A Startup Is Trying to Buy PayPal… Craziest Deal of 2026! | E2312

This Week In Startups is made possible by: MongoDB - MongoDB.com/ai Rippling - Rippling.ai/twist Agree.com - agree.com Today's show: Hustle Fund's Eric Bahn and Chapter One's Jeff Morris Jr. join our venture capital roundtable directly following the news breaking that Stripe wants to buy PayPal. The possible transaction highlights how little interest today's leading startups have in going public any time soon. With host Alex Wilhelm, Bahn and Morris dug into when startups should burn the boats à la Fin, physical AI as the next frontier, the power of Silicon Valley groupthink, and how startups are approaching the first days of widespread, company-specific AI evals! Guest links: Eric Bahn https://x.com/ericbahn Hustle Fund https://www.hustlefund.vc Jeff Morris Jr. https://x.com/jmj Chapter One https://chapterone.com Show links: Stripe wants to buy PayPal https://www.reuters.com/business/finance/stripe-advent-offer-buy-paypal-more-than-53-billion-sources-say-2026-07-15/ Flex https://www.flex.one/ Webflow https://webflow.com/ Oak https://www.oak.id/ Fin selling to Salesforce https://www.salesforce.com/news/press-releases/2026/06/15/salesforce-signs-definitive-agreement-to-acquire-fin/ Erebor https://erebor.bank/ "Competition is for losers" https://www.wsj.com/articles/peter-thiel-competition-is-for-losers-1410535536 Chapter One piece on the value of software https://jamesin.substack.com/p/some-areas-weve-been-investing-in Timestamps: 0:00 Eric Bahn (Hustle Fund) & Jeff Morris Jr. (Chapter One) join the show! 1:35 Stripe's reported $53B offer for PayPal 6:29 Why founders don't want to go public 10:08 Agree.com - Stop chasing invoices and automate your entire contract-to-cash stack. Go to https://agree.com and tell them Jason sent you to get 50% off for life! 15:16 When is it time to burn the boats? 20:12 MongoDB - AI-assisted and agentic coding is helping you build faster than ever. Start building at https://MongoDB.com/ai 21:16 Does VC need an AI revolution? 30:50 Rippling - Thanks to our partners at Rippling! Head to https://Rippling.ai/twist and get the only AI built to give you full visibility across your startup and take complex actions across your entire business. 32:36 Custom evals, open source models, and avoiding vendor lock-in 36:37 Physical AI and robotics as the next frontier 37:42 Oak's $60M seed and startup clusters chasing agent infrastructure 40:43 Escaping the Silicon Valley "think tank" 48:28 Consumer AI's missing wave 1:00:18 AI regulation and how to get it right 1:12:01 Where to find our guests and wrap-up! Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.com Check out the TWIST500: https://www.twist500.com Subscribe to This Week in Startups on Apple: https://rb.gy/v19fcp Follow Lon: X: https://x.com/lons Follow Alex: X: https://x.com/alex LinkedIn: ⁠https://www.linkedin.com/in/alexwilhelm Follow Jason: X: https://twitter.com/Jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis

• 0:00Speaker 00

  Hey everybody, welcome back to Twist. My name is Alex. Today is July 15th, 2026, and that means it's a Wednesday, and that means it's time for yet another venture capital roundtable. And we're recording this right after news broke that Stripe wants to buy PayPal. We're also going to talk about AI's next bottleneck, how startups can build defensible software, and more. But to help me grok the market, I have brought along two crackerjack VCs, and they are Eric Bond, one of the founders of Hustle Fund, a super early venture capital firm, currently investing out of its fourth fund. Hustle Fund is backed companies like Webflow, Agree.com, Boom, Charter Space, and others. Eric, welcome to the show.

• 0:37Speaker 01

  Thank you so much, Alex. Happy to be here.

• 0:38Speaker 00

  We also have Jeff Morris Jr. He's the founder of Chapter One, investing out of its third fund. Chapter One has backed companies like Superbase, Mercury, Flex, and Metify, a company that I also love. Jeff, welcome to the show.

• 0:50Speaker 02

  Great to be here, Alex. Thanks for having me.

• 0:52Speaker 00

  This week in startups is brought to you by MongoDB. AI-assisted and agentic coding is helping you build faster than ever. Start building at mongodb.com/ai. Rippling. Don't settle for AI, that's all talk. Head to rippling.ai/twist and get the only AI built to give you full visibility across your business and take complex actions across your entire organization. That's R-I-P-P-L-I-N-G dot AI slash twist. Sign up for exclusive access today. And Agree.com. Stop chasing invoices and automate your entire contract to cash stack. Go to agree.com and tell them Jason sent you to get 50% off for life. We'll get to all the startup stuff and founder stuff in a minute, but I want to start with

• 1:36Speaker 00

  this enormous possible transaction because it blows my mind. The news is that Stripe, a private company, may purchase PayPal, a public company, for more than $50 billion. So just first reactions, Jeff, let's start with you. Did this deal surprise you? Do you think it's smart? What was your kind of first page analysis?

• 1:52Speaker 02

  Yeah, I think it did surprise me. The, you know, when you picture PayPal and you hear the name, you just think they're not acquirable, right? And you have this private company Stripe who really is in world building mode and wants to, you know, as a private company, buy this iconic payments platform. And it's just, you know, I think it's just a headline that's very surprising. It shows, I think, some of the benefits of staying private, though, because Stripe can do these things without the scrutiny of the public markets and presumably has a balance sheet to make it happen. And so, yeah, it is surprising. I think it's, you know, it's like one of those things you hear about the PayPal mafia and you think, think of this iconic company and now it's, honestly, it makes me feel

• 2:34Speaker 02

  old. It's my first reaction. And, you know, it's just a sign of the times. Everybody's out to world building. M and A is obviously a great way to do that.

• 2:45Speaker 00

  I was going back through very early Stripe coverage, and I think when they were worth like $20 million, both Elon Musk and Peter Thiel, two of the best known PayPal mafia members, also backed it. So this is actually kind of revenge of the same team, if you will, Jeff. But your point about, you know, the flexibility of the private markets allowing you to do more stuff, that to me implies that you think that if Stripe had been public when they executed this transaction, and it's not confirmed yet, et cetera, et cetera, that it would have been poorly received by the markets?

• 3:13Speaker 02

  I wouldn't say it'd be poorly received. I think there's just more considerations as a public company when you do anything M and A related. And, you know, Stripe, Stripe being a private company, just has flexibility to take a longer term point of view on what this acquisition might mean for their company going forward without impacting their, you know, public market, market cap in the near term. And so I think when you're private, you can just do things that are a bit more like YOLO. And this is definitely, for me, a sign of the times.

• 3:45Speaker 00

  Yeah. All right, Eric, I know you and your firm invest incredibly early. So you're the person that we should always talk to when it comes to super late-stage unicorns buying public companies. But I'm really curious what your thoughts were about this. PayPal's share price peaked above 300 back in the 2001, 2022 era, and they're going to go private maybe for 16 and a half bucks a share, kind of a shocking decline in, in worth.

• 4:07Speaker 01

  You know, I've been watching a lot of Hamilton the musical with my kids, and King George is my favorite character. And his, you know, song always has that chorus, which is like, oceans rise, empires fall. So, you know, PayPal is just one of those canonical brands out of the dot com 1.0 era, really. And it's in some ways unsurprising. This is just how the nature, nature works in Silicon Valley. You know, you see these incredible rising stars, and then, you know, at some point they become white stars and sort of fizzle out and so forth. Then, you know, this actually wasn't really surprising news. I figured that PayPal was going to be a target at some point because it seemed like they'd been stagnating over the last couple of years.

