Why Anthropic is watermarking every word Claude writes | E26

Summary

This Week in AI – Episode 26 Summary

Episode Title: Why Anthropic is watermarking every word Claude writes
Host: Jason Calacanis ("JCal")
Guests:

• Will Brick, CEO/Co-Founder at Exa
• Josh Sirota, CEO/Founder at Aragon
• Kash Ali, CEO/Co-Founder at Tax GPT
  Newsreader: Lon Harris
  Release Date: August 13, 2026

---

Overview

This episode dives deep into Anthropic’s decision to watermark all text generated by its Claude models—a response to the EU AI Act’s new transparency code. The roundtable of AI founders discusses the technical, ethical, and societal implications of watermarking, the broader regulatory and competitive landscape, and recent big news including Mark Zuckerberg’s new AI manifesto and Texas’s moratorium on new data center approvals. The panel closes by examining the potential “end of software” and how AI is reshaping the future of work, company value, and entrepreneurship.

---

Key Discussion Points & Insights

1. Anthropic’s Claude: Watermarking AI Output (Starts at [03:38])

• News: Anthropic will watermark all text from Claude models, using C2PA open standard, to comply with the EU AI Act (from Aug 2, 2026).
• How it Works: The watermark is embedded in the text and travels with any copy-paste—even through moderate edits. Applied at the model, not app, level.
• Background: The shift echoes moves by Substack, Instagram, and Chrome extensions to label AI-generated content. Watermarking originated as a way to track document leaks in big tech companies.

Panel Discussion

• Regulation vs. Opt-In:
  • Jason Calacanis: “I would rather this be something we opt into rather than this creepiness that, you know, it's forced on me.” ([10:52])
  • Will Brick: “Maybe there'll be regulation for any company running those models too. So maybe this is the start.” ([11:40])
• Impact on Users & Open Source: The panel predicts watermarking could push developers/enterprises toward open-source models that don’t watermark. There’s concern about user control and unintended friction.
• EU Perspective:
  • Josh Sirota: Highlights cultural resistance to AI in the EU, compared to the US. “It's almost going to be like a societal norm… not to use AI as much as other parts of the world.” ([07:32])
  • Kash Ali & Will Brick: Agree watermarking is low-lift, might have positive effect by increasing transparency, but may not meaningfully reduce “AI slop” or stop copying.

Notable Quotes

• “The new trend’s gonna be ‘no clanker’ spaces… spaces where I can just say, ‘No bots.’” — Jason Calacanis ([16:24])
• “We need proof of human.” — Jason Calacanis ([20:16])

---

2. The Dead Internet & ‘Proof of Human’ (Starts at [19:50])

• Discussion:
  • The “Dead Internet Theory” suggests bots massively outnumber real users, threatening trust and utility.
  • Calls for “proof of human” mechanisms echo KYC (know your customer) efforts in banking and age verification on the web. Sam Altman’s Worldcoin and the risks/benefits of biometric verification are touched on humorously.
• Automation Transparency: The group reflects on mandatory bot disclosure and platform labeling (e.g., Twitter/X’s “bot” icon, SoundCloud labeling AI music), their importance for public trust, and challenges ahead.

---

3. Zuckerberg’s AI Manifesto: Decentralization vs. Regulation (Starts at [23:01])

• Essay Recap: Zuckerberg’s “The Future Is For Everyone: The Path to a Positive AI Future” outlines concerns about centralization as the main AI risk, not runaway superintelligence.
• Key Points Discussed:
  • Decentralized AI Preferred: Competition and open models are crucial. More entities building AI is better for progress and safety.
  • Jobs + Automation: AI will enable personal and professional growth, not just job loss. “Recent statistics suggest... people will gain the ability to do many new things before their current jobs change.” (Zuck, cited at [23:38])
  • Open Sourcing & American Leadership: Meta supports sharing intermediate training checkpoints and closer, but flexible, government-industry collaboration, opposing heavy top-down review.
  • Hypocrisy Noted: Jason and others point out the irony in Zuck’s pro-open advocacy, given his record of aggressively protecting Facebook/Instagram’s data.
    • “When you’re behind, go open. When you’re ahead, be closed.” — Jason Calacanis ([44:11])
• Debate on Regulation:
  • Will Brick and Kash Ali contend that although open models might slightly increase risk, the positives (innovation capacity, democratic access) far outweigh the negatives. Bad actors historically have limited reach, and intelligence agencies remain adept at tracking them.

Notable Quotes

• “...it's so interesting... in that 6,500 words, that was the second thing I teed off of which was him saying ... if you can observe it, then you can learn from it. That was his way of saying I can steal any data I want to be competitive.” — Jason ([41:51])
• Panel consensus: American open-source models need a regulatory rethink to keep pace with global competition.

---

4. Texas Puts the Brakes on Data Centers (Starts at [51:12])

• Topic: Governor Abbott halts new data center approvals for audit—concerns over electricity grid overload (474 GW requested, 5x current peak).
• Perspectives:
  • Political Motives? Some panelists suspect the pause is partly performative, aiming to reassure voters.
  • Pragmatic Need: Others point out Texas’s painful history with brownouts and storms means the audit is common sense. Pollution, water use, and local climate effects of data centers (temperature rise, smog) are important to consider.
  • Will this backfire? Could produce copycat regulatory pauses in blue states.
  • Josh’s take: “The only thing I would say... the way this is written is he's freezing everything. The question... is... do they move capital to other states?” ([61:58])
  • JCal: “If the audit is, you fill out a form and say, we’re bringing our own energy... then we’re done.” ([62:41])

---

5. The End of Software? Prompt-Driven Development (Starts at [63:56])

• Provocation: Will AI agents and conversational interfaces eventually replace conventional software—and programmers?
• Josh Sirota: Predicts the “end of logging into apps” similar to how CDs gave way to streaming. Humans will use agents, not log into SaaS tools. The valuable asset will be underlying data, not the application interface.
  • “A human logging in, manually manipulating records inside of CRMs... is dead.” ([64:26])
• Kash Ali: Cautions that enterprises and SMBs still value packaged software for cost, support, and stability. While no-code/AI tools democratize tinkering, “people like packaged software where maintenance and cost are someone else’s problem.” ([66:41])
• Will Brick: At Exa, actual code-writing is rare: “Agents do all the checks before the code goes to production... we’re still hiring like crazy for software engineers... software engineering is becoming software managing.” ([72:06])
• JCal: The real differentiation—the new moat—is founders’ persistence and promise of ongoing support, not just code or features.

Notable Quotes

• “The new moat is founders who are so deranged your customers believe you’re never quitting, you’re never leaving.” — Jason Calacanis ([77:17])
• “[Software engineering is now] managing their agents... using their judgment and conversations with customers to decide what to build, that’s not going away anytime soon.” — Will Brick ([72:06])
• “The TAM of software development is the population of planet Earth.” — JCal ([81:24])

---

Memorable Moments & Quotes (with Timestamps)

• [10:52]: "I would rather this be something we opt into rather than this creepiness that, you know, it's forced on me." — Jason Calacanis.
• [16:24]: “No clankers, no bots... that opt-in for bots should be the new rule.” — Jason.
• [19:50]: “Dead Internet theory... the Internet, at some point, maybe already, filled with just not real humans and nothing is real. And that's a very bad thing for society, right?” — Will Brick.
• [44:11]: “When you’re behind, go open. When you’re ahead, be closed.” — Jason.
• [64:26]: “A human logging in, manually manipulating records inside of CRMs... is dead.” — Josh Sirota.
• [72:06]: “Software engineering has more becoming like software managing... everyone is managing their agents.” — Will Brick.
• [77:17]: “That's the new moat, is having a founder who is so deranged. Will, Kash, Josh—sound familiar? ... Are you so deranged that your customers believe you're never quitting, you're never leaving? ... That's the moat. I'm not leaving.” — Jason Calacanis.

---

Other Key Topics & Timestamps

• [38:54]: Law enforcement/AI monitoring for bad actors; joking about LLMs flagging teens searching “how to make a bomb.”
• [51:12]: Texas data center moratorium; balancing business growth vs. public utilities.
• [80:51]: New universal TAM for software: “the population of planet Earth.”
• [83:28]: San Francisco’s vape ban as a parable for common sense vs. regulatory excess.

---

Takeaways

• Watermarking is here to stay (at least in regulated markets), pushing new debates on open-source, regulatory overreach, and user control.
• Transparency and “proof of human” might soon be default expectations in digital spaces.
• Zuckerberg’s AI vision: Decentralization is the path forward—at least as long as Meta’s chasing rather than dominating.
• Data center proliferation and resource strain may soon become flashpoints in both red and blue US states.
• The role of conventional software is eroding. Prompt-driven AI agents will upend SaaS delivery, but human judgment, curation, and founder tenacity will remain the true advantage.

---

Panelist Contacts

• Exa (Will Brick): DM on Twitter/X
• Aragon (Josh Sirota): Email [e-mail]; “No resume needed, just show something exceptional.”
• Tax GPT (Kash Ali): taxgpt.com/careers; open to remote hires across North America

---

For more deep-dive AI discussions and startup profiles:
Sign up for the This Week in AI newsletter – new paid product with exclusive AI startup profiles launching soon.

---

End of Summary.

