked down now. I'm not saying, like you can't. Like, it's just interesting. Like, Linux is open. I don't think Apple's an open company.

[C]

I can put anything I want on a Mac.

[A]

Yeah. Okay, we've gotten off track here. So have we resolved this? Obviously there's openness at different layers, right? I mean, I think that. And it's interesting because Mark's vision isn't really explicitly about open or close. It's about putting the technology in the hands of people, which I think is aligned with your personal computer vision, Dave. But we will see Meta reported earnings that were disappointing, but mostly because they spent a lot of money in lawsuits and laying people off severance.

[B]

So what happened to the stock price, just out of interest?

[C]

I didn't see this. What's the. What's the headline, Jess?

[A]

I believe it went down.

[B]

I can't look it up because I don't. I don't have a computer I can use right now.

[A]

Oh. I'm also the, the scribe of the Internet for you to chucklehead to a shitty Internet connections. Because this has only been on your calendar for like three months. But that's okay.

[B]

Well, it's just like an AI situation.

[A]

Let's see, how is Meta trading right now? Meta. Meta trading. Down a percent on the day.

[C]

That's not that bad.

[A]

So down. Down a bit.

[B]

Oh, good. We can still afford the Breville. Thank God.

[A]

It's not that bad, but no, I don't. I don't think it's that bad.

[B]

No. A percentage that they did with the market expected.

[A]

And the Capex guidance, like, shifted slightly. Like the low end shifted slightly. But. But we're not talking about any astronomical swings here. Last week, which you didn't talk about.

[C]

I mean, you have to talk about the macro, which is the entire market is down today. Right? It's a down day.

[A]

So that's a great segue market wise into not so much today. But the information had a major story this week that we don't have to go deep into because I know you guys are all fans of lithography as much as I am, but our team out of Asia had a piece that China domestically has been making big progress in extreme ultraviolet lithography, which is how actually, sorry, not the extreme one, the other one, just deep ultraviolet lithography, which is a key step in chip making.

[C]

Right.

[A]

How you actually etch patterns on wafers. The extreme one, or enhanced or extreme should know this extreme is much more cutting edge. And China is still far behind on that. But this story obviously majorly moved the market for asml, the company that currently has cornered the market on this technology. But I read a lot of analyst reports about overall sentiment in AI that cited it as well. So I think it's really interesting. Not that we go deep into the tech stack or US China AI developments, but. But this is a big deal.

[B]

We know way more about Star wars than we do about lithography.

[A]

Yeah, but this is a big deal. And I think it is also significant, just more broadly to the point of the jittery nature of investors right now in terms of how this will all play out. But I think I've said this before, but I'm always surprised, like, you know, in Washington, tech execs go to Washington to say we have to be China and AI at all costs, privately or even semi privately. The exact same people say, obviously the US and China are going to both be leaders at AI. They sort of reject that the war narrative makes any sense. And. And then, you know, three weeks later they pick up the talking point that's expedient in the moment, which I get. But I. Having witnessed that, I. This has always been very top of mind for me. And, and so I think this, this story is really important in that regard. And.

[B]

Wait, Jess, are you saying that people say things in public that they don't necessarily believe?

[A]

Yeah, I do, I do.

[B]

Oh, my God.

[A]

We can't say who because, you know, we'll get really in trouble. But like, my jaw dropped hearing some of these people just sort of like, you know, a Moderator tees up a question about the war, and they're like, oh, it's a stupid question. We're both going to win. Next question. Like that is how dismissive they are sort of privately about this. So, I mean, yes, it's true that, like, China has not, you know, there have been areas of manufacturing or technology where China's had, like, a national push and not made progress. I think jet engines is one. Someone was telling me. Not an expert in that either.

[B]

Yeah, that's correct.

[A]

But it seems, like, completely obvious to me that, like, lithography, they're going to figure out, and they're figuring it out, and I think that's interesting. I'm also damn impressed with my reporters because that is not an easy story to break, let me tell you. Okay, let's talk a bit about compute. Dave, you said you are ferrying COMPUTE across San Francisco. Would you like to elaborate on that at all?