• 4:45Speaker 00

  Yep.

• 4:45Speaker 01

  The Stripe thing is interesting as a private company doing this transaction, or at least intending to do this transaction. And the, my suspicion is that PayPal kind of calcified in its own, within itself, right? So like, you kind of find these company arcs where they're startup-y, they do cool things, they innovate, they break things, and then they professionalize, you know, a bunch of MBAs start to join, and then it starts to become about the employees trying to figure out how to get promoted for themselves. Innovation kind of stops. And they know that they have a really good brand and asset, for sure. But my sense is the culture is totally broken. So maybe Stripe can do something really fun here with this asset.

• 5:23Speaker 00

  Yeah, I was reading their last earnings call, which is the transcript because I'm lazy and don't like audio. The CEO, the installed the new CEO, I think it was in February, and he was like, there are, and I'm going to air quote this, it's close enough, like, places where we can make large cuts in the company. So I think on the point of being kind of calcified internally, that makes a lot of sense. Eric, yeah, I'll stick with you, Eric. Do you think that there's any way to build a large technology company that doesn't end up in kind of Yahoo PayPal territory, given enough time?

• 5:51Speaker 01

  Microsoft kind of did it, right? They've been around since like the 1970s, and they, you know, had their ups and downs, but they seem to be, seem to be generally trending in the right direction. And I suppose like an Oracle as well. I think it's been, it's been done. You know, there have been enterprises even beyond tech that are sort of innovative, I guess, that have lasted a lot longer, maybe like a GE and so forth. So yeah, certainly. But at least within Silicon Valley, you know, the examples are admittedly quite few. So, you know, I'm struggling to find more than five or 10.

• 6:21Speaker 00

  And then Jeff, on the, on the take private aspect of this, you know, it does seem that every time I talk to a founder, they just have no interest in being public because they don't see any upside to it. Do you think that seeing a, one of the latest stage, highest value private unicorns take a public company out is just kind of a sign of the times? Or is this more of a starting gun to see more of this kind of a transaction happen?

• 6:41Speaker 02

  I mean, I think it's a, there are two realities in Silicon Valley. One is you have the SpaceX IPO and the rush of IPOs that will soon follow. You know, obviously OpenAI, Anthropic, et cetera, which will create, I think, a new class of entrepreneurs who really does want to go public and give employees, for different reasons, you know, in those cases, maybe there's not private capital available to, to fund their, their infrastructure projects. But I think, I think there's actually going to be more companies over the next year or two that view going public as being a great thing. Whether those stock prices hold up, you know, SpaceX is now trading today below their IPO price for the first time since the IPO.

• 7:25Speaker 00

  Yeah.

• 7:25Speaker 02

  And so going public doesn't mean that you're just going to have, you know, an easy path going forward. You still have to, to perform. I do think there's been a kind of fear of going public over the past two or three years. And, you know, Stripe is famous for being one of the companies that really does want to stay private. And so for a long time, Stripe has been one of the cultural, kind of like North Stars within Silicon Valley for how, you know, like best practices for, for how a company should, should, should see the future. So it's, you know, I think, I think it's going to be a mixed bag. There are going to be, there's a ton of targets on the public markets, especially

• 8:06Speaker 02

  within SaaS land that are going to struggle. And whether they like it or not, and even on the consumer side, like you hear rumors of Snap going private again, right? And so it's, it'll be very interesting. And, and we'll see what, what that means for a lot of companies in, you know, in the next year or two.

• 8:23Speaker 01

  I have a question for like both of you on this too. So like, what does it even mean to be private at this stage? I mean, so I can go into my, my Solium account. I have a little bit of Stripe shares. And actually, there are places for me to trade, you know, and transact and get liquidity from, from my equity at the, at the same time. And even things like Anthropic and OpenAI, they're producing some sort of vague quarterly reporting that's actually affecting public equities whenever they, they announce, right? So it, this line seems to be getting so blurry. So if there's liquidity available and actually, I guess, like some public sense of, of performance, then I don't know what, what the benefit is becoming for public companies, maybe outside of like some fundraising upstream.

• 9:08Speaker 00

  Well, on that point, we saw the, the blow up between USVC and Androl now a couple weeks back when they said, hey, you know, we got Series H shares. And Androl said no, and they said, well, yes. And then there's big back and forth about who was to blame for that. And that to me is just an indication of why the private markets are different. I mean, they, they don't have the same part of transparency in terms of data. It's harder to get price discovery because the market's less liquid, so you have less price efficiency. So to me, like a bunch of reasons, Eric, are why they're bad. And if you look at the most valuable companies today, they, they all grew, you know, their last 99% while public. And so to me, it's just, it just, it feels weird to see adolescence in the corporate world extend forever as venture capital firms get larger. And also, I can't benefit in my, my index funds as easily. So I think there's still a reasonable argument

• 9:51Speaker 00

  to be made for going public. But, you know, I think, I think Jeff's right, you know, I think Stripe set this North Star and everyone's emulating them because who doesn't want to be a Colson brother, you know? Does anything break this, this trend? Like, what could happen, Jeff, that would actually get people to want to list again? Like, is there any like prestige to it?

• 10:08Speaker 03

  Our show is filled with helpful and practical advice for founders, but there's another reason to become a regular Twist listener. Amazing deals on essential products to help you run your company more effectively. So I'm super excited to announce our new deal with Agree.com, the all-in-one contract to cash flow workflow. Go right now to Agree.com, sign up, and tell them you're a J Cal listener, and they'll give you 50% off for life. Agree is the number one fastest way to go from contract to cash. That means gathering these signatures, invoicing, billing, payments, and revenue recovery. No more jumping between four or five different platforms, just to write out a contract, get it signed, then

• 10:48Speaker 03

  set up billing, and start sending out invoices. 99.64% of all invoices on the entire platform are paid within just 10 days. So if you want to stop chasing invoices, go right now to Agree.com, and if you tell them Jason sent you, you're going to get 50% off for life.

• 11:06Speaker 02

  I mean, I think the capital markets might force that for some companies, just in terms of the private markets being exhausted and unable to finance those really expensive fundraising rounds. The, Eric made a great point, though, which is

• 11:23Speaker 02

  The secondary markets really do blur the lines between what's a public and private company. And you know, as Eric was saying that I was thinking as a CEO, maybe you want to become public just to get all the secondary noise out of your life. Because that's managing the secondary transactions. If you're a private company CEO is a huge pain in the butt. And if you're anything in any critical industry company, doing anything, you know, selling to the government or anything similar, the you actually have to control the secondary markets in a really efficient way, or that can come back to bite you on the contracting side when you're selling to the government.

• 12:04Speaker 02

  So there's a lot of foreign investors in countries that, you know, these founders would prefer not to have on their cap tables who are desperate to get access to those names. So just controlling the secondary markets is a huge, huge challenge for these founders.

• 12:25Speaker 01

  Excellent point, too. I mean, like you could have like a Chinese oligarch with a Singaporean entity trying to invest in like a defense company in the United States, right? So all these like sort of shell company things. So I'm sure like if the US government, they're like, why is Androl taking this money, right?