Transcript

[A]

Are you going to be here in four years because some business is going to be dependent on that product? Those people are lunatic, committed founders. That's the new moat. But are you so deranged that your customers believe you're never quitting, you're never leaving? Every founder, every vc? Oh, what's the moat? You know, anybody can write code. The moat is you. You're the moat. Will you quit? Let's stay focused on the mission. How absurd obsessed are you with the problem, the product, the market, the solution that you're building? That's the new mode. We're going to Mars. We're going to solve tax. That's what people want. Thanks to our friends at PayPal, the exclusive sponsor for this Week in AI. Try the payment and growth platform that's trusted by millions of customers worldwide. PayPal Open. Start growing today at PayPalOpen. Com. 3, 2. All right, everybody, welcome back. It's this Week in AI, another roundtable from your boy, J.C. howell. What's this one about? Well, it's in the name, folks. There's this Week in Startups. Hey, that's about startups. Then on Wednesdays, we do this Week in Venture Capital, we talk about venture capital. And then we do this Week in AI. I got three pokers in the fire here. And of course, all in, where we talk about politics, media and technology as well. With this roundtable, I focus exclusively on AI and we go deep, deep, deep every week. Thanks to our friends at PayPal for underwriting the first year of this Week in AI. What a great thing they did for me in getting this off the ground. Every week we like to rotate amazing builders. So this isn't journalists talking about builders. This is builders talking about what they're building and what they're seeing on the front line. That's the innovation I did in podcasting. Started 16 years ago, Lon 17 years ago with this Week in Startups. And I quickly realized, hey, having journalists as guests, not as good as having people who are actually running their companies. And then I'd have journalists online and they would be asking me questions because they didn't have any information. Hey, no dig to journalists, their important role in society. But you gotta go direct. And they'd be like, well, you tell us what's going. You tell me what's going on at Uber or Robinhood or in venture capital, because you're on the inside. You're in the room where it happens. So we got three great guests today for this Week in AI, who two out of three of them. I made an investment in. They're doing really great. These are all companies that are building the future. Let's introduce our guest, Lon Harris. My editorial director is with us. He'll be our newsreader today.

[B]

Sure thing.

[A]

I'll be world's greatest moderator asking questions along the way. But introduce our guests, it's going to be amazing.

[B]

Well, first we've got from Exa, he's the CEO and co founder. It's Will Brick. They're building a developer facing search engine that's specifically for AI agents. Jason. Then we've got Josh Sirota of Aragon, he's founder and CEO. Over there they're an agentic operating system for enterprises prompt interface that sits on top of your company's existing tools. Finally we've got Kash Ali from Tax GPT. He's the co founder and CEO. They are an AI operating system for accountants and tax professionals.

[A]

Well, at least I'm not an investor in exa. How did I not wind up getting on your cap table? Will, what's going on here?

[C]

Ignoring my DMs on Twitter.

[A]

I'm going to slide into some DMs here and get myself a little, a little tasty poo. I need a slice. I mean this. Listen, the whole podcasting thing is a front for me to get smarter and me to build my network and to share with you what I'm investing in and why and meet those people, but also find new people to invest in. All of that occurs through the launch for fund and the syndicate dot com. Enough with the plugs. Lon, what's first in the docket?

[B]

We were talking about it in the chat room before the show started. We got to jump right in. Anthropic has announced they're going to watermark text from Claude models. They're saying this is to comply with the EU AI Act's transparency code that took effect on August 2nd. It requires AI companies to mark AI generated or edited content in a way other systems can identify. So Anthropic says it will comply with all models released after August 2nd. And then they will go back and update their older models as well. They're going to use the C2PA open standard. The text. This is what's interesting. The watermark travels with any text that's copied and pasted out of Claude. So later somebody could go back and paste your text into some sort of a system that could read this watermark and it'll tell people like, oh, this was original content or, or this was written by Claude. Here's a quote from Anthropic. Because the watermark is part of the text, it will travel with the text when it's copied and pasted elsewhere and may persist through some editing. Watermarking will be applied at the model level, which means it will be present no matter which claude, product, or surface the text comes from. Not clear right now. Jason, how much editing you have to do to get rid of this text.

[D]

Watermark.

[A]

Yeah, you know, I. I know about all this because watermarking is super important. I have an investment in a company that. It was on the program. My friend Sky Dayton introduced me to it. I forgot the name of it right now because they changed their name, but somebody will look it up. And they were making watermarking Lawn not for determining if something's AI generated, but maybe five or six years ago, determining if somebody leaked a document at Microsoft, at Tesla, at Meta. So they started watermarking every document in the organization. These watermarks are a bit harder, I think, to remove for consumers, but this is a definitive trend. Now, Substack started using software to tell people, hey, we think you're about to post AI generated content. You know, we want to label that, and we want you to know that we know that you know that we know you're using Claude to write your substack. And that's not bueno. Then there's that company that has a Chrome tool. I don't know why they charge for it, but it lets you see and label things. You already have labeling on Instagram. You know, I like shark videos. I watch them, Josh, with my daughter. She loves shark videos. And now Instagram has forsaken me. They're giving me AI slop of sharks. And I'm like, ooh, this is like an actual great white shark, you know, biting a free diver who's taunting them. Excellent. I've been waiting for this. Cause I root for the shark, I root for the tiger. And then the shark takes, like, seven bites out of the person. And it's like, ha, ha. This is AI. And it's just. It's terrible. This is an important reaction. I will just end Jon Josh. The EU has never found a possible regulation that they can't put on everybody. I just think we should all assume this is happening. I don't know why we gotta watermark everything, but Sora was so convincing, Josh, that they put it on there. Pangram was the name of the company that does this. Pangram's got to make it free for, like, the first hundred things you check so that it's actually useful. It's not worth paying for for me, but I would use it and share my pangram stuff. What do you think of all this, Josh? It seems like a reaction to fake news, actual confusion on the part of users. And then I don't know exactly what the EU's ax to grind here is, but it does seem like a good thing if people are doing it. If people are watermarking stuff and platforms are identifying stuff that feels like a virtuous thing to do, I don't know if we need a regulation for it.

[E]

Yeah, I mean, I think it's a super interesting point. So I would zoom out on two commentaries, I think. First is this overall perspective on Eugene so was pretty interesting, I think. Last week I had dinner with someone who has their own venture fund and is seeing what's happening all over the world and with someone else who built a company that we all know, Microsoft. And the conversation was really around how is AI going to play out in different parts of the world? And when it comes to eu, just in general, I think there is a perspective on how it might play out where it is almost going to be like a societal norm. It could almost be cool to actually say, hey, we are implementing as many regulations as possible and we're taking the position that we're not going to use AI as much as other parts of the world. And that actually might be a perspective that they might take. And from our standpoint, we just actually started selling in EU and working with companies in France and Spain and others. I think there is a real caution actually to using AI. And so I think this stems like this particular, you know, watermark really just stems from this overall theme of within the eu.

[D]

This.

[E]

It's very, actually it seems it's like very anti AI. It's very detail. Right.

[A]

It's like we got to contain this thing is the vibes in Europe.

[E]

Yeah, exactly. Right. And so I think this is just yet another kind of mechanism. This, this watershed on doing that. And you know, curious, you know, Will, like, what are you seeing? I know like exas. You guys work, you guys are everywhere. Do you guys have any thoughts on what are you guys seeing in you and in general?

[C]

Well, so first of all, we haven't seen yet users asking so much about like, hey, can you remove AI generated content? I think they will, I think they will increasingly do that because as the world gets filled AI generated content, people are going to want to know, okay, is this like real? Is this biased? And so like there is a trend where People are starting to ask, and then it's going to be huge. So I think for the first of all, I think this is great. I think this is a great thing for the world. I think it doesn't seem like a big lift. It's something that shouldn't be so hard for them. And so I'm hearing about this and I think it's great. By the way, I don't think you need Claude to do this in order to have this functionality. You could easily train models that detect whether something's AI generated. I think they are obviously well best placed to get. Instead of 99% reliability of whether something's AI generated, Claude can have 99.99. It's all. It seems like it's statistical embedding of, of the classifier inside the text, which is really cool. Anyway, to your question, though, like, I don't hear much about this kind of thing yet, but I think it will be a big thing. I think you guys are saying it's decel. I don't feel like it's actually that it's. It's not really going to decelerate anything because it's not. It's a, it's a very small lift. So I think very good for the world. Happy they did it. Shouldn't. Shouldn't block Claude that much.

[A]

Will, should it be a regulation or should it be opted in your mind? I guess that's where I fall on. Like, I would rather this be something we opt into rather than this creepiness that, you know, it's forced on me. And you gotta think this is gonna become a big push. Will, for open source that doesn't watermark my stuff. Cause if I'm writing code or I'm making images, do I really want this creepy watermarking or not? I would like to have the option, say, hey, watermark that. Put Buy at Jason in the corner, you know, because it's a clip from my podcast. I want it watermarked. I want to know when somebody uses it. But there are other times when I'm like, yeah, I don't. This isn't anybody's business. I should get control of that. What do you think, Will? Of the regulation versus opt in?

[C]

Yeah, I guess that that is like the cultural difference. It's like, oh, should we just let. Let the models cook, do whatever they want? I. I think, I don't see why anthropic. I think it's better for the world. And then it's just a question of like, okay, is it just like a Slippery slope of regulation is like the question like okay, sure, like this regulation seems like clear net positive and not it doesn't hurt the model companies but it opens the floor for more regulation. Oh, your model has to do this and that. So I think in this specific case I don't think it's such a big deal. But yeah, the cultural difference between the Europe and America is pretty big. I agree with you that it will push people towards open source. Maybe there'll be regulation for any company running those models too. So maybe this is the start.