[C]

No, but.

[B]

Okay.

[A]

You can say no. I mean, but okay. When I think of compute, that's like

[B]

when you used to do the sneaker net, when it used to be faster to move out, to throw a hard drive in the, in your. In your passenger seat than it was to move it over the Internet. Those were the days.

[A]

I just feel like you used to

[B]

be able to ship hard drives to S3. Remember that? Do you still do that?

[A]

When I think of compute, I think of data centers, which don't.

[C]

Well, what's the question, Jess?

[A]

Okay, here, I'll ask a question. So my. I also actually want to briefly talk about X, which seems to become love

[B]

X mildly relevant again.

[A]

And also, guys, not since I have a strong Internet connection. I have been able to multitask throughout this podcast. And I just went to X to look something up and Mark just followed me on X. So this is a very exciting day. Mark Zuckerberg's following me on X. Like, more important is that is, like, he's on X following people. But to wrap up the compute point, the, like, hedge fund, Gavin Baker types of the Internet and tech investing seem to have, like, very tense debates about what's going to happen to COMPUTE in, like, the near term being like, next 18 months. And some people think that it's just going to get wildly more expensive, AKA much better margins for the. The businesses powering it. And others think the opposite's gonna happen. And, you know, it's pretty tight. Like, people are like, some of their analysis is based on, well, these longer terms contracts are gonna come up and then they're gonna have to renew at Much higher rates. I, I kind of assume that the labs have like figured that out and aren't just gonna have to like triple their AWS spend over. So I don't really buy that, which was sort of part of Gavin's analysis. But Dave, you're closer to this. Like, do you think there is a clear pattern on compute pricing over the next, say, 12 months, six months? What, what are all these investors in a tizzy about?

[C]

I don't really understand what the market is doing right now. My intuition is that there's something else that's causing this to happen in the markets. You know, it's, it's strange. Like you've got RAM stocks that are super down. Like Micron is like, I don't even down like 15% or something like this. There's a bunch of those types of analyses that are running around. I can't comment on what's going to happen with this contract or that contract. But here's the thing that I know from the front lines, which is that everyone is compute constrained and everyone, you know, at every manufacturer is trying to develop manufacture chips, ram. The entire supply chain is operating at, you know, complete capacity to fill the demand. That is, I don't see any signal that the demand for tokens is going down. I only see bottlenecks. I mean the actual. I was mostly joking with you earlier, Jess. The reason that, you know, I'm in San Francisco is that at openclaw we've been highly constrained on compute around the Mac operating system. Our ability to do multi agen agenda engineering is constrained by the amount of RAM and the amount of Apple silicon we can deploy against running agents. And so we've actually now worked in partnership with Apple to develop a bunch of Mac studios in custom configuration that is enabling us to unblock our engineering team's ability to be, you know, extremely productive. And so I only see, you know, engineering bottlenecks right now. And so I don't totally understand what's going on with all these analyses because my empirical experience on the ground is that we're constantly trying to figure out how to get more tokens, get those tokens at an efficient cost, deploy the right tokens against the right, you know, tasks and workloads. And that goes across the stack from frontier tokens to directly renting V3 hundreds and V2 hundreds to the Mac studios that we need on the desks of engineers to manage the agents in the first place. It's constrained across the board. And so I don't see where the dip in Demand is personally and I don't see it across my portfolio either. So again, I work at a different part of the stack, but that's what it looks like, you know, from my vantage.

[A]

Dave, your connection's getting a little spottier, so we might wait a second because it's going in and out a bit. But what to be clear, what are these Mac mini, What part of the process are they being used?

[C]

You know, in order to run multiple agents at once, your Mac has to manage each thread, right. So if you're running, you know, 1, 2, 3, 10, 100 agents at the same time, whether they're running through Codex, PI, openclaw or whatever, you have to have memory and compute available locally on your machine in order to run that process. And so you're only constrained by the power of the computer sitting on your desk to manage multiple at the same time.