• 12:41Speaker 00

  And Androl's like, we're trying not to. We're staying private and we're exercising pretty strict control over their equity. What percentage of founders of companies that are later stage are actually okay with having their secondary shares trade relatively freely? Or is the Androl position here the common point for most founders?

• 12:58Speaker 02

  I think the Androl position is going to become the more common point because you're going to see more and more stories of investors who are either from, who aren't welcome on the cap table, getting access to that, that name, or you know, I think you'll just have founders who, who want to just control this process because it's become, I don't know, you've seen the SPV stories now where people just disappear, like the manager of SPV suddenly doesn't respond to emails, they can't be found. And I think on the investor protection side, you're going to have a requirement to, you know, on the SEC side or something similar to come in and clean this part of the market up because it's so.

• 13:42Speaker 00

  I think crypto rug pulled to SPV managers could flee. I think that was really the, the order of operations there. You gotta set some precedence. Now, just before we hopped on, we were talking about Flex, a company that you guys have both backed that just announced a series B, one round, 70 million, 1.2 billion dollar valuation, clearly in the fintech space. This is a major piece of fintech M and A. Do you think it's going to lead to increased interest from incumbent firms? We just saw bank earnings come in really strong to buy what we might call the middle class of fintech startups, Eric?

• 14:15Speaker 01

  Oh, that's a very interesting take. I don't know, like, you know, I think the eye on a lot of investors, I guess everyone's like just focusing right now on just like pure AI companies right now and like gross age related. And so fintech has always been this category that I think in the last like two or three years has been slightly overlooked, but it's one of those, I think, going concerns, there's models that have like really clear going concerns of how money comes in and goes out. So consolidation, I think, is always something that's happening in the finance world, like, you know, with these big banks. I never really considered whether, you know, there's like a harvesting strategy taking place for modern fintechs, because I always felt like that's happening more on like regional

• 14:59Speaker 01

  banking, you know, some of these like a kind of classic rollups that are taking place. But I think it's a pretty decent hypothesis. I just haven't really put much thought on it.

• 15:07Speaker 00

  Well, I'm hoping it's the case because it'd be really fun to talk about some stories that are not just AI, because I write a newsletter and I'm just always like, good morning, welcome to the speaking Alex thinking about stuff. It's all AI again. And that gets kind of boring. All right, let's talk about, let's talk about AI. So one of the most interesting deals we've seen, I think in the last couple of quarters, was the exit of Finn to Salesforce, Finn previously Intercom. They famously kind of burned the boats and pivoted the company towards agents and renamed themselves after their agent. And it worked out pretty well. They're an example of a SaaS unicorn that was struggling, finding its footing in the AI era and having a pretty solid exit to a major company, just kind of a win for everybody. Eric, I was going through Webflow's history and I know that the company went through a pretty big shakeup.

• 15:51Speaker 00

  I think it was this May, cut some of the staff and really kind of re-architected where they see the future going. So I'm curious about like, when should a startup founder know that it's time to kind of like let go of the past and be willing to set fire to kind of build values they can scale up to kind of the growth expectations we now see in the current era?

• 16:12Speaker 01

  Yeah, I've been tracking Webflow for a long time. In fact, it was the very first angel investment I ever made many years ago. So Vlad is a good friend from 2007, when he started this company. And they really nailed it during the Web 2.0 rise with just how they architected this like software within the browser. It was kind of revolutionary and you know, had all this wonderful control. And what they're finding themselves is kind of rug pulled by this AI era where there's a different kind of paradigm here now in terms of how people want to design, they don't want to learn how to like put CSS together and so forth. So it was really painful to watch that one because I think that the culture of Webflow during its peak was really, really cool, but it did get kind of bloated.

• 16:55Speaker 01

  And per what we discussed earlier about PayPal's cultural ossification or whatever classification, they were experiencing something similar here, too. So it was somewhat of a brave call and one yet to be seen as successful that they had to do some pretty major cuts to try and make this company lean again, so they can actually start to feel a little bit more of that startup-y vibe once again. So I think it was necessary for them to go through this, but again, to be, to be said, whether they catch up to some of their competitors at this point.

• 17:28Speaker 00

  Yeah, when I was working at Crunchbase from pre-series B through after our series C, I was amazed at how much the company grew in terms of staffing and then how much process got built, almost like automatically or naturally as it scaled up in headcount. And I don't mean to be a cynic here, and I don't mean to be an AI doomer, because I'm not, but it does seem that whenever we see inefficiencies, whenever we see cultural ossification, to use your phrasing, it's the humans that are the problem. And so are we just kind of moving towards a world in which it's almost like the fewest, like the highest revenue per employee is going to be the most efficient and least slow company, as in humans should only be added when they're like so painfully necessary? Because otherwise you're going to end up with everyone going to meetings about meetings? That's just, that, that is where this seems to be going, Eric.

• 18:14Speaker 01

  I think so. I mean, you know, the vision that I'm currently subscribing to is that all of us become some form of individual contributor as a key component of our jobs that's orchestrating all these agents in our work. And that's kind of the life I'm trying to build for myself. And the really neat thing about that is, you know, when I have command over, I guess, all these agents, you know, I can, I can feel a little bit startup-y in terms of the work, you know, just, just because I have so much capacity to try different things or experiment, even while I'm sleeping. So I hope so, but I don't know. I mean, like, I feel like the ossification timeline has been a recurring thing since like modern white collar work has, has happened. So maybe we get to a new normal for how much throughput we're expected to produce and then

• 18:58Speaker 01

  it ossifies in a different kind of form.

• 19:00Speaker 00

  So that's dead on. Talking to my dad about his early professional days, they had like a typing pool and they'd all these people that carry documents around the office to their next station. And today, like, we would think that's absolutely insane because everyone's down charged with their own email and scheduling and so forth. And so I think we've already become more IC-ish, but maybe this is the next iteration of this. But Jeff, when Eric was talking about becoming more startup-y, you were nodding your heads. I want to get you to weigh in here.

• 19:23Speaker 02

  I was just thinking, we talk about burning the boats on the company level, but I think every employee at each, every company needs to burn the boats on what they think their job is and what they're going to be doing, you know, going forward. And it kind of surprises me, I think there's, there hasn't been as much of a, like a, just a rush to learn all the new tooling amongst my peers as I thought there, there might be. I'm obviously a VC now, so I think our jobs are relatively different than working at a startup, but you know, I think there's a chance, it's like pretty easy to become AI native within a company even today, where you, you just need to be like the one who's

• 20:04Speaker 02

  most interested in AI amongst your peer group, which is actually relatively easy within most orgs.

• 20:12Speaker 03

  The founders I'm talking to are shipping code faster than ever, thanks to AI. But one big important question remains, if every feature you're writing ship has to slow down for an overwhelmed database, what's the point? But fear not, the solution is MongoDB. Instead of wrestling with rigid schemas and painfully reworking antiquated data structures and formats, MongoDB's native data model perfectly mirrors the language LLMs are already speaking. MongoDB gives you the flexibility to ship at the speed of AI. The acid compliance guarantees you actually get to sleep at night while it scales to handle massive Fortune 500 workloads. And the best part is developers swear by it, literally. I can't use the actual words they said in this ad, so let's just call it a really great database.

• 20:57Speaker 03

  Start building at MongoDB.com slash AI.