[A]

Yeah, then you would have to go after the hosting company. Well, because it would be like, you know who's hosting it for you now? Amazon Web Services and you know, Google Cloud and DigitalOcean they got to get in on it and they got to start watermarking people's open source. I mean, God, it's just friction. Go ahead Lun.

[B]

Yeah Cash, we were talking before the pod actually started like do you think if word gets out that this is out there in the world that it will actually cut down on people copying and pasting things out of Claude? Like is there a utopian version of this where people learn this is happening and then they, they do start, you know, using AI with themselves in the loop more?

[D]

Yeah, I believe there is a lot of AI slop out there from that lens. It makes sense. Google extremely now push a penalty on AI generated content in the last three, four months whether it's on YouTube or Aeogo. All of SEO they have very strong filters on that and then the quality is coming back and more human touch is coming back which is great for the Internet overall regulation of how EU's are forcing it. I won't comment on that but from my understanding I think probably will be a little bit of a net positive. I think from the business standpoint a lot of the time the client deliverable having this watermark they might give Claude OpenAI introducing another tier to produce work without the watermark I don't know how they will figure it out with a regulation standpoint. But for an enterprise, let's just say JP Morgan or Stanley are sending you a document and there's a watermark of Claude there. I don't know what will your response will be push back to them. We saw PwC customers are pushing back that you're using AI more and more so you should bring down your cost and everything. So this will be an interesting development to see overall how it will impact overall the business, whether it's going to be a net positive for the Open source models, more proprietary models or better revenue for Cloud and OpenAI and all of that. If EU is forcing Cloud, OpenAI has to do the same and Google probably eventually going to follow the suite. So overall I think it might be a little bit of net positive, but the ramification of that in the real world yet to be seen.

[E]

I think I would also just add like, it's not even so much the question is like, why is eu, out of all places, the ones that are introducing this regulation? And I think it just goes back to the point of like, this is one data point in an overall series of data points that's showing us in Silicon Valley what EU's perspective and kind of position is relative towards AI. So this is not a place where I would say AI has been diffused to a large extent inside of. Right. And so for them to already be taking this position of saying, okay, we're going to add yet another regulation, you know, around AI. Yes, I agree with literally everyone's commentary that.

[B]

Sure.

[E]

Is it, you know, good or bad? I don't think it actually matters that much whether someone has to say that this was generated or not. I think I would lean towards. It's kind of more annoying than anything, right? But sure, I think overall it's more an interesting question on, you know, what is the EU's perspective going to be on just diffusing AI throughout the territory and like, how, you know, AI forward are they. I think this is just another example showing that there's a real, there's a lot of work to do there from a societal perspective to get them to really lean in. So I think that to me that that's the more interesting piece here.

[A]

I think there's going to be a trend, Josh, in society where people are going to seek out no clanker spaces. Right? Like, remember in Star wars we had the hey, you're kind are not welcome here. And Luke Skyler's like, what are you talking about? He's like, you're droids. They're not allowed in the cantina. That's going to be the new thing. No clankers. And I want spaces, whether it's, you know, in my slack, in my social media feeds, on YouTube where I can just say, you know, I want a button. No clankers, no bots. And I think that opt in for bots should be the new rule. Reddit is getting so polluted. My replies are so polluted on Twitter with, you know, clanker spam that I put a $1 subscription to charity on my Twitter. So That I could enjoy it again. I have a million and a half followers. I didn't. I didn't know I'd gotten that many followers. But I couldn't use Twitter or there was no reason for me to look at the replies because it was so filled with spam bots.

[B]

Yeah, you know, just crypto schemes and.

[A]

Yeah, all of it. But then, you know, the crypto ones were easy to identify because there was like a hash and a link and you just block that person forever and you hide their comment. People using AI to say, hey, let's just write random comments to jcal and troll him in the hopes that they get followers, which is what they're doing. That to me, was the problem. I could actually enjoy my Twitter. So right now I just say subscriber only comments. Anybody else can quote, retweet it if they want. I'll see it. And it has increased my enjoyment. I bet you other people are going to move to this. And I think Reddit should be the first one where you can go on Reddit and you can have a clanker on Reddit, but you have to pay a fee and say it's a clanker, say it's a bot and associate that bot with another paying account. This should be the new rule. You can have a clanker. It's gotta be associated with a human. So no like, clankers running around amok without being associated with a human. Because I want to create a JSON bot on Twitter and have it tweet quotes from the show that are interesting or the archive. And I think people would like that. And I would like it. I'd follow it. But it's got to be associated with somebody who's human who can say, hey, this clanker's going awol. It's, you know, posting false stuff or it's spamming people.

[B]

You know, like this is like, it has that little robot icon you can put on it so people can see this is automatically generated content.

[A]

Is there a robot icon on Twitter?

[B]

Yeah, I think I've seen other people have it where it's like, you can label, like, this account is a bot.

[A]

Oh, yeah, they do have automated.

[B]

Yeah, this is an automated account. But it's like, it's cool. It's cool. It's not trying to trick anybody. You know, where you. You do actually see this already is like SoundCloud and those kind of like, because they were being inundated by so much like Suno, you know, produced music, like AI generated music has become such a huge thing that places like SoundCloud were getting like tens of thousands of submissions a day. And so now it's like they had to very clearly be like, here's the AI generated stuff. Here's the actual songs by people singing songs.

[A]

Actually, this does exist on Twitter now. Automated accounts are now labeled. Oh, wow. I've never seen a little.

[B]

It's like a little droid logo guy appears on them. I think I've seen.

[A]

There it is.

[B]

I've seen it in the wild.

[D]

Yeah.

[E]

I mean, you.

[C]

Have you guys heard of Dead Internet theory? Is this idea that like the Internet, at some point, maybe already it's just filled with just not real humans and nothing is real. And that's like a very bad thing for society, right? Like we, We. The Internet is very valuable. It has some problems, but it's mostly valuable thing. If we have bots, you know, 99 to 1, it's bots to humans. You can't trust anything. Then we lose the Internet. So I think anything that could help avoid that without too much regulation is good. I think this is a totally reasonable.

[A]

We need proof of human. Like, how do you get proof of human? Yeah, and it's. It sounds crazy, but, you know, there's opt in, there's regulations, but like, is there a way to do proof of human? I know, like, I've done banking stuff recently, and it made me. It makes me take a picture of my driver's license. Both sides? Yeah, like opening a bank account and then it made me take a picture looking into the phone.

[E]

Right.

[A]

And that was like. It's only in the last six months that this has happened where I've had to do this proof of human for bank accounts. But, like, maybe we need to have proof of human.

[B]

Well, that's what they're doing. That's how they're like age gating all of these. Like, if you want to join, you know, Discord or whatever, you got to like prove that you're over 18 with the.

[A]

Oh, do they do that now?

[B]

The facial scanner stuff? Yeah. And it's like, like again, Texas, if you want to go to certain naughty websites, you got to do that. You know, that's. That's becoming more and more of a standard already.

[A]

The face. Tell me more about these naughty websites in text.

[B]

I'm told that there's a route here that I would never know myself. I've heard it through third parties that.

[D]

Yeah, didn't the Sam Altman had that World Coin project where you can go scan your IRS and your eye and

[A]

they give us Some tokens. Seems like a reasonable thing that people would love to contribute to Sam Altman's collection of companies. Read the room, Sam. It is actually a great idea, but it's kind of the wrong person to do it. You know, like Zuck or Sam doing it.

[C]

Yeah, we actually have one in our office. Someone. Someone donated it to us, but we haven't used it.

[A]

The orb. Somebody donated an orb?

[D]

Oh, my Lord.

[A]

I would like to. When. I mean, listen, I don't wish failure on any founder, but when that thing fails, I would like an orb for my collection. I'm like General Grievous. I want to have, like, a collection of, like. I really want the Theranos machine. I will pay 10 grand for a Theranos machine. I'm just saying it right now. Anybody gets me, anybody get. Here's. Here's the thing. Anybody can get me one of those machines. I would either give you $10,000, or I will buy you a grand Seiko watch and take you to Ramen Katsu Kari, whatever you want. In Tokyo, I would literally take you out for the afternoon, pick a grand Seiko watch with you, and give you 10 dimes in cash to buy a Seiko watch. And I'll buy you lunch.

[D]

Wow. I saw the coolest memorabilia in a VC office. A SpaceX engine.

[A]

Oh, I can get one of those. Yeah, there's enough. I mean, I've been to Starbase, and there's, like, 20 prototype engines all over the place. Yeah, that would be a fun thing to own. All right, let's keep going through this, doc, all right?

[B]

We got a lot of other stories to cover, so we should talk about this Mark Zuckerberg Future of AI essay. It's titled the Future Is For Everyone, the Path to a Positive AI Future. I don't like using future twice in the title, but that's. That's Mark's choice. He argues that super intelligence's biggest risk is centralization. Not.

[A]

By the way, he did that to prove he was human. He made.

[B]

Oh, you think so obvious. Yeah.

[A]

Yes. He made an obvious error that any editor would have changed. Putting the same word twice in the same sentence. Or two. You're going to get that edit. He did that to be proof of human.