[B]

But why are you doing anything locally? Like, I don't understand. I mean like I run very powerful computers in the cloud for things I want to do. I would never run anything locally on my machine. Like why do I do things? Why is anyone doing anything locally?

[C]

You need both. But like why your machine has to be talking to the ones in the cloud?

[B]

No, I have one machine in the cloud that manages the other machines in the cloud. Like why does it have to do. It's just a terminal on my desktop. I could do it on my iPad.

[C]

But you still have to talk to that machine.

[B]

I don't know. This is. I kind of get it, but not really. Like, I like why you have to have things on your desktop. Like I could code anything I want off my iPhone. Just tunneling into one machine and then having it manage a bunch of other machines.

[C]

Yeah, I mean, but you have to have that machine, whatever machine you're talking to, whatever machine you're coordinating or orchestrating with has to be very powerful. And you're right, sure.

[B]

I just. But why does that to be. It doesn't have to be local. I mean like this is the thing. Like I don't to me increasingly like my laptops are truly dumb terminals, right? Like I have laptops, they're actually I have one that's an off the shelf like cheapo error and whatever. Like I don't feel constrained by that at all now I'm not doing serious, serious engineering. But like I am curious what anyone's trying to do locally versus just like having, you know, an instance in DigitalOcean or AWS that then manages a bunch of instances that can be as beefy as you want it to be if

[C]

you're running 50 Codex agents at once from your Mac. Like you have to use a Mac with a lot of ram.

[A]

Like that's who needs to run Codex

[C]

agents in order to do engineering. It's like we're. That's, that's how we're doing engineering on openclaw.

[A]

Right. But like, what per. How many people are running 50 agents at once?

[C]

A lot. A lot.

[A]

Okay.

[B]

Our entire engineering team.

[A]

Well, the non engineers here. The non engineers here.

[B]

I just. I get it. I get it.

[A]

Abstractly, I have to update. By the way, meta stock is down 10%. I need to.

[B]

Oh, return the Breville. Return the Breville.

[A]

Return the Breville. Okay, so we're confused about compute. We see huge demand. The compute conversation continues. Let's talk about X for a second. Do you like Sam, you said you love X. You know, journalists have always been a love hate relationship with X, but it is really clear now that to me, like X is not just the town square, but it's really like the AI town square. Do you guys feel that too?

[C]

Yes, for sure.

[A]

And obviously we've had, well, Mark. Mark reinvigorated his ex presence.

[C]

It's the tech town square. It's not just the AI town square.

[A]

Jensen joined. Bernard Arnault also joined. So it's also the piss against the media town square. But yeah, I'm curious what you guys are saying.

[B]

Look, I think honestly I'm very pro. I mean, as a user, I've consistently used X since 2007. Although admittedly when I was at Facebook, I didn't engage that much for a few years. But it's always been great. I think Nikita has been doing a great job and I think it's better now than it was before. I think the conversation is stronger there than it ever has been. It's also really funny. I mean, I also use threads. I use threads out of loyalty. And I will say that it's pretty funny.

[A]

I end up in threads. I don't go to threads, but I spend time in Threads. I click on things on Instagram.

[B]

Yeah, yeah, yeah. Because you click on things on Instagram and then you end up. But here's the funniest thing. I very frequently will double post. So I'll post things on Threads and X at the same time. And it's really funny because on threads, threads is so left wing. It's like wild. Like anything I post there. Like I will have the left wing, far left communist mafia after me. Which is funny because like threat, because X is actually right leaning, at least in my purview. Right. So you have like this very funny thing where like you can post it, you can, you know, based on what you say and who you want to antagonize, how to completely flip out threads with anything that's even slightly right leaning or pro capitalism. Whereas like X is like all that. And so it is a really funny thing. But overall, like I think X has never been stronger. I think it's a great.

[A]

Are you saying there's no neutral platform?

[B]

No, there's no neutral platform, Dave, where

[A]

I feel like even when you. That was sarcasm, even when you had your I we, you like sort of half quit social media. But you, you stayed with X, right? Why?