• 21:00Speaker 02

  Like, I kind of think it is on the individual level is like, yes, come easy to burn their own boats, but so do, so does every single employee in Silicon Valley at this point. And, and that's, that's kind of what I was nodding my head at.

• 21:11Speaker 00

  Yeah. Do you think that VC needs to go through a similar evolution slash revolution? Because I know that some companies, some firms, I should say, having more data focused than, you know, hand sourcing focused over time, but there's always seemed to be the kind of the, the edge of venture versus kind of the core of it.

• 21:26Speaker 02

  I do, yeah, I absolutely do. And I think the newer firms, it's very similar to like the PayPal Stripe conversation where you have the opportunity for your newer firm to really shift your strategy and become, you know, I think by necessity, most emerging managers run pretty lean firms. And so to, we've over the past, you know, however many years, become more software and data oriented just to compete. I guess there, there hasn't been a choice. It's like, you have to do this. And now, you know, seeing the bigger firms, like the big platforms, like the question is like, how many people do they really need to be efficient? And is, is, you know, having a

• 22:11Speaker 02

  A 500-person org, the right solution. And I think there's just going to be a barbell adventure, just like there is in in the private markets on the company side.

• 22:19Speaker 00

  Are you trying to imply that very large platform VCs that are multi-stage, multi-focus, multi-adventure might have some cultural ossification of their own? Because that would be a pretty reasonable point, I think. Yeah. Okay, I'll just, you guys have to be nice. I don't. When Andreessen announced their latest batch of new media partners, I was a little bit perplexed. It seemed to be a little bit afield of the overall mission. But if you have a lot of these, I guess you can afford a lot of side quests. Eric, your firm is seven people, according to the website. How much automation have you guys put into place to keep that number of people relatively small? Because it's just a couple of partners and a little bit of finance, it looks like.

• 22:56Speaker 01

  Yeah, I mean, I'll admit that there's actually a bit more. So seven full-time, but 22 total. And the majority of our team are contractors working on media and network and events, but only four investors. So we are building lots of software. If you talk to my co-founders, Elizabeth and Sheen in particular, they are ridiculously sleep deprived because they've been rushing with their Fable Access to ship tons of code and are, they're producing code every day. So for, you know, I think Jeff is making an amazing point, which is for earlier-stage, kind of smaller AUM funds that are resource constrained, Scarcity has a wonderful way of forcing innovation and doing more with what you got, right?

• 23:41Speaker 01

  And right now, the best hammer that we have in our hands is like these are these AI tools where we could just like constantly vibe code, things that we can do with our data outreach, that's a little bit more automated, even a little bit of deal assessment as well. So it's a fun place to be. There's a reason why I'm actually in my garage right now is because I can't fucking afford like a nicer office. But yeah, it's a, it's a cool place to be. And I'll just make another comment too about something that we're kind of dodging is like, it's amazing how ass-backwards the VC industry is right now, in the sense that most people, as Jeff is saying, are not actually using these tools in a big way, right? Like, I think the majority of VCs have not touched Claude Code once.

• 24:22Speaker 00

  Wait, no, no, no, no, that's impossible.

• 24:25Speaker 01

  No, I really do think so. I mean, they may have opened it and put like, how do I like make a brisket or something like that? But I think they're not actually committing anything to like GitHub or Vercell or anything like, like they're not actually producing real software. And it is crazy how much pen and paper I'm still seeing during these meetings with other VCs, you know, and versus, and I'm not sure how any of this is getting transcribed into like, you know, institutional knowledge and data and so forth.

• 24:50Speaker 00

  Surely the people telling me that everyone needs to be AI-native and move twice as fast, those are the people who are leading the charge in their in-person interactions, right? They wouldn't be talking to the boss without walking the walk.

• 24:59Speaker 01

  My boomer parents, like my dad was a smoker and he was a doctor. She was also like, you should probably shouldn't smoke, you know? And, you know, like, say what, like, do as I say, not as they do, kind of situation.

• 25:10Speaker 00

  I don't even mean to poke fun. It just, it surprises me that the people who are often writing very large checks into technology companies haven't, it's like not test driving a car before you buy Ford. It just feels a little bit, do you, do they have people that tell them what, what's cool and what's not? Is that like a, like a job in VC now? Like to be like a whisperer or kind of a manuensis to an elderly VC who doesn't want to learn? Is that what, is that the renewed venture capital job path? Because I think that would be a good one.

• 25:38Speaker 01

  Jeff, you should take that.

• 25:41Speaker 02

  Yeah, I think to Eric's point, actually amazes me because using the software and the tools is actually a great way to win the deal. And so that's happened to us. We like Betteroff, which we invested in last year.

• 25:53Speaker 00

  Yeah.

• 25:54Speaker 02

  We just built a simple web app and user authentication and went to the second meeting, showed the founder what we had built with their infrastructure and suddenly you're on the cap table. Like it's, it's like the, and by the way, that takes with Claude Code like five minutes now?

• 26:10Speaker 00

  Yeah. It's not like that was a six-month project.

• 26:13Speaker 02

  No, it literally took us five or 10 minutes and the founder is, he was amazed, right? And, and so it's like, it's almost faster than reading the deck to actually use the product. And I don't know, I don't know why that is. I think there's a rat race within venture where you stack your calendar and you have like zero time to experiment or, or think. And that's just sort of like the, the culture that exists within most venture firms.

• 26:39Speaker 00

  About going and using AI technologies, one of the things that's come up quite a lot is what to do at the startup level regarding models and evals. And I don't want to beat to death again, the idea that yes, open source models are improving and the gap's closing to closed source and all that. But there's been some interesting commentary lately about how startups shouldn't just bring their own intelligence in-house, but also design their own evals around it because no one knows their company better than them. And that makes good sense to me, but it also seems to be technically tricky and demanding quite a lot of startups that may not have those kind of in-house AI chops. So Jeff, in your portfolio, how are you guiding companies to not only avoid vendor model lock-in, but also to ensure that they are doing the right evals for their use case

• 27:23Speaker 00

  and not just depending on some benchmarks that, you know, SpaceX AI or Anthropic put out?

• 27:28Speaker 02

  Yeah, I think this is a really new conversation. And quite frankly, like over the past two or three years, there hasn't been a ton of startups who focus on evals. Like it's really just about keeping up with your, the competition and showing revenue and growth. And so there's been, like, this is, this is like a more recent conversation, just frankly due to the number of startups who have gone absolutely destroyed by OpenAI and Anthropic. And so, you know, I can count maybe on like one hand, how many, how many teams in our portfolio that are like pre-Series A are, are building their own evals.

• 28:09Speaker 02

  It's a really small number of companies, but there is, I think, a new conversation around having more openness to using things like open source models, just because you don't want to, to be destroyed by giving your data to, to the larger foundation model companies. So, but I would say it's, it's a very small number of companies at this point who, who build their own, or do their own evals in-house.

• 28:35Speaker 00

  I want to get Eric on this, but Jeff, on the point you just made about, you know, seeding your data to the major AI labs, I went through every major AI labs data use policy and they all say, we don't trade on your data, we don't trade on your prompts, we don't trade on your outputs. But then Satya Nadella, the CEO of Microsoft, said that even how frequently you're doing tool calls can be information that is useful. So to me, this conversation is less about seeding your data to the ALS, but really the meta data? Is that correct? I just, I'm literally just trying to understand this whole problem better.