[B]

Oh, maybe like. Like putting the three dots instead of the M dash. That's what I do. So what he's building. He's building a personal AI agent for everyone. We got a little quote here. Everyone will have an exceptionally capable personal agent that understands you, your goals, and everything you care about. Your agent will work 247 on your behalf to improve your relationships, health, career, finances, home management, hobbies and more. He's also saying they're going to build powerful coding video design tools for everyone ways to launch businesses without teams or funding. PhD level tutors and coaches for any subject. They'll make contributions to science and it's going to be free or low cost access for everyone. Zuck also Jason noted this about AI job loss. He said recent statistics suggest it may be more likely that individuals capability growth could match or outpace automation in which case people will gain the ability to do many new things before their current jobs change. So we talked about this a little yesterday. The theory Jason's prevailing theory was this is buttering people up so that they won't oppose new data centers. I'm wondering if the if the panel agrees cash what do you think about that?

[D]

There is no Claude watermark so maybe to prove that to also that it's written by. I think for me the most important part in this was him introducing a structure to self regulate. I mean this pitch is not new. It's very like everyone will have a super safe like very smart agent working for them in all aspect of life PhD level tutors but giving board a structure to kind of decentralized regulation and giving board a tool that they can build something some safety guardrails. So I'm not sure the I think that was the most important part for me the bigger debate that is going on overall the world do we need to regulate AI for how much? I'm not a. I believe that the private sector will figure it out on our own. I think that was giving a starting a structure we don't need an FDA for models and everything. So I think for me that was the most important part of this conversation this whole essay.

[A]

What do you think the most important part was of this manifesto? 6,500 words. Josh, was there some piece other than everything that resonated with you particularly the one I was telling Lon resonated with

[D]

me was

[A]

if you embrace these tools you are going to be so valuable to your company that you'll outpace your boss's ability to fire you. I'm paraphrasing but essentially we're all getting fired and replaced by AI Unless you use AI tools and you're the fire and or person who knows how to use them. Which kind of goes back to that Jensen quote. You're not going to get replaced by AI you're going to get replaced by somebody using AI. But I Thought that was quite hopeful. And he kind of talked about the velocity in which you embrace the tool being the determining factor. Like, can people's embracing of AI out triple or better said, outpace the jobs that are obviously going to be retired?

[E]

I think this is a fantastic piece and an overall wonderful manifesto. And specifically, he touches on some of the following things. First, he says that right now the quest for super intelligence is really dominated by, frankly, two companies. And so from his perspective, and I would agree, I think for humanity, it's going to be much better to have a lot of different companies approaching this from closed models, open models, there's a whole new wave of building company models. I think the idea that a collective amount of companies working towards this overall vision of the superintelligence, I think is really great. And then I think, furthermore, he says every single big shift in human history, we've seen a lot of these transformative advances and we always have the fear that people are going to be left behind. But every single time this has happened, we've had greater prosperity, health, freedom is what his perspective is. And I think that that's very true here with AI. So it's a very optimistic manifesto. It's very, you know, pro human, I think. And I think what he's saying is all of the inventions that we've seen, you know, have come from, like he says, quote, you know, brothers in a bicycle shop, right, who believe people can fly, or the kid in the garage who thought that the personal computer could be for everyone. And what I think he's laying out is that AI is going to help those people, right, which is just us, regular, all, you know, just people be able to bring their ideas to life. I resonate with that a lot. And I think that this is going to be, we have the chance. You know, everyone on this call is working on AI in some, you know, way, shape or form. I think we have the opportunity. And what he's saying is to really shape, like, what this huge shift for humanity looks like, and that this shift is going to empower, like, everyone to bring their ideas to life. And then I think he lays out some of the points that Meta specifically, you know, is going to try to help empower this on this vision. But overall, I thought this was a fantastic piece, a wonderful manifesto from, from Zuck, and lots of this. You know, ultimately it's a big question, right? Like, the question overall is, hey, as this plays out, and if we do think there's going to be a huge shift, what happens to humans? What happens to the work that people are doing. If you look at companies today, Will's company, phenomenal company, Cash's company, you're building big companies with way less people. And then larger companies. We haven't yet had this huge AI diffusion moment inside of them where agents are doing everything but know we're working towards this. You're going to have larger companies that need way less people. And so then the question becomes, okay, what does everyone do? And his manifesto says, hey, you know, everyone now has the opportunity to use AI, you know, bring their ideas to life. And, you know, who knows, right? Maybe this is going to help the next Steve Jobs or the next, you know, HP or the next, you know, whatever it might be. So I think ultimately very, very positive piece. And good for Zuck, good for publishing this.

[A]

Will, you want to jump in here as to what your most salient moment was in this? I like your take, Josh. He does talk about, like, new jobs, and he talks about, like, making custom toys, making custom biologicals, making worlds, which is like his attempt to pump up the metaverse mistake. But I have to give Zuck a lot of credit for just, like, being like, you know what, that whole metaverse thing I spent 50 billion on, it's a complete waste of time. AI is actually the paradigm shift. It's one of the few times you've seen, like, a leader in real time just say, I made a mistake. I don't know if he actually said I made a mistake, but by his actions, Will, he was like, this was a blunder. Everybody stopped working on the metaverse. That 50 billion is going to go to H1 hundreds, A2 hundreds data centers. We're moving on. And the stock rallied massively. But what. What was your takeaway from it, Will?

[C]

Yeah, I mean, first of all, I always approach these things with a skepticism because any business leader writing a vision of the world is obviously going to be exactly aligned with what maximizes revenue for them. And like, you know, it's very hard to post something that. That goes very. That that would actually hurt you, hurt your business and your tribe, which is ultimately what business leaders care about if that vision came to light. So that's the healthy skepticism. I happen to completely agree with everything he wrote. So I really liked these Same. Same as Josh. And so I think. And by the way, I think he completely believes it as well. So I think in this case, what meta is building in AI is actually very aligned with what the vast majority of people want and what's good for the world. Right? They're like the Fundamental part of this piece is like centralization versus decentralization. And centralization financially benefits a couple companies, whereas decentralization benefits millions of companies. Right, so.

[A]

And humanity and.

[C]

Yeah, well, well. Okay, well, that, well, that, that it's a lit. Depends on. I mean, there are, There's a, There's a counter argument which I, I tend. I agree with you, Jason, but like, I think the counterargument is, Is a. One of. It's like this is like optimism. And then the counter argument is like one of fear. It's like, oh, no. If it is decentralized, then people will create cyber weapons and bioweapons and it's uncontrollable and we need some centralized power. And there's an argument there. I disagree with it. I think.

[A]

Why do you disagree with it? Explain why you don't need to worry about the cyber hacking, the biological weapons being created with this. Because he does address both of those. He has his own thesis. But let's unpack that piece. Well, your opinion matters too.

[C]

Yeah, yeah, yeah. If you jump into the details. Okay, what. Play it out. How does it actually work if there's a cyber, like some cyber threat or biological threat that. From decentralized models, like, can we handle it? And if, if you play out those details and it is. And you could just. Oh, anyone can make like a Ebola at the. That spreads at the rate of COVID Like in their, you know, microwave whatever is usually the thought, like nuclear weapon in a microwave type of analogy. And then, and then actually just destroy the world. That's very bad. I don't think that's the case. I think the, like, biological weapons are, Are limited by physical resources. So it's actually really hard for someone to just like, really just like, have a humongous effect on the world. Just, just, you know, in their, in their house. I think it's still possible, but I think that the, the positives outweigh the negatives there and there's always a risk of anything.

[A]

Right.

[C]

And so on the biological weapon side, I think it's mostly protected. And yeah, his, one of his arguments is that if people have all these different defenders, it could actually help. And on the cyber side, I think it's. It's even a less of a risk in the sense like, yes, there will be huge damages. There's some mass, like spreading of some, some virus, some intelligent virus, but that, that one feels like the kind of thing that we will learn from, handle it and be better off because of it. Yes, like this, like this Hugging face attack. Scary, but amazing that it happened right? Like there was no, no one was hurt. And we learned a ton from it. And now we're going to prevent against swarms. This is how humanity learns to prevent disasters. We unfortunately we have to make the mistake.

[A]

And then, and then yes, Fukushima was a huge mistake. They built that seawall too low in the wrong place. They were told, this is in the wrong place. You're going to get a hundred year storm and this is going to be under salt water. And they said yolo, we don't think so. And 50 years later, literally almost like clockwork, the tsunami hits and we're left here like, why are we so stupid as a human species? And he points out two really important things, Will. Well, if you really care about the biologics, like, well, why don't you just not let people buy the source materials and regulate that a bit more. And then on cyber, which is just the general term for cyber hacking, for cyber warfare, for hacking, he's like, well, the code's gonna be two things. He doesn't say the code's gonna be written by AI and be perfect. I'm adding that piece. But, but he kind of insinuates that that software is going to be kind of perfect because it's all going to be vetted this year. Like if it keeps going at this pace, like what software is not going to be vetted? Like you'd have to be an idiot to not have your software vetted by these latest models. Even if it's Cobalt and if it, if the majority of software is going to be written by these systems, hey, it's going to be pretty strong from the beginning. Yeah.

[B]

Will.

[C]

Yeah. You know, I mean, I agree, I think you would have to, to be against this, you have to make the case that like no, like the one event is going to kill us all. Like it's like the first biological, the first time. And so like I think the analogy of like, can you, if you can make a nuclear weapon in your microwaves, would we regulate that? And there's an argument that you would because then one person can actually do so much damage. But I think that the AI, like local open source models, at least right now are very far away from being as the blast radius being as significant as a nuclear weapon, so. And way less than that. Right? So, so like, you know, I think that the positives of decentralization are way overpowering the positive centralization at least right now. Maybe if the models get way better than my change. But at Least right now.