[C]

Because I agree with you that X is the, I mean it's the network of technology. Right. Like it actually always has been. You know, X was born in the tech community. It, you know, I remember actually in the early days of X, it's funny, I'm, I'm sitting here in South Dakota park and I remember when Twitter started in south park and people would call it the, the nervous system of tech. Like you'd be able to get in there and you just kind of see the, the neurons of tech firing and it's meandered. Like Sam said, it's like gone in and out of various different phases and the, the feature set has changed and the people working on it have changed. But you know, since it launched in.

[B]

Well, the feature set really hasn't changed very much.

[C]

Yeah, I mean they've added things, taken things away, they launched payments.

[A]

I didn't touch that with a ten foot pole.

[C]

But yeah, there's parts of it that have gone and some, you know, whatever. And I agree, Nikita's doing a great job and Elon's doing a great job. But you know, I think it's. If you work in the tech business, if you're a technologist, it's where the conversation is. And so it's not on threads, it's not on Reddit, like it's definitely not in Facebook groups. Where else would you go? It's the only place that you can go to have that conversation in the world right now.

[A]

Are there? The other example of this was Demis Asabas posted his treatise on AI regulation as an X post. And so that was also really interesting to me. Does super technical discussion happen on X?

[C]

I actually didn't understand why Mark Zuckerberg didn't post his treatise on X as a post rather than behind a subscription paywall in the wall Street Journal.

[A]

Well, I think he did both. And don't, I don't like what you do with your subscription paywalls. But so behind, you know, a premium community, the message got out. The message got out. But there's definitely like a vibe shift on rival executives. So if you guys were in charge of X, what would you do? Because I assume it's still a bad business and in case you've forgotten, it's really a business nested in an AI company, nested in a rocket company that will also be nested in a car company. So it's pretty abstracted away from major revenue streams.

[B]

I don't, I think it's fine. I mean like look, it was quite a good vibe by, by Elon, right? And like I think it doesn't have to make money, right? Like I don't like to me it's like not a great business. It's fine for it not to be a great business. You know, paying 40 what billion for, for like a power play that isn't a good business is like a thing very few people in the world can stomach or do. And like obviously there's a lot of pain involved in like rationalizing it but if the thing kind of makes money and like who cares like let it like it's a great power lane, it means that when Elon wants to post something everyone will see it and that he and he can't be deplatformed. Just as Trump has tube social media and look again, it's an expensive non financial asset that way but it's a pretty valuable one is my view. Right. I don't think we'll ever see it be an important business but that doesn't mean it's not an important product.

[A]

Yeah, just interesting. Good job.

[B]

X hanging on in there by doing nothing. Basically what they did is they like took a thing, they made some obvious fixes, they made it not lose a shitload of money with a bunch of people who weren't doing anything. All of which takes guts and like what makes sense and they didn't fuck it up so nice.

[A]

That's a good business strategy. Don't fuck it up.

[C]

It's interesting. I mean one of the things that they added to X over the last, I think it's been in just the last two weeks was they launched this X money feature which has gotten almost no press and nobody's really talking about. I don't know how successful it will be but you know, it's a big money moving feature but it doesn't make

[B]

any you can't make any money. It's like how Venmo is an important product but never will make money and never has. Like, it's not like, it's not an important business.

[A]

Still, like, someone to unlock me out of, like, to put me back in my Venmo account. If anyone listening knows how to do that.

[B]

Oh, yeah, this is funny. So this, for anyone listening wants to know. This is very funny. I literally am like user ID on Venmo. I'm. I'm the third name on the ip. My handle used to be Sam. Three letters. Because again, like, I had handle number three, which has all sorts of fun stories. I got kicked off, like, and I can't get back on now Jess has been kicked off and can't get picked on. It's like wild. And so I actually messaged the founders of Venmo. I was like, guys. The actual founders, like, guys. And they were like, we don't even know anyone, so we can't get you back anymore.

[C]

I know who you should talk to.

[A]

I've given up. It turns out Zell works. I've given up.

[C]

No, no, guys, I know who you should talk to. I actually was trying to transfer some crypto through Venmo and it got really messed up.