• 29:06Speaker 02

  That was exactly, it's the tool calls and being able to see even the customer spend within your platform. Like you, you can tell who's inflecting within, within your customer segments and it's pretty easy to, without ingesting their data for model training purposes, to, to kind of have directional data to, to where you should spend time internally.

• 29:31Speaker 00

  So what you're saying is that Stripe's going to become the world's best VC firm in time? That's what I just heard. All right.

• 29:36Speaker 01

  They're already pretty good.

• 29:37Speaker 00

  Well, they are in fact, because why, if you don't go public, you can do whatever you want. Eric, custom evals and how your portfolio companies are approaching this? I know that Hustlefund has like 65,000 portcos, so just maybe a kind of a pastiche, if you will, just blend it all together.

• 29:52Speaker 01

  Don't be hyperbolic. We only have 700 portfolio companies, right?

• 29:56Speaker 00

  I'm so sorry.

• 29:58Speaker 01

  No one gives a shit. At pre-seed, it's such a zero-to-one thing. Like the only thing that these companies are trying to do is get to product market fit and they'll use any model that's cheapest and available to them. So open source is becoming more of a common thing that they're discussing, just for the sheer cost of it. But I'm constantly being like begged for, like, you have any Anthropic credits or OpenAI credits or whatever? They'll, they'll use whatever they can, right? Now, I think this problem starts to become real after product market fit and you actually have like a true business and, and a real sense of like, we have to now create like boundaries and a moat, right? So I can, I can understand why, like at the enterprise level, this is like an acute issue. But at least in the pre-seed world, it's, it ain't no thing. No one's really talking about this in our portfolio.

• 30:43Speaker 00

  But you invest at the pre-seed level and the companies keep growing. So I presume you have some visibility into what the same cohort are doing at Series A and beyond.

• 30:50Speaker 03

  Most AI tools can give you business insights, but they can't actually do anything about it. That's the difference between AI that talks and AI that acts. But Rippling AI is built differently. It's the only AI built on your live global workforce data. So HR, IT and finance are connected from day one. That means it's not just surfacing insights, but you can actually take actions across your entire organization. Let's say you want to focus on talent retention, just ask Rippling AI, who are my top performers this year? You'll instantly get a workforce report, comp ratios, performance reviews, engagement metrics, all the data you need. Then Rippling AI will recommend a retention strategy, including a 10% spot bonus for your top performers.

• 31:34Speaker 03

  So don't settle for AI, that's all talk. Head to rippling.ai/twist and get the only AI built to give you full visibility across your startup and take complex actions across your entire organization. That's R-I-P-P-L-I-N-G dot AI slash twist. Sign up for exclusive access today.

• 31:54Speaker 01

  Yeah. Yeah, yeah. And barely having these conversations, but they are starting to happen, you know? So the, the way that I'm sort of trying to wrap my mind around this, and I think Alex, you're a Formula One fan, right?

• 32:04Speaker 00

  I am indeed, zim-zim.

• 32:06Speaker 01

  Yeah, exactly. So spa this weekend. So I kind of view it like this, which is, you know, all the cars right now, if you think of each car as different kinds of frontier models, are within like percentages of each other, right? So you got like your, your Red Bull, you got your Ferrari, whatever, Mercedes, and maybe someone's an Aston Martin in the back, right? And so, you know, they're all kind of close to each other, but what the, and that's fine and that's sort of publicly known, but what the teams don't want to share is like their pit stop strategies for the next race, like the racing line they're going to take, like tire management and so forth. And like, that's kind of how I'm viewing like these evals, which is, you know, once you get to that level of like high-end, like the pinnacle of motorsport, that's the stuff that really starts to make huge differences, I guess alongside of the car, which is probably the biggest one.

• 32:49Speaker 01

  And the, the companies that we're sort of dealing with at pre-seed at Hustlefund are still like in their entry level Miata spec racing thing where like, you know, none of this stuff is like that relevant. But it's starting to happen. You can start to, you can start to see around Series A too, just like whether we should use some of the capital that we have now to invest in more of a moat around like evals. And that's barely starting to happen.

• 33:14Speaker 00

  Okay, so we're hitting this one right when it begins, but going back to your F1 analogy. I think the argument about the need for custom eval is not really in a protective sense, but in a way to better understand how a model applies directly to your business case. It's not like we're just talking about their pit stop strategies. I think the cars are running on different tracks. And that's why I think the scores being 1% off between the leading models, sticking to the analogy, don't matter as much. Because if, if something is much better for your company, or you can tune it, post-train it, fine tune it, whatever, then you can move a lot faster. I guess I, I thought this was going to become more quickly a requirement to survive, given what Jeff said about the number of companies that OpenAI and Anthropic have either accidentally or purposely stepped on.

• 33:58Speaker 00

  And that seems to be happening pretty frequently. So to me, any defense, sounds like the right way to approach this, Jeff. But am I being alarmist?

• 34:05Speaker 02

  I think when you're an early-stage founder, as Eric said, you're, one, if you start a company, you view yourselves as being different than OpenAI and Anthropic for a variety of reasons. Like, there's nobody in, in like a accelerator cohort who is a, you know, AI-native application in their company who's starting a company and saying, hey, like Anthropic or OpenAI is going to beat us if they want to. And so there's a, like a naive, it's almost like this, like blissful naive approach to building a company that you have to have. And then, you know, secondly, I think, Eric, you can convince yourselves, like, when that does happen, we'll have enough scale or defensibility that they won't be able to compete with us.

• 34:49Speaker 02

  And so I don't think, like, the early-stage founder psyche just isn't geared towards worrying about this problem until it, until it actually is a problem, which is normally when you do have some product market fit.

• 35:02Speaker 00

  Okay. So it sounds like this is more of a late-stage point that it'll be curious to see, or interesting to see, how far down the startup age this goes in terms of what people do it. But I think we're going to need just better and easier tooling. I looked up, there are companies that will help you with, you know, internal AI evals, but they seemed much more enterprise-tuned. So maybe someone should build that for like, Series A companies or earlier. I think that'd be super interesting.

• 35:24Speaker 01

  Yeah, Jeff, you and I can vibe code that this weekend. It's a good idea.

• 35:28Speaker 00

  Let's talk about that for a second. You had a funny tweet, Eric, that was like, everyone in Silicon Valley staying up to the glow of their laptops until they run out of Fable 5, Fable 5 access.

• 35:37Speaker 00

  How much better has Fable been for you and Elizabeth compared to preceding models for vibe coding work? Because it seems about the same to me, but I'm not doing difficult work with it. So I'm not quite sure that I'm getting full bang for my token.

• 35:48Speaker 01

  It's great. I mean, like, the, the stuff that's doing right now is going across all of our codebases, refactoring things, showing like where we've been inefficient and so forth. And, you know, Opus was insanely good, too. But I think just the level of sophistication of just like, oh, you know, this is not very refined code, you know, here's, here's what we're noticing and things like this. It feels like the exact same kind of feedback that a senior engineer used to give me when I was working at some of these big tech companies. So this is AGI. I think it's actually happened within coding, at least in this domain. It's amazing.

• 36:19Speaker 00

  What's the next sector where we reach an AGI-ish level of performance?

• 36:23Speaker 01

  Jeff?

• 36:26Speaker 00

  I've never heard someone say pass before with such, such confidence. All right, Jeff, you tell me.