[D]

And interestingly, if someone can cook up next Ebola in their microwave, if everyone has access to the same tools, can someone also cook up the vaccine cure in the microwave? If AI is decentralized. So I completely agree with you on that part that the open source and decentralization of AI is very positive, net positive for the world. And more people having access to these tools are going to increase the capacity, I guess. And just playing devil's advocate, the only action that remains is a rogue state that has way more resources in the physical world also. So.

[A]

Yeah, but cash, aren't they all been trying to build bad things in the world from Al Qaeda to North Korea to Iran, like they've been at this already. Luckily, terrorists and evil people tend to be stupid and bombastic and they telegraph what they're going to do. And Israeli, US, UK, 007, all of our intelligence agencies bring the best and brightest. So to your point about the white hat black hat, the black hats in this regard tend to be dumb and they signal what they're going to do and they're easily identifiable. And with LLMs, you can be certain if somebody's typing in. How do I make a fertilizer bomb? The first thing Claude is doing, Dario has like a bunch of speed dials and he's watching everybody trying to do something bad. And he's got like this war room that he sits in.

[B]

He's got the phone that rings.

[D]

Totally.

[A]

He's like, oh my God, there's a 14 year old that just said how do I make the Oklahoma City bomb? And he just calls Emil Michael or who's the guy in charge of the Department of War now who does a lot of push ups and drinks a lot of beer. Yeah. He's like, pete, Pete, come regulate me. There's a 14 year old who just said, how do I make a bomb? Let's knock his door down. It's like it's a, it's a raid. 14 year old doing what 14 year olds do, which is look for boobies and type in things that send off alarms on the Internet. Like come on.

[D]

Maybe he ask Andy, Andy Jesse to do make the call. He doesn't even make the call by himself either.

[A]

He does a conference call. He's like, you know when you get the FaceTime call from the group chat and somebody asked dialed the like start a zoom. And everybody's like, does this guy even know he ass dialed us all? That's Dario. He just has like Trump Besant everybody. He's Pete. They all get on the same call, which is bad.

[E]

Jason, I think a huge, important piece. Just back to the Zuck like that we went over. I think one of the things that stands out the most is in the ensuring American leadership section. And so what he says is a very clear stance against what some of the labs are taking. So he says, and this is a direct quote, which is, I propose a collaboration model based on sharing intermediate training checkpoints and technical staff. Given how dynamic and competitive the AI landscape is, I think it will be more productive to have a close, proactive collaboration between Frontier labs and government, rather than a rigid process and review timeline. And so this is very clear against some of the labs, because what the some of the labs have proposed is, hey, we need a government committee that basically we release the models to. They have 30 days to review these models and then they can approve or reject it and then we can release it. And what he's saying is absolutely not right. What we need is just close collaboration. We do not need these committees, etc. And then I think the second point, I mean, you know, Jason has been a beacon for open models, like for, I think for longer than almost anyone at this point.

[A]

It's been two. We pulled the clip from all in from two years ago where I said, listen, I think open sour. Remember that clip lan that you found? It was like episode 50 or 100. I was like, you know, and I think open source might just run away with this. And they were, all the other buzzers were like, crazy.

[B]

Yeah, it was from like summer 2020.

[A]

You're an idiot. Yeah, remember, Jake Howe was right about everything.

[E]

It's honestly, it's been, it's been Jason and then Mark Andreessen, I think Mark Andreessen for two, like, he's been saying, you know, I don't think that the open, you know, race is over by any means whatsoever. Right.

[A]

Yes.

[E]

And, and, and, and Jason, these have been the two voices on the side. And so now Zuck is coming in under this American section and he's saying, right, that foreign labs have a whole bunch of competitive advantages against US American labs, namely with all these regulations that American labs some are proposing, have to be complying with. But he says, quote, for the US to lead an open source, we need to rethink our policies in several areas, including distillation, data use and training. So I think this is a really big point. Curious to hear what you guys think.

[A]

It is a big point. And it also, it's so interesting, Josh, in that 6,500 words, that was the Second thing I teed off of which was him saying, he says if you can observe it, lon see if you can find this. He says, like if you can observe it, then you can learn from it. That was his way of saying I can steal any data I want to be competitive. But he embedded it and he framed it as competition versus China, which is like the best strawman ever. It's like I get to steal because hey, the Chinese steal. It would actually be better if you license the stuff and just paid the New York Times for their content. Like that would be reasonable. Just because China steals the New York Times doesn't mean you have to. Just cuz they steal the Lord of the Rings doesn't mean that you get to steal it in the West. I thought it was like a really important point. Could you find what he said?

[B]

I did. Some have tried to frame distillation as harmful, but I think it is important to protect the principle that, that you can learn from anything, you can observe. This is how the world works and the US will not be able to lead if we restrict ourselves on this front.

[A]

Okay, so I can observe all the accounts on Instagram and make a database of every Instagram account, all of their stuff.

[B]

Not like that.

[A]

No, no. You can't distill Instagram or the social graph from Facebook. He stopped that. And he sues every startup I know because I've invested in a couple of them. He sues every single startup. This is the hypocrisy of Zuckerberg. Now I'm getting fired up again. If Zuck really feels that way, let us distill the social graph. The number of times Zuckerberg has sued Israeli, specifically companies and companies in India and the Philippines. Those three. He keeps trying to sue them for creating databases of influencers. And I know this because we had a startup, I wouldn't say which one, that had a heck of a time with the threats from Facebook, which was their partner, and then they kept threatening them like, oh well, we're going to pull all your access and we're going to sue you into oblivion because you have a database of influencers. Like F off, I can distill your database. Give me a break.

[D]

Jason, a question for you. On the same mindset, if they were one of the leading labs and they were part of the duopoly, let's just say. Will this be Zuck's take?

[A]

This is a great point. What you're pointing out, Kash, is a truism we've had in Silicon Valley for about over Two decades since open source actually worked. When open source first started to hit the consciousness of enterprises in the early 2000s, that's when MySQL Hadoop and a bunch of these databases started to go from. That's farcical to think that an enterprise would use them to enterprises saying well, we don't want to be locked into some big tech service that they can rug pull. Just like people don't want to be locked into Fable, which was actually rug pulled multiple times. That's why enterprises eventually said, hey, this is cheaper, better and more stable. Think about that. Open source controlled by nobody was better and more stable in the mind of enterprises. So we came up with this truism. When you're behind, go open. When you're ahead, be closed. So he is behind in terms of models. So he's for open distillation. He's ahead on the social graph. So he's suing everybody into oblivion when they try to distill Instagram. I'm going to ask that somebody come up with a startup that distills Instagram and Facebook and I would love to back that or somebody come up with an open source product that does that. Wouldn't that be an incredible thing, Josh, if you could go to GitHub and there was an open source project, I would put a bounty out for this. That just said here's the social graph as we've read it from a bunch of browser extensions or whatever that scraped it. Oh my God, that would be awesome.

[B]

I mean imagine if you could, could really like search and, and fine tune your searches of who you were looking for in terms of influencers across Instagram and TikTok. I mean you have, you're. We're locked into their algorithmic system. That would be amazing.

[A]

Such an.

[E]

Yeah, well, you know, will on, on. On the call like if, if exa had the social graph right of if it was open source. So that would be curious to hear what your perspective is there. But just quick remark on my side, I think. Yeah, I mean I think there's obviously right the catch 22 from his perspective where if you're going to take this position then you know, hopefully the actions will reflect that. This is. But over overall, overall like his point on American open models I think is important and hopefully his actions, you know, in broader perspectives will reflect this because we, you know, this is new, this is a new race for America. We have Mira's company that's firing, you know, seems to be starting to start firing now. We've got a new Company that was just funded. River, you know, hopefully it's a good thing for humanity. Zuck, if they can figure this out, if they can come out with an incredible open model, this is going to be a fantastic thing. Right. For the world, actually. So hopefully, you know, his points on openness and American, you know, open source, you know, I'm rooting for them. I hope that they can come out with something amazing. It'll be good for a democracy, It'll be good for humanity. And then to your point, Jason and Lon. Yeah. Right. So then fine. If you believe in open source, you know, help the community and. Right. And open source different assets that other people can use. Right. To bolster some of their models. Right. So curious Will like your perspective on that.

[C]

Yeah, I mean, everyone wants the thing that they don't care about to be open. It's exactly what Jason said. Like every. At least, like it explains everyone's position. It's like people in the lead want to be. Yeah, I mean, of course, like we really care about giving people access to information. And so like, it's almost like we're kind of enabling all the open source models to have a fantastic search engine. Right. Like every open source model, there's no open source search engine. It doesn't make sense. You could run a model like pretty cheaply on your laptop, but you can't run a search engine, so you kind of need to use some search engine. We want to enable all these open source models, wherever you are, whether it's in a company or on your local laptop, to like have access to whatever information you need. But yeah, I mean, within the limits of, you know, what companies are willing to make publicly available, I think the balance is correct. Like people, like, maybe there's an interesting thing between access to data and then distillation. They are different. Like, you know, lots of companies have all sorts of data that they rightfully so don't want to be out there. But then if you do, if you have a model like can. Can you con. Are you. Are you allowed to. So basically, if you have access to data, you know, you should be able to control whether that data is available to others. But. And if, but if you're a company that has access to a model, should you be able to control the way people use the end results of the. The model? I think it actually is kind of different. I could, I could see a world where like, data is more controlled, but model outputs should be less controlled.