[A]

And actually you should be kicked off. We've done.

[C]

No, they have a crypto feature. They've got crypto.

[B]

They do, they do.

[C]

It's a very long story. I was trying to pay somebody back for something that some concert tickets. And they in crypto more expensive. No, they were more expensive than Venmo's limits for what you can transfer. And so I was using. Yeah, yeah, crypto and the crypto somehow got lost and to Venmo's great credit, they actually credited me the money that they lost. And so I've got a good connection there now.

[A]

This would be most helpful, Dave. Thank you.

[B]

The number of customer service problems I've had to solve through much bigger company CEOs recently is hilarious. But if for a reason Venmo I can't fix.

[A]

No, we're. We're hopeless. Have you guys been following the saga of Lillian Wang?

[B]

I don't even know who know what's.

[C]

I don't know. I don't know who that is.

[A]

I will regale you in the saga, which is currently 80% of my X feed. So Lillian Wang was a co founder at Thinking Machines Lab, Mira Moradi's startup that is slowly but undoubtedly starting to release products. So we know what they're actually doing. And earlier this Week, maybe the end of last week, she penned a lengthy missive, as one does in this age, about her mental health suffering by working so hard. So she was leaving Thinking Machine Labs, and this was met with, you know, Mira cheering her on for taking care of herself and, you know, a whole wave of all that good. You know, take the time you need.

[B]

That's what you do when you fire someone.

[A]

Praise. Well, yes, I read in the information today that she started at OpenAI. Started back up at OpenAI.

[C]

That's funny.

[B]

Oh, the bastion of mental health.

[A]

So we are in that phase of the cycle. Much of the commentary is exactly what you said, Sam. A strange move for your mental health. Now, if we're being generous here, I can see how being at a bigger company and having a more defined role might be.

[C]

OpenAI is a big company, guys.

[A]

If you're just leaving a company to go to arrival two days later, don't pen an essay about mental health. That would be my advice. Keep it to yourself. Who cares? Because now poor Lillian Wang has made herself, you know, this symbol of something, including, you know, basically this phase of the AI talent war. But anyway, it might be pretty smart.

[B]

Let me. Let me. Let's play. Let's game. Let's game this out.

[A]

We're talking about her. We're talking about her.

[C]

My question is, what are all these people going to do that signed this. This other letter?

[A]

Which letter?

[B]

What's the other letter?

[C]

There's this other. There's this other letter where, like, a thousand people that work at Frontier Labs have signed a letter saying, oh, the slowing down letter. We must slow it down. And.

[A]

Yeah, I hate these letters. I hate.

[C]

I don't understand this letter, because if you want to slow it down, just quit. You don't have to work on. You don't have to work on AI Anymore.

[B]

No, no, they want to. They want to get paid the same for doing less.

[C]

If you don't want to work on capitalist AI, you are totally welcome to come join the actual nonprofit AI Club. You can come work on nonprofit AI at openclaw. We're happy to have you.

[B]

That's even more stressful. You have no resources, and you have to compete. Like, you come work with me. We'll settle Mac Minis around the Bay Area. Like, what the.

[A]

I also, so many questions. Are you still. Are you in a waymo. Are you still paying your Lyft or Uber driver? Or are like, what is your situation right now as we wrap? I. I want. Yeah, like, you're just sitting in the back Of a car.

[C]

I'm in my Tesla.

[A]

Your Tesla has been driving you around?

[C]

No, I'm sitting on. I'm sitting in south park in a parking spot.

[A]

Oh, I see you're parked. I thought you were moving for the first part of.

[C]

Ah, so great.