• 36:31Speaker 02

  I mean, I think it's going to take a while, but it seems like the model companies are shifting their focus towards physical AI. And in some ways, that's a recognition that they're pretty close to AGI on the coding front, or on, you know, kind of on these more digital native use cases. So I think, I think that will be interesting because those new domains, whether it's robotics or anything, physical AI, require you to spin up whole new teams. OpenAI already has a robotics team. And so I think that's, the next narrative is going to be, hey, we conquered coding, and now we're going to, you know, focus on the physical world.

• 37:08Speaker 00

  Okay. Well, we'll, we'll see. I'm, I'm very bullish on that. We had, we've had the CEOs of 1X and a lot of the humanoid robotic firms on, and some also some purpose-built robots like Dusty Robotics and so forth. And I, they, they all seem to be making such quick progress that it seems to me we're going to get to that being not solved per se, but to maybe a modern self-driving level of sophistication in the next couple of years. And that's going to be just super duper exciting. All right. I want to talk about startup clusters. A company called Oak, Oak.id, just announced 60 million in funding, and they came out of stealth, and they're working on essentially a shared identity layer for agents and humans, which a couple of companies are working on.

• 37:50Speaker 00

  And so one thing I'm seeing, Eric, it seems that there's a lot of capital chasing newly discovered problems that we run into, like, oh, now we have agents, we're going to need to integrate them into identity. Oh, we have a lot of agents now, we need to orchestrate them. And then like six to 10 companies get spun up to work on that. They raise money, they all seem to kind of go pretty quickly.

• 38:09Speaker 00

  In prior cycles, did we see this level of intra-startup competition for new possible problem spaces? Because it seems to be more crowded and faster than before.

• 38:19Speaker 01

  I think you're framing it the right way, which is, these are, VCs in some ways are offering a solution for a problem that they're trying to find. Right? So if you're like this large, this is my interpretation, what you're saying, like, you have these mega funds, so you raise like billions of dollars, and your LPs know that AI is hot right now. So now you're finding like this very niche set of areas on the AI stack that you want to invest in. I'm not even really paying attention to this specific space that you described what Oak is doing. So I don't have any real knowledge of it. But what I suspect is happening is that, you know, you're trying to put money to work. And there's just so much capital sitting in these mega funds. So I hear about this news like all the time. It's just like a crazy amount of money.

• 39:03Speaker 01

  I have no sense of like what the TAM is, or whether this is going to be like subsumed in some sort of like broader kind of governance model or whatever.

• 39:13Speaker 01

  But like, it's, it's a funny thing, because this, this feels almost like driven by just like, I have all this money, like, how am I going to put it to work? At least on the venture side of it. So maybe not answering your question.

• 39:22Speaker 00

  No, no, that's close enough. Jeff, something over you, do you agree with Eric that we're seeing maybe more startup clusters form with higher levels of funding, strictly because higher AUM firms need to get that capital to work?

• 39:33Speaker 02

  I think it's primarily because there are so many known problems that exist, and it's really, I think, once any company has traction, that's known faster than ever. And so then the cluster forms. It's like, you know, it's, and you kind of like repeat and rinse that, that cycle. So I don't know if it's necessarily like VC-driven. It's just, like, there's like such obvious problems, especially within software, that people spend time on. It's actually to the physical AI conversation, I find the clusters are slower to form within physical AI than the software, which to me is an advantage. These are harder companies to build.

• 40:15Speaker 02

  There's normally a hardware and software component, and often they're within very kind of like old, traditional industries. And so there's an advantage if you actually like look outside of Silicon Valley to find companies where often this, this kind of like fast follower effect doesn't exist, which is, if you look at like, we've done eight investments over the past year, and I think one of them has been in Silicon Valley.

• 40:43Speaker 00

  Wow.

• 40:43Speaker 02

  I think six of them have been in Los Angeles. And so we, as a firm, believe that, like, you know, some of these like old sayings are very true. And the one that always comes to mind is Peter Thiel, competition is for losers, right? And if you believe that quote's true, there's a lot of losers out there who are, you know, who are, who are other capitalizing companies or starting companies that, and so, so I think that's kind of like something we just think about a lot is, we don't, we don't want to be existing, or investing in categories that are easy to follow into.

• 41:23Speaker 00

  Yeah. One of your partners wrote a post talking about this, and they said, you know, agentic coding doesn't make software less valuable, but instead it quote, moves the value to software solving the hardest problems it can reach. And it sounds like you're saying that to, to get to a point where you don't have instant fast follower competition, you have to either go pretty far afield or literally get some wrenches out. That's fair. And the partner in question mentioned General Intuition, Alfred and Erebor as examples of places where they're working on solving data friction, iteration friction, and deployment friction. So applying that, Jeff, to the companies you backed down in LA, why were they the right choices for chapter one right now?

• 42:03Speaker 02

  Yeah, I think starting with Erebor, so they're obviously building a new bank. And to the earlier fintech conversation, it feels so obvious to us that fintech and banking is one of the categories that you can't fast follow into. And because there's regulatory reasons and also trust and security reasons why fintech should be a, I guess, like a hotter category today. And so when we, when we see a company like Erebor, Get their banking license fastest ever in the US. They have Palmer Lucky, they have, you know, this amazing cap table to kind of like consolidate AUM into very quickly. It's, that's like a pretty easy investment for us to make.

• 42:46Speaker 02

  And then if you look at general intuition, I don't know if you guys have followed their journey, but they had this company called, or it's called Metal TV. If you're a gamer, you probably know what it is, but you use it to clip your, your video game highlights and they've found a way to use that data to train physical world use cases around defense, drones, et cetera. And that's just like a, a data bet is you have this really unique data set that they have access to fairly unobvious. And then you have a team based in primarily the UK. They can recruit really well out of DeepMind. They're not within this Silicon Valley think tank.

• 43:27Speaker 02

  And so you have people who are doing really unique things with unique data sets. Alfa was another kind of hardware, software bet, teams being out Tesla, working on a fairly unobvious solution to, to helping car manufacturers integrate software more, more efficiently. So, yeah, they all, I think it's a combination of, of data, regulation and team that we look for and often that exists within regulated industries or really difficult industries to, to build within.

• 44:00Speaker 00

  Chapter one is based down in LA, right?

• 44:03Speaker 02

  Yeah, I think it's a pretty generous term because it's, we have a distributed team, but I'm, I'm based in Los Angeles. I grew up, I'm actually today in Menlo Park and it's funny because people think of me as being like this LA investor where I really, I've spent a lot of my life in, in the Bay Area, but I, I live in Los Angeles now and, and spend my, my, most of my time there.

• 44:23Speaker 00

  Yeah, but you said the Silicon Valley think tank. To me, you're describing in very polite terms, essentially an epidemic of group think?

• 44:31Speaker 02

  I think, I think there's a, a culture that exists within the Bay Area, which becomes very insulated and you can just, like, I actually like to just compare it, like what billboards do you see when you drive on 101 versus any other city? And, you know, if that's what's staring at your face when you're driving to the office every day, then you get to the office and you're talking about, I don't know, like different AI, AI topics all day long and that becomes all you think about. And to me, like that can, there's a, a cost to your kind of like originality of thought that exists in that environment. And so I'm careful, I, I love the Bay Area, so I don't want to, I'm, like, there's all

• 45:14Speaker 02

  these Twitter wars, it's like Bay Area versus LA or Bay Area versus New York. I truly don't care, like, I just think that's like, people do that for clickbait on Twitter. But for me, I do my best work when I'm not living in the Bay Area, at least at this point in my career.