[A]

This is a great episode because it got me inspired to like, because you Know, if you want to put out your manifesto and there's items in there that are incongruous with your past behavior. Mark, somebody clip this and send it to Mark Zuckerberg at his email. And I would sincerely like a response from him. And I. So I just tweeted, serious question, fink date. If you're pro distillation, does that mean that Facebook and Instagram will allow scraping of the social graph? Now, yes or no? And here's a link to him saying, we're going to sue you into oblivion, which he's done over and over again. I think that's an important question. You know where I stand on this. If you're breaking the terms of service of LinkedIn and of these social graphs, it's their right to their data. They should be allowed to get a licensing fee. So If Claude or OpenAI want to use this hard earned data from Instagram, they should get permission, just like they should get permission from Reddit or the New York Times. So this is like a very important moment because he has legal cases where what he said in the manifesto works against his legal cases. The same thing with Dario and Claude saying, oh, my God, you're distilling, you know, Fable, you're distilling Claude and then running to, you know, daddy, President Trump and saying, protect us and give us some, you know, protection.

[B]

They had that database of like hundreds of thousands of pirated books. And they, you know, that they, they were scanning, like, that's okay. But taking their models and studying them for your own model is not okay. It is.

[A]

It is a $1.5 billion fine. The largest copyright infringement in the history of copyright infringement paid by Anthropic just in the last 30 days. So take from that what you will. Let's keep moving through the docket. There's another couple of big stories here. Yeah, we like the end of software stories. Really good. You got some other ones you like, Lon, let's do it.

[B]

Sure. All right. Well, I do, I do want to bring up Texas, the great state of Texas, where you and I are right now. JCal, they're temporarily halting new data center approvals. On Monday, Governor Greg Abbott ordered the Public Utility Commission of Texas and the Electric Reliability Council of Texas, ercot, we call them here, to audit all proposed data center projects in the state that connect to a public power grid. Governor Abbott has issued a moratorium on approval for all data center projects until the audit is complete. It's unclear how long that's going to take. He's asking for basic information in the audit, who owns it? Will the data center receive tax breaks? How does it use a generate power? How does it use water for its cooling operations? And so on and so forth. Ercot currently tracking 1800 data center projects representing 474 gigawatts of electricity, or more than five times the grid's record for peak demand. 90% of all new power requests right now in the Texas Q are data centers. Here's a quick quote from Abbott before I throw it to you guys. Our top priority is to protect Texan safety and quality of life. Any project that fails to comply with the requirements set forth by the PUCT and ERCOT and by state law must be denied connection to the Texas grid. Simply put, Texans must come first. So I guess my question to you, J. Cal and the rest of the panel. I thought Texas was supposed to be the leg come build anything here pro business. Is it a bad sign that even Governor Greg Abbott is saying, hang on a second, we got to hit pause.

[E]

There's probably like two, two cases for this. You could take the positive optimistic one and then you can take the more maybe cynical like political one. So we can start with the cynical political one. So I think from a population basic, we have to remember, okay, when you go, I'm in, I'm in San Francisco, I, I, you know, here, right. Like everyone is, you know, pro AI and all of this stuff. But when you step out, okay, it's, this is not necessarily the sentiment even in America. Everywhere. Everywhere in America. And so I think from a cynical perspective of maybe gathering more votes, this data center is going to become this lightning rod issue. I think of a lot of these campaigns that are happening, states that are pro building it, states that are against building it. And what he's asking for is information for things like, okay, how is this going to impact electricity bills? How is it this going to impact water use? How is it going to impact noise and you know, local jobs and all of this kind of stuff. I think the second thing inside of here which is interesting is the leveraging and gatekeeping. So he's saying that an audit must be completed before any project moves forward. So it kind of turns the governor's like appointees into basically these huge regulatory cap. They have so much power now, right? Every hyperscaler has to pass these people that, that, that he appoints. So who are these people? What do they know about data? What do they know? Like this is going to become pretty big thing. And I would say like on the having it both Ways side. Right. Like he made Texas kind of like years of marketing itself as the place for data centers. Right. Like this is, this is a huge position that they've said. And so I think the other side is that this is an audit perhaps where like there's maybe there's more fluff here. And it's like a nice political stance to take saying, hey, we're gonna take a little bit more.

[A]

A reasonable one might.

[E]

Yeah, right.

[A]

Might be my take on it. If I'm gonna give the most charitable. And you gave the cynical, which is great. I think the most charitable one is people in Texas are very sensitive about the brownouts we've had. People have died. You know when this happened during the two weeks of unbelievable winter, we have surrounded by 100 degree temperatures. We have these like two frosty weeks. And listen, people's homes, the plumbing, the electricity is not designed for ice in Texas. So if you do get an ice storm during that two week period, which happens every couple of years, it's cataclysmic. So people are sensitive to it. Yeah, there should be an energy audit. Obviously there should be and sounds quite reasonable to me. The question's really, I think cash gonna become very quickly pollution from these. In order to get these up and running today, you need to burn natural gases, you need to burn fossil fuels in whatever form you want to do that. And that's why people are buying turbines. Elon personally bought a turbine company that was very under the radar. He bought it for like a billion dollars. Really quick, the boom, supersonic making that jet plane decided, hey, let's do a mini pivot to do the engines because there's so much demand for these. These things pollute and they raise the temperature of areas. So now that's like a very nuanced

[B]

cash

[A]

thing that people have to I think contend with and answer a very valid question, which is, hey, you want to go fast, you want to burn fossil fuels, is it going to just be dirty and people are going to lose years of their life and we're going to have a smog issue or is that going to raise the temperature in Utah? There's issues in Utah where like certain valleys, the temperature raises 2 degrees, 3 degrees on average for the miles around it. And what impact does that have on the environment? And so you don't have to be like a climate nut to not want to have this kind of pollutants. What's your take on that, Keshe?

[D]

Yeah, I think I agree with Josh from pitching this a little bit of a Lightning rod issue. When you go into a little bit more details of this story, they are tracking 1800 projects and 1800 and that demand of electricity will be 474 gigawatt which is five times what Texas already produce. So I mean if you want to see, take a look, give them a little bit of a favorable understanding. But they were trying, what they are trying to understand is what is going to happen if all of these 1800 projects go through and we have, we are producing five times more electricity. What will be the risk to the pollution, water, all of those. So I think that's, that's how I understood it. They are trying to audit it. But from the AI company's perspective, when I say it, okay, Texas used to be like you can always go build there. Then where else you go build, where's the home where you can actually like scale and everything. So let's see how that audit go. But I think it's not in my backyard type of mindset. It's actually true. It can be a true crisis for the quality of life on people. 1800 projects going on and five times the electricity. And I didn't read what time frame that these projects are looking at but it has to be a little bit more thoughtful approach. And I think government is trying to do that audit. But from the AI company's perspective, where do you go from there?

[E]

Yeah, I think a couple things just to start. So first is a lot of this is centered around something called ercot. So what is ercot? So it's the Electric Liability Council of Texas, which is quite interesting because it's a non profit that actually operates 90% of Texas electricity grid and it's about 27 million customers. Right. So you have this ERCOT committee, right. That is a non profit. It's actually not run by, right. The Texas government and they're responsible for 27 million residents. And so what they're saying is okay, hey, they're regulated. You go up one notch, which is where he's at, Governor Abbott is at by this thing called puct, which is the Public Utility Commission of Texas. And so the Public Utility Commission of Texas, that's where Abbott can appoint a commissioner to basically oversee kind of this ercot. Right. And what ers, what ERCOT is saying to your point, cash is hey, we have 474 gigawatt connection requests. This is five times higher than the highest demand that Texas has ever hit. Like we need to ring the, like what are we doing? We need to actually not pause but we really need to get this under control. And as far as the data centers, what they're saying and them being ERCOT is they're saying out of the 474 gigawatt of requests, they don't know how much of that is coming from the same data center. So they actually don't have visibility into that. So if they don't have visibility into it, they're saying, how can we plan the grid for the 27 million customers that we support if we have no idea basically where the requests are coming from? Where are their duplicate requests coming from? How can we better route this traffic? Right. And so ultimately, I think there's a lot of stuff here that, again, I think that makes a lot of sense. Right. Especially for. To your point, Cash, you know, the residents of the state of Texas, right. I think that they have the right to know these details for. And I don't think this necessarily goes against actually Abbott's, you know, in Texas's vision of, hey, come here and build. I don't think anyone's saying by any measure, hey, no, no, no, we're closed for building at all. I think what they're saying is like, hey, these are two things can be true at once, which is, hey, you know, we're going to be the place to come here and build. Right. And we're going to have the granular details to know. Right. How we can make these projects a reality.

[D]

Yeah. I think my fear is that a lot of blue state can take this and misread it and say, like, well, Texas did that. We need to figure it out on our own. And that kind of become a de facto ban overall. Building new data centers and these new regulations and new structure emerge by that time.

[E]

Yeah. The only thing I would say, like, to add to that, the only, like I would say, in my view, the area that is like, interesting for discussion. Curious what you guys think is the way this is written is he's freezing everything. So the question I think here becomes, okay, hey, right. It literally halts every single shovel. Okay. In Texas. And so I think the question here is like, okay, well, there's been a bunch of these that have already been funded. They've gotten the capital needed for this, They've gotten the approvals ready for this, and they're ready to go with the shovels. And now all of a sudden, hey, what happens there? Do they move the capital to other states?