[B]

But look, in all seriousness, like, this whole AI slowdown thing, it's exactly like people wringing their fists for, like, climate change, right? We're like, we should use less power. It's like, it doesn't matter what you say. It's a global world. If someone else will just do it. Like, it's like. So there's this weird thing where, like, I don't know, I think a lot of it comes down to, like, people's sense of their unique talents and abilities, right? Where everyone has been told that they're like these special snowflakes that only they can do what they do. And that is, like, a big part of, like, how they just. By their lives and things like that. And so then they're like, oh, well, we're the only special people. So if we all say we should slow down and that's meaningful. It's not. There are plenty of people running on China can do the same. Right? And so I just think it's all a bunch of theater of, like, people trying to wrestle with the fact they're probably really overworked and really stressed out and don't know if their stock is worth anything and, like, want to be told they're doing good, but maybe they're not, and, like, a bunch of existential stuff.

[C]

So I do think this is all,

[B]

like, emotional existential noise from people that should be ignored.

[A]

Yeah, these letters don't sit well with me. Not because, I mean, if I were to argue the other side in developing these technologies, there are important choices to make that, you know, could broadly fall under, like, the responsibility of the model and that kind of thing. And I think it's important to draw attention to those. But, like, if you really feel like you're not at a place that's doing that, you just shouldn't work there. You shouldn't, like, attach your name to a letter.

[B]

I think what it is, honestly, is, like, all of these people have Oppenheimer complexes and think they're working on the Manhattan Project, right? And, like, there's a. This is, like, well documented and, like, it's well understood. And look, it's. The Manhattan Project is well documented. The fact they all have Oppenheimer complexes, right? And, like, they think they've invented atomic weapons.

[C]

Can you describe what that means? Like, what does that mean? What is. What is Oppenheimer complex?

[B]

They think they're, you know, they're. They. They believe they're Oppenheimer. At least the leaders believe they're Oppenheimers, right? And that they're on this, like, existential mission where they will unleash hell, right, in the form of atomic weapons. But they have to do it, because if they don't do it, someone else will do it.

[A]

But they're the one person who can do it in the best possible way,

[B]

but they're the one person responsible enough to do it. And then they're gonna launch the atomic bomb, which is AGI, and say, I have become Vishnu the Destroyer, and they're going to have an emotional complex about it, but be the most powerful man in the DOE in the history of the world. So if they're not Oppenheimer, they're von Neumann. They're like, in the game. And so then. And there was a bunch of angst about developing the nuclear bomb, as there should have been, and there was a bunch, but it's not the same thing, right? For a whole bunch of reasons. The atomic weapons were much harder to create. There was a much smaller community of people, like, literally dozens that knew what they were doing, and then huge armies of people who didn't know what they were doing around. It's like. But people really desperately want to be Oppenheimer in this age, or they want to feel the emotional importance of their lives like that. And it just keeps playing out over and over and over. And you're just like, guys, you're not Oppenheimer. Like, this is a very different game. This is a very different era. And AGI, while in theory, if you're believing, if you really believe you're building God is on the level of atomic weapons, you're not. There are gonna be lots of gods, right? It's more like a pantheon. And, like, it's gonna be super complicated

[C]

to bring it back to. One of Jess's questions at the beginning of the episode, you know, where Jess, you were asking, talking about Sam interacting with his agents. Do we think that there's a crisis of meaning going on that's even worse than the one we had during social media?

[A]

I read Sam's X post on that. I read Sam's X post on that.

[B]

This is the most viral thing I've written in, was like, my whole thing about AI is a crisis of meeting.

[A]

Well, Dave doesn't read everything you have. It is pinned, though. It is pinned.

[C]

It Is not. That's not what I'm referring more to these folks that are working in AI in San Francisco and, you know, interacting with agents all day, isolated to zoom. You know, they actually don't have an emotional outlet. So, like, maybe the emotional outlet is coming out through X and through these types of letters. And, you know, I'm actually kind of curious about what the cultural societal thing is that's going on here beyond just the Oppenheimer.

[B]

If the question is, are they all. Is everyone going crazy? I think the answer is probably, yeah.

[A]

Let's pick up these deeper questions. Let's pick them up. We need Brett as part of this conversation before we wrap. Anyone got some, like, culture advice? I got. I have one rant. One rant. You guys are in your many hats. You are also organizers of conferences. I am also an organizer of conferences. What do you say to the EA that reaches out to you on behalf of someone you invited to your conference last year to get the dates for next year's conference just so it's on their calendar?