• 45:29Speaker 00

  Yeah, yeah, Eric, I'm curious what your thought is about this because on one hand, people say, you know, if you have ambition, you have to move to the Bay Area. But we are, we are seeing a lot of cool companies being built, not just around the United States, but also around the world. And I could, I could list off some names that I like, but how do you avoid that kind of like poisoned thinking by supping too long and too deep at the communal well of thought?

• 45:49Speaker 01

  Well, first of all, Jeff is totally wrong. The real answer is Miami, right? Everyone should be there, right? You're on the road, you're a loser.

• 45:56Speaker 00

  All right, all right, wait, wait, wait, I have a take about that before you actually give your real answer. My thought about the Miami thing is that people just liked sun and partying. And so they're like, we'll just take our laptops there and buy nice houses and then it'll be just as good because we're the people that matter, blah, blah, blah, blah, blah. Anyways, you were saying.

• 46:09Speaker 01

  Okay, first of all, I'm going to roll that back, that joke, which is, you know, one of our best companies that Jeff and I share is Flex, this, this bank that we, we sort of briefly touched on, they're based in Miami, they're doing real work, you know, there's, there's a mantra within our fund that we see at Hustle Fund, which is that great hustlers look like anyone and come from anywhere. Full stop. And I really just truly believe, just like Jeff does, which is, you know, you can build great companies anywhere. There are multi-billion dollar businesses built in Wisconsin, you know, or in Indonesia, like, you know, markets with very different kind of parameters and so forth. So I guess when it comes to group think, you know, it's hard to escape this bubble. I live in Silicon Valley, I've been here for 26 years and all my friends, all of my

• 46:51Speaker 01

  neighbors work in Silicon Valley companies, right? But I think one superpower I do have is I'm not from here. I grew up in Detroit and a lot of, I think my friends are, you know, still based in the Midwest these days and I catch up with them. So there's an element of touching grass, but this is kind of like the fun part of the job as a VC is like, now I'm realizing it is a necessary part of the job to touch grass, you know, to, to get out of this bubble, like travel a little bit or go to some of these exciting hubs like Chicago or Atlanta, they're a little bit more overlooked because you surely do find founders who are less Kool-Aid, if that's a verb, into like a very

• 47:32Speaker 01

  specific way that companies should be built, have a very different kind of approach for how they're building or burning capital and so forth, and they can build great businesses too. So, yeah, I subscribe to what Jeff is saying.

• 47:43Speaker 00

  Yeah, I kind of came of age in the Chicago tech scene, back in the early days of, of Uber and kind of during the, the Groupon boom. And it was so cool, but then it didn't seem to have that staying power. Then everyone kept moving to Silicon Valley and it seems like such a missed opportunity to build something with a different perspective, a different LP base, a different, you know, local industry focus that what could have been, it just didn't quite, quite bear out that way. And if you're listening to Eric talk and you're thinking overlooked, man, we should build a venture capital firm called Overlooked Ventures, too late, someone already did that. So that brand is.

• 48:14Speaker 01

  That was James' fund.

• 48:15Speaker 00

  Yeah, is taken. Okay, but let's, let's leave the bubble and talk about something else. Now, Jeff, you're maybe best known for your time at Tinder, during which it became the number one grossing app on the iOS App Store, which is legitimately an amazing accomplishment. And Eric, you just put some money into Sage Haven, which is a play on, I think Safe Haven, which is a texting company for kids. And it struck me when I was just thinking about this, that you guys have kind of more consumer bona fides than most VCs that I talk to. And I'm curious i- if you think that there is maybe right now a dearth of consumer facing startups, just because everyone wants that, that sweet speed B2B AI markup. And Jeff, why don't we start with you?

• 48:55Speaker 02

  Yeah, I think it's probably been the biggest surprise of this AI wave so far that we haven't seen more consumer companies that actually leverage AI in interesting ways. I think I caught up with Josh Elman last week, who obviously joined A16Z and is a longtime friend and him going to A16Z, I think is actually just a great catalyst for more people to build. It's, it's funny, I think the natural, you know, you, you flock to like building a vertical AI company and then you realize two or three years later that maybe that's not easier than building a consumer company. And, and so I think, I think it's just, it will happen. And I've been, I personally like, I've been spending a lot of time trying to think about

• 49:40Speaker 02

  what the form factor might look like that actually makes, because the first iteration of consumer AI companies was like, let's take Instagram, use the feed, put some like generative content in and hope that humans like to, to just like 100% consume generative content. And I think that's just so far from what people want every day. And even the, the X algorithm change yesterday was interesting as a, everyone's clapping, but like that to me was a sign that people actually do want to connect with real people and real content. And so we got away from, from kind of like what the core value of consumer should be because AI was, it felt like this new superpower that people were trying to experiment with.

• 50:24Speaker 02

  And now I do think we're going to get more consumer startups and there's also a lot of really cool things happening within consumer hardware now as well.

• 50:32Speaker 00

  I, I was literally just going to ask, do you think these next breakout consumer companies are going to be hardware or software first?

• 50:38Speaker 02

  I think both. And we had talked about how much easier it is to iterate within hardware today than it was many years ago. And, you know, I think the wave of consumer hardware startups that have actually succeeded over the past several years is up, is very understated. So if you look at Whoop or, or anything kind of like health base, that category has been awesome. What hasn't been as, as great is, is consumer social, consumer marketplaces. But if you look at consumer health, I think there's been really amazing examples of companies that have really broken out over the past couple of years. And now, you know, we'll see what the next wave looks like.

• 51:16Speaker 00

  Is that because the incumbent companies in healthcare were so, going back to the word ossified, that you could just maneuver around them pretty quickly? Whereas the companies in social and other parts of technology are a bit more quick moving, like I'm not saying that Meta is nimble, but Meta has managed to give a pretty strong lock on a big chunk of consumer social forever.

• 51:35Speaker 02

  I think it was, to me, the companies addressing a new zeitgeist that. Incumbents weren't paying attention to. And so if you look at consumer health, like that's a big chunky category that people are wanting to spend money on. And you have a lot of income that's aging, who want, you know, want better products. And so to me, it's, you know, it's more so like, hey, there's health seems like such an obvious category. If you just look at like Maslow's hierarchy of needs, where, which that's like the most basic way to find consumer startups where, and that was why I actually joined Tinder a long time ago. I was like dating, you know, that's pretty high up on that list and all these companies

• 52:19Speaker 02

  are really terrible. It seems like this is a pretty good place to, to, to build a consumer company. And we're just getting back to that now. I'm actually really excited about like AI native fintechs or AI native, like anything that involves money and AI, I think is going to be awesome. And if you like, there's been some great examples of, of companies who are, you know, giving your agent a bank account and letting them basically be your financial manager. I think is, is going to be, it's like so one's going to nail that. I don't necessarily think it's going to be Robin Hood too, just because their product stock is so, you know, so busy at this point. So, but yeah, I'm, I'm, I'm pretty, I'm pretty pumped about everything that's happening right now.

• 53:03Speaker 00

  Eric, first of all, just give us the quick TLDR on why Sage Haven and then the broader question about creator, sorry, consumer focused startups being possibly the next wave.