[A]

So the devil's in the details, basically. If the audit is, you fill out a form and say, we're bringing our own energy byoe then we're done. I'm doing like, Crusoe Cloud is doing a nuclear power plant. Other people are going to the Permian Basin and putting their data center next to nat gas fields. It's like, yeah, we're good. And we all know Elon has access to solar and batteries. Like, people are going to be able to just check a box. Yeah, we're good. And I bet you Abbott is so pro development that he's just gonna like. I don't wanna say rubber stamp, but be like, okay, great, you're good. And if you said, I don't have a plan, then it's like, okay, come to us when you have a plan.

[B]

Yeah.

[A]

And maybe this will actually accelerate the ones with the plan and then penalize the ones that are yoloing it and wanna start building without a plan. Which would be good for the ones who do have a plan. Yeah, it actually seems reasonable if it's executed.

[B]

Well, worth noting that this only applies to data centers that want to connect to the public grid. If you come in and you're like, I got a. I got a plan. It's on site. We got our own power. You're. You're still good to go.

[D]

Yeah.

[E]

Very important distinction.

[B]

Yeah, that's all right.

[A]

Let's keep moving.

[B]

Yeah. Well, I wanted to talk a little bit about. Is software dead? Are we at the end of software? Everyone from Wall street to our own panelist Josh has been suggesting conventional software coding days are numbered. Soon we'll just all be building things for ourselves directly from prompts. I want to know what everybody thinks about that. And, and Josh, because you sort of made this prediction. What's the timeline? What's the horizon like? At what point will everybody be putting their coding away and their software away and just start asking a chatbot to make things for them?

[E]

Yeah. So I think it's already happening. And then like, the caveat is I use the analogy, you know, when we used to listen to music, someone, you know, we used to get a CD and you'd have to go to this record store and buy the new album and we can listen to songs. And then today, obviously, you just click a button and you stream whatever song that you want. The analogy, I think, in software is I think that a human would log into a piece of software is going to be the same thing as using a cd. I think in a couple of years, if anyone says, oh, I'm logging into Salesforce or I'm logging into this business application, I think will be like, that's Crazy, right? So for software, two things will play out. Like one is, hey, if I have a CRM or this big enterprise application of my company, the question is do I need humans logging in and accessing the interface? So what I mean by is it dead is I think that a human logging in, manually manipulating records inside of CRMs or ERPs, I think that is dead. Right? And the question that a lot of the companies we work with is hey, you know, I have 2,000 licenses for a CRM, I have a thousand licenses for the CRP. Why can't I just have, you know, five licenses? And then why can't agents just access all of the underlying data and go and do this record? So it's a little different than the hey, I'm going to vibe code my own enterprise CRM. I think that that's a little bit away,

[A]

but it does challenge the per seat model and moves you to consumption, which cash you have a per seat or consumption or both model. Because that I think what the AI startups are deciding to do is what's going to impact the people prior. And I'm sure there were accounting, not copilots, but accounting software that were on a per seat basis that were extremely expensive. And then you have I guess either a choice to pick a business model here, what have you chosen and what's worked and what doesn't.

[D]

So we have the hybrid model where per seat basis you bring your team in and then you pay for the outcome based pricing based on outcome. So if you need to prepare a return that a human Preparers will take four hours, you can pay $30 one credit and get done in 15 minutes. And our whole goal is to add productivity into the teams. So imagine a very complicated return from 1065, 1120 that a human preparer reviewer takes the whole day or maybe even the whole week to prepare. All of a sudden their productivity is 3 to 5x and they are able to do 3 to 5x more and people happily pay for that because what we are selling to are financially very smart people. They see the ROI and for us our software is impacting the human cost, the resource cost, we are dipping into that. So this hybrid model works for us and I mean we've seen it within the six months with like a third of our revenue is agent actions already within six months, so working very well. But taking a step back, what Josh was talking about and the greater debate of is software that or is it going to be dead? I think my opinion will be against it. I think all of these enterprise is still in the decision Paralysis, build versus buy safety, change management. So many of these issues, they are not moving at all or moving very slow. Everyone is confused. Internal politics, all of these. They look at the internal team and they say, oh, we can build it in our house, in house. Why do we need to do that? Right, so just keeping enterprises on side. There are 33 million small businesses in the country that take 46% of US economy between 1 to 25 million in revenue. I don't think they're going to code all of their own CRM. I mean you can, technically speaking you can have your, you can build your own email server, but we don't have server rooms in our basements or in our spare room. You can actually have your own protocol of private messaging. But if we don't build it, we can wipe code it. We all can wipe code it. We are happy to give our data to WhatsApp and Signal and all of those things. I mean signal is different. So I believe people like packaged software where the maintenance and the cost are someone else's problem and it becomes truly economy of scale. People like more productivity, cheaper and faster. If I want to just say I'm a small business owner and I want to code my own CRM or anything, what will be the token cost of that? Would it be the maintenance? Electricity? Are we going to go?

[A]

There's a lot of hidden cost on maintenance alone and then there's business disruption. I think the SLA is kind of what you're alluding to here. Cash. And I think that's where I get concerned. My team has been vibe coding incredible software projects and I go to my team and I'm like, who's going to maintain this? It's fantastic that you've built it. Who's going to maintain it? Tell me about the security. Lovable does a really good job because they have a perspective on security databases built into it. But if you're just vibe coding with the coding tools, I find them quite random in how they apply stuff. And then the people who are building, they don't know code, they don't know what's brittle and not. But I do think software as a problem set is about to be solved. What do I mean by that? I think the concept of writing code is going to be abstracted to the human language and it's starting to happen even with developers who are pushing code and there is an agent that checks it. And are they actually looking at the code? Well, now are people who are even able to look at the Code? Are they even bothering to look at it because they trust it so much? I guess. Is it the end of code? If developers describe what they want, somebody else in the organization checks it in and uses AI to proof it, and neither party looks at the underlying code. Does that mean the end of code?

[C]

Yeah. I have a lot of thoughts here. I'm trying to organize on my head as you guys are talking.

[A]

No, it's kind of mind blowing when you think about it. At what point do you just say, you know, it's kind of like music? At what point do you say there are no more musicians? Music is solved. It's just people pressing buttons, beep, beep, boop. I sampled a bunch of stuff and nobody in this quote unquote band actually knows how to play the songs. So when we go on tour, we're also hitting the play button like a DJ does.

[C]

Yeah. Okay. I give the perspective from exo. So we have, you know, a lot of engineers and they're all just using agents all day. They're very rarely actually looking at the code itself. So to your point, it's like, no, people are not writing the syntax of the code. I feel like this has been true for some time. I don't know how Exit has always been very much on the forefront of these kinds of tools, but we're purely agents. Agents are looking at the code. Agents do all the checks before the code goes into production. It's just rare to actually look at the syntax anymore. So in that sense, if you define software engineering as looking at the syntax and writing the actual language, and yes, it was dead maybe months ago, but at the same time, we're still hiring like crazy for software engineers. So how do you explain that difference? Well, I think software engineering has more becoming like software managing. Like no one's really. I mean, it's semantics, but like everyone is managing their agents and just. And. And using their judgment and conversations with customers to decide what to build, that's not going away anytime soon. So anytime, like the agent has to rub up against the physical world, whether it's customers, like feedback or like, where is the world going and building for that, like, then you need human judgment and we want to hire the smartest people to manage all these agents. So that's like one. One example, you know, one direction to take the discussion. But what else is interesting? Yeah, by the way, it also, I think you mentioned on someone mentioned all on podcast now it's a lot easier for someone else to come in and just immediately get started, which is. Which is cool. Now it's like, it's like, you know, humans can come and go and just all be operating on the soup of like, of like it's like, it's almost like you're operating in product space and so no one actually needs to know what like the underlying code is.

[A]

It's happening in venture as well. So when somebody joins our company they can say like you know, Josh or Cash might send us an update on, you know, how the business is doing every quarter or something. We set it up now that Lon or let's say an associate in training at our firm, we have a training firm for associates. They can say tell me the investment history and a summarize everything going on with Aragon or tax GPT and it will just bring it all back to them and it will search everybody's email, everybody's notion and the entire slack. So it'll go back to when these guys were in the accelerator and what their presentations were, what notes I gave them on their presentations, what venture firms we invested in. So it's kind of like that will all the institutional knowledge is now freely available in the corporate Oracle and that's the same thing in code, right? Or the sales team. The sales team. Now it's like you just ask Claude Tag. I don't know if you guys have set that up. Tell me the history of LinkedIn advertising or Squarespace advertising with us. And it's like yeah, Squarespace was on for these five years. The cto, the chief Marketing Officer left, they took a two year break, now they're back, they're advertising this and here's the codes they used and these are the products and the promotions they did over the years. It's kind of mind blowing for anybody to be able to come into an organization and be up to speed.

[C]

Just quickly on that note, like it's really important to have all. If the model has all the context that you have, then sure, it arguably might have really good judgment. But the problem is there's so much context that's missing. Like any conversation you have in the office that's not recorded, if it doesn't have that crucial information, just going to have really bad judgment. So for example, we tried to get the frontier models to do our Q3 planning and they were terrible. Why is that? Because they're just missing all the little nuances and even emotional states of people. As I walk around the office, there's so much context in an organization that's not just like what's recorded in your notes. So I think until we somehow fully give these models context, there will still always be a role for human judgment.