[C]

Oh, what? Amazing.

[A]

Far before invitations have gone out. Far before invitations have gone out.

[B]

We were on the same email about this one. The person who did this we love and was obviously invited. Obviously invited. It's actually a brilliant move and I kind of want my EA to start doing to a bunch of conferences I'm not even invited to. Because if you just like, I'm going to reach out to Alan.

[A]

Alan, Jessica is looking forward to Sun Valley next year. Just confirming.

[B]

It's like, that's a great way to get invited. I was like, oh, fuck, we already invited them. Right? Like, so I think it's a great move. Very, very smart. And I'm in.

[C]

Super smart. Super smart.

[A]

So, like, both in one case, it was the person themselves.

[B]

Dave, when is the. When is the lodge next year?

[A]

Yeah, exactly. Well, I just. I laugh. What case was the person themselves?

[B]

We can invent conferences we want other people to put on and just ask for when the date is for them.

[C]

Yes.

[A]

Like, I'm just doing so.

[B]

You know what I mean?

[A]

Ten months from now. Ten months from now. When is your Davos dinner, Jessica? Yes.

[B]

Anyone you want to have dinner with? You're like, hey. Your EA reaches out like, hey, when? Or is Sam. Sam wants to know when the date is for our annual dinner. You know, blah, blah, blah, at Per Se.

[A]

And you don't want to like, ring this, like, poor EA's neck.

[B]

You find you message Benson's assistant and you're like, jessica would like to know when Our December dinner is.

[C]

There's an even better version of this, which is.

[A]

Okay, tell me. I love this.

[C]

Proactively put the thing on their calendar. With Google Calendar, you just put the event into their calendar.

[B]

People do this to me.

[C]

We did this for our holiday party a couple of years ago. We actually just put the event on everybody's calendar, and it works like a chart.

[A]

This is different. This is different.

[C]

No, no.

[B]

Yeah, but it's related. And I'm with Dave. This actually happens to me all the time, is people put events. And the problem is I have someone who does a lot of scheduling for me, and so sometimes it's unclear whether it's something I asked for or not or something that's approved. And, like, every once in a while, I'll end up in one of these. I'm like, fuck, I didn't for this. The same thing with emails. The best. The best move in an email if you want to meet with someone is start in the middle of a conversation as though you have already been talking to the person. Because then you're like, oh, like. And then it's a lot of work. I have to, like. Like, figure out if we actually were talking before. Like, whatever. So, like, the pickup in the middle with, like, all these assumptions is, like, a very smart BD move, I think, for anyone fundraising or doing sales, imply you're already talking, right? Don't start. Don't say, hey, like, I'd like to pitch you on a new thing that's immediately gonna get deleted. But if you're like, you know, love to continue, like, continuing a conversation from last week, you're like, oh, shit, what are we talking?

[A]

Picking this back up so good. Well, let's just say. Let's just give props where props are due. Major power move of the week is having your EA just run the dates by, you know, making sure we block it out for something you haven't yet been invited to.

[B]

Yeah, brilliant.

[A]

So take that tip for what you. What you will, people. We're bringing you all the news, all. All the inside baseball. And I think now, if you lasted this long with this shitty audio, God bless you. I actually want to know who you are.

[B]

No, it'll get fixed in post.

[A]

Please message me because you will be a forever fan.

[B]

And, Jess, I will let you know that the dinner has been put on the table, so we should probably eat

[A]

that indoors or outdoors. What's the weather like? I'm just, like, so confused, you know,

[B]

you're not that far from me indoors. And all the children are calling over.

[A]

So I'm enclosed like most people work inside. Okay, with that, we'll see you back here next week for another episode of More or Less.

[C]

Bye, you guys. Later.

[A]

If you enjoyed this show, please leave us a virtual high five by rating it and reviewing it on Apple Podcast, Spotify, YouTube, or wherever you get your podcasts. Find more information about each episode in the show notes and follow us on social media by searching for or lessavemorin at lesson at jlesson. And as for me, I'm Brit. See you guys next time.