[E]

Just two things. So first, like on the. Is software dead? I think it's like good, you know, JCal to define this. So if something's not growing, in my view, it's dead. And when you look at the SaaS era, CRMs, ERPs, all of these kind of business apps, the thing that will remain is the data that they have. So actually the underlying DBs I think are really, really, really valuable assets here. But as far as humans interacting, logging in, taking actions like getting more licenses, all of this stuff, it is so dead in my view over there. And second, I think as far as the actions that will start to happen, I was just talking with someone that is about to raise a bunch of money for this idea to actually train small models. CRM model, an ERP model, so on and so on and so forth. And how do you take all of the context from the underlying DBS and put it actually even into the weights of a small model? This stuff is also pretty interesting as well. So where all this data goes, whether to will to your point, to Jacob's point, whether it's this memory, whether it goes into these small company models, this is tbd. But I think the fact that a company would need to ever spend tens of millions of dollars on some enterprise piece of just software, I think is, is, is, is over. It's not a question of if, I think it's just a question of when

[A]

I'm going to agree and say and then what are you paying a software company for? Yeah, paying for them not to quit making and supporting the software. And that's worth paying for. And I saw this in startups where people would be like, oh my God, you know this, I'm going to make the 17th meditation app, I'm going to make the 27th personal CRM system. And it was like yeah, you could raise 500k, make it and are you going to be here in four years? Because some business is going to be dependent on that product or some individual is going to be dependent so longevity. And the founders promise that they will not quit. Like say what you want about the valuation of SpaceX or Tesla, is there anybody in their right mind who thinks Elon's going to suddenly quit trying to get to Mars or he's suddenly going to quit solar and batteries or suddenly going to quit self driving like those projects he's been late on but never wrong and so that's why people buy those stocks and that's why they pay 30 or 40 or 50 times revenue for them. Same thing with Palantir. Do you get the sense that Alex Karp is ever going to leave that company and stop jumping up and down on his chair and speaking faster than me and being able to run circles around me and the other besties at the all in Summit and not let us speak? No. Those people are lunatic committed founders. That's the new moat, is having a founder who is so deranged. Will cash Josh sound familiar? And this isn't mentoring advice, but are you so deranged that your customers believe you're never quitting, you're never leaving? It's like the wolf of Wall Street. I'm not leaving. That's the moat. I'm not leaving. The show goes on. We're going to Mars. We're going to solve tax. That's what people want. They want a driven, absurd founder, and that builds the moat in your business. Every founder, every vc. Oh, what's the moat? You know, anybody can write code. The mode is you. You're the moat. Will you quit? And I can tell when founders are gonna quit. I can tell because then they start asking for, like a crazy comp package. They want to work on a side hustle. They want to start an incubator or be a scout. Like, when I start seeing that and founders like, hey, I'm a scout and I'm thinking of starting a micro fund and a podcast. I'm like, is the podcast about tax? Yes. Cash do it, or is the podcast about you trying to recreate all in or whatever? Like, let's stay focused on the mission. How obsessed are you with the problem, the product, the market, the solution that you're building? That's the new mode. Thanks for coming to me.

[D]

Exactly. I mean, on a side note, on this software thing, I think what an incredible job Claude OpenAI and all of these coding tools has done, Cursor has done to make everyone dream. We went to cello shop and the celloist is building their own CRM to manage their customers better. An Uber driver is building something my CRO and his wife is a psychiatrist and therapist, and she built a bot that goes into other psychiatrist videos and give her inside topics to create videos about.

[A]

Amazing. So she created a research assistant, a chief of staff that would normally cost $100,000 that doesn't sleep. Yeah.

[D]

And so it's such an incredible explosion of ideas. People can dream and build. So I think, I mean, just a year or a year and a half ago, the dominant conversation was oh, what is the TAM of coding agents and, you know, generation of code? Nobody's questioning that. That they put a software engineer in everyone's pocket, and our goal is to put a tax advisory in everyone's office, in everyone's pocket.

[B]

Done.

[A]

I mean, you know what the TAM of software development is the population of planet Earth, which is the population of food, electricity, smartphones, and now software, like everybody wants that. Yes, there are some people who don't have it right now. You know, they're still working on food, clean water, et cetera. But the number of people living in abject poverty is dwindling, and the number of people connected to the Internet on high speed with access to these tools is skyrocketing. There's reason to be optimistic. I'm sorry if we're getting a little crazy, but, you know, I just did. I don't know what you guys are on, but this is not a commercial for Breakers. Coffee, four milligrams, but, boy, does this get me active. What are you. Oh, you're on coffee.

[B]

We should reach out.

[A]

Will's on coffee. I'm on four.

[B]

If they want to come on board as a sponsor.

[A]

All right, cash is on. Zyn. He's got a. He's got. You know, I got to tell you, as your investor here and as your guide on the side here, I'm going to tell you, you got to get rid of the vape, okay? No vapey. I don't want any popcorn. Lun, if you're going to do nicotine, my professional opinion is do it with that. So you got the Zen. What do you want, Josh? You want to get.

[E]

You want to give it Celsius over here.

[A]

Okay, Celsius. Very nice. Blood type Celsius. Oh, plus a little sesh. All right, so we. Look at that. We got Lucy, sesh, and Zinn fired up after that. Now we're fired up. We all have to go back to work.

[D]

I have a funny story about the vapes are banned in San Francisco. So I had a guy, so I went to the shop.

[A]

Wait, what do you mean banned? Like, you can't buy them?

[D]

You can't buy vapes? No. Yeah.

[A]

What vapes are banned in California or just San Francisco?

[D]

San Francisco, particularly. So I went to go to San Mateo.

[A]

You go to Millbra?

[D]

Last time I bought it from Monterey. I mean, we were on trip at that area, so I bought it from Monterey. So funniest thing happened. I went into the shop, and they said they banned it. And the city, the official. When they Came in, they had the pictures of someone vaping outside of their store.

[A]

I was like, all right, cool.

[D]

And I didn't have the Zen either. So I went, came out and someone offered me to buy.

[A]

It's a good owner right outside the.

[D]

This is the most San Francisco thing ever. You can't buy a vape. I was a journalist, Jason, as you might know. And I used to smoke two pack of cigarettes every day. And seven, six, seven years ago, it helped me quit juul and all of that. And this common sense regulations, like it's my choice. I am choosing the less of lesser of the two evils. You can buy the cigarettes, you can get offered, you know, like just the Market street in San Francisco. But we're gonna, we're gonna ban the vapes and the Zin. I think. Don't hold me on this, but I, I, my understanding is that you can't buy flavored Zen either in San Francisco.

[A]

This is a very unique San Francisco moment. It's crazy. The funniest also, I mean, I just, I got like three great tweets and clips from this episode. Another amazing episode of this week in AI Go to this week in AI AI. Just sign up for our newsletter. In about two weeks, we're launching a new paid product. Every week we're going to profile two AI startups that I pick and my team, Deep Researches. We write a deal memo, we do a mini interview, and you're going to get that. It's going to be 300 bucks a year. We haven't started it yet, but if you're interested, sign up for the email now and get the free product. We'll put the first couple of profiles up for free, I think. But that's a new product that's going to pair with what we're doing here on the show, which is if you're a venture capitalist or you're a business development person, you know, you're doing M A, you want to buy companies, you're going to want to go to this week in AI AI to get those hundred profiles a year you guys are hiring Will, where should people go find out about an incredible career?

[D]

Yeah.

[C]

Or just DM me on Twitter.

[A]

All right, there it is. Josh. Where should people go if they're interested in joining you and your company careers?

[E]

Or email me josharagon AI with anything exceptional that you've done? No resume needed.

[A]

No resume needed. Yeah. Okay, Cash, what do you got?

[D]

Go to taxcube.com careers. We're hiring for incredible exceptional sales hunters and customer success managers.

[A]

Is that remote or in the office? Will you take somebody in Arizona or in Toronto?

[D]

If they're a killer, anyone exceptionally anywhere in North America will take them.

[A]

Okay, there it goes. You might want to look in Tokyo too. There's some great expats living there.

[D]

I mean, the Biscuit, sorry to get the base case is you make at minimum 200k.

[A]

And plus if you, I mean, listen in San Francisco. Hard to get salespeople. Too much competition. What I learned from Jeremy at Yelp, they moved to Arizona. I had this other startup. This was one of the best sales hacks I ever had, ever heard from a sales startup. They were like, we're going to start a sales office. So they had their team make a list of like what they thought were the five best places to do it. One of them was Arizona. So I'm going to look up on my team, like where Yelp's sales office is. They put out a bunch of job descriptions. They did a bunch of interviews. They went on a tour for 10 days over two weeks. Two days in Phoenix, Arizona, two days in Salt Lake City, two days in bowl, one day in Boulder, two days in Austin, one day in Dallas, whatever it was at Nashville. And they just set up interviews for these like very high paying jobs. And they were just like, yeah, you know, we went to Salt Lake City and so many people showed up. We got all these Mormons. These people were like knocking on doors to tell people they need to convert to Christ. And I was like. And they just person had this epiphany like, if they can do that, they can sell my CRM system or whatever sales automation software it was. And after that test, they decided to just hire like five people there. And then they looked at where they lived and they just picked an office equidistant from those five people that was the most equidistant popped up, a sales office. Anyway, I thought that was one of the great tactical, practical hacks I'd ever seen. And that's what great entrepreneurs do. Will, we gotta talk. I mean, how am I gonna get on this, Captain? I gotta figure out how to go in a time machine and go back and make up for for lost time.

[B]

We got to make it three for three. Today's panel, we only went two for three.

[D]

Absolutely.

[A]

We'll see you all next time on this week in AI.

[C]

Bye.

[A]

Bye.

[B]

Bye.