• 53:11Speaker 01

  Yeah. So I don't do a lot of investing in consumer and I'll be also really clear that Sage Haven wasn't the deal that I led within my team, but I do have good context about it. I was a product manager at Facebook and Instagram for several years. So, you know, had some, some experience here, you know, Sage Haven, think of it as like a very safeguarded way for parents and those in your community to message your kids, right? And, and it's sort of like a parallel stack for, for iMessage that's designed for safety for your children. So as I have an 11 year old, I have an eight year old, you know, I'm really worried about like who and who's able to communicate with my kids. And this is like a really nice solution to create those kinds of guardrails.

• 53:54Speaker 01

  So I'm also really bullish on consumer AI too, even though we don't invest as heavily in this space.

• 54:02Speaker 00

  Sure.

• 54:02Speaker 01

  I like the Maslow's hierarchy framing that I think Jeff is putting together here because, you know, I've been thinking a lot actually about the X algorithm change that happened a day or two ago and like how suddenly so joyful the feed feels like, like all my old friends I see, I can see Alex's shitposts and.

• 54:20Speaker 00

  No, I'm, I'm, it's so true. I've had more fun on Twitter in the last 24 hours than I've had in the last six months.

• 54:27Speaker 01

  Correct. Yeah. I think like, it's almost coming back to like pre Elon days in terms of like how the feed feels. And as I've sort of been processing this, like just, just taking like that example, just social networking, I think people like it, you know, when, when from like 2005 to 2015, let's call it, it was a much more joyful experience to be on all these kinds of platforms. But then it kind of turned into this extraction phase post 2015, where especially tools have accelerated the company's ability to, you know, rage bait you into like, you know, getting more doom scrolled into the content, like, like extracting more of your data to do all sorts of interesting monetization and so forth. And we're starting to find that kind of pushback. And with the X algo change yesterday, I kind of like discovered that I really needed that

• 55:12Speaker 01

  of just like, oh, I can, there's a place where I can have fun again, you know, because, you know, that's just something that I need. So who knows whether X is going to be the ultimate platform for that, but there's got to be better tools with this current environment that can be created by some startup or some company to do that joyful thing, but hopefully in a way that balances extraction in a way that isn't like harmful, right? And I'll just like leave this last caveat to you of just like, maybe the need in, in terms of like a new socials in this AI world thing is, I think the majority, just qualitatively of XPMS I know from Meta are really militant about not letting their kids use social media, including with my kids, just because like we saw like what these tools were doing and like the addiction and all stuff.

• 55:57Speaker 01

  So that's always like the, the crazy irony of these kinds of products. You don't, you don't, you get high on your own supply.

• 56:03Speaker 00

  The, the most insane and radicalizing thing for me and self critical point was watching my children discover what a phone is and then their interest in stealing mine from my pocket and running away with it. Cause we're.

• 56:18Speaker 01

  That's everyone.

• 56:19Speaker 00

  We try to have a low screen household. The children don't have iPads. They watch very little video content, usually during a tough diaper change, they can watch Brock and Toad for a minute, but like not much, and they're just desperate for it. And it's made me really look in the mirror and, and ask myself, why am I so plugged in? And the answer is Twitter and, you know, the industry that we live in. Cause I don't think technology's ever evolved faster in terms of what we need to stay on top of than right now. So I feel the need to be plugged in, but I mean, I don't want my kids to live a life on a digital treadmill like that. And, you know, I'll just ask this because we're here now, Eric. What's your take on the, the restrictions on social media access we're seeing both at home and abroad? The right move?

• 56:58Speaker 01

  Yeah. I mean, I think the UK or some like European country said like at least 16.

• 57:02Speaker 00

  Yeah.

• 57:02Speaker 01

  I, I support that. I think even 18 potentially. I mean, if, if we're not allowed to give our kids cigarettes because we think it's harmful for their health, I think actually.

• 57:10Speaker 00

  That should be allowed. That should be a hundred percent.

• 57:13Speaker 01

  Yeah, exactly. Yeah. You know, exactly. True child of the nineties, you are over there. But, uh, yeah, I mean.

• 57:22Speaker 00

  1989, baby.

• 57:23Speaker 01

  There you go.

• 57:24Speaker 00

  Five months. Oh, yeah.

• 57:26Speaker 01

  Yeah, yeah. So, yeah, I remember the eighties very vividly actually. So, um, it's harmful. You, I mean, I, I really do think that, um, you know, brains are developing, especially for men until we don't stop developing until we're like 25 or 26 and then maybe women a little bit earlier. And, you know, this is the reason why like, like marijuana is like sort of dangerous. Like you don't, you don't, it's fine. I think if you're a fully formed adult because your neural pathways are mostly, mostly there. But like, I'd be really frightened if my, my son started taking like gummies pretty regularly now, because I think that's going to fuck up his mind. And I think it's just a similar kind of snow crash moment where like something about these data that, and this, this consumption is really messing with, with human brains.

• 58:09Speaker 01

  And there's probably enough evidence and scientific support for that at this point. So.

• 58:12Speaker 00

  I think that's, that's all pretty reasonable. I didn't think though of cannabis gummies as the prompt injection of young minds, but there you go. Jeff, I want to extend this point to something else. There's a lot of talk right now about AI regulation, trying to get this right. And I don't think we need to argue about the exact last one or two things that have happened, but directionally from where you sit slightly outside of the Silicon Valley zeitgeist, which I think kind of leans in one particular direction when it comes to this point, what do you think we should do at the national level, which is I think the real question today, to ensure that not only do we have a non cybersecurity destroyed world, but also ensure that startups don't end up regulated out of the cutting edge? Cause that's my real concern is that we're going to end up just granting too much power

• 58:56Speaker 00

  to incumbents and squash the little guy.

• 58:58Speaker 02

  Yeah, I think that's a huge, big question to unpack. Look, I'm like, I'm the most pro America person you can, you can probably meet. And I, my biggest concern is that you have people, especially today with a lot of the, the data center build outs that do not have the right information. And there's just going to be, you know, the, so it's self inflicted to be, to be clear, being there's been a lot of kind of like scary visions of the future that the biggest companies have been portraying around job loss. And it hasn't felt very productive for the industry.

• 59:40Speaker 02

  And I think it's, and so, and that's just, you know, I was actually pretty shocked that Dario and Anthropic, like a lot of their early comms, to me, even being in the, in the industry were really scary. And so now you ask the kind of like general American public to support AI and you already shot yourself in the foot.

• 1:00:00Speaker 00

  Plus repeatedly.

• 1:00:01Speaker 02

  Yeah. And so, but I think it's important to, to look at the bigger picture, which is if we don't figure out how to create a more positive narrative around AI and to take on these big infrastructure projects, we will fall behind. And that does come at the expensive national security. And ultimately I think, I don't know how it's going to play out because it doesn't seem to be making a ton of progress. And you read this in the news every day or, or you talk, you just talk to young people, right? People graduating college and their views on AI are really negative.

• 1:00:40Speaker 00

  Yeah, but they're also enormous hypocrites. There was a recent study from Brown, which is up the street from where I live here in Providence, and a professor gave out a midterm, it was take home, everyone did really, really well, they didn't believe it. They gave an in-person midterm, nearly everyone failed. And so to me, the, there's, there's an irony to the college kids saying, you know, we want authenticity in life. We really want to be real humans. We want to paint on our hands and then they, they cheat like mad on their homework and, and don't learn anything. And I, I think that points to just people are going to use AI a lot because if even the haters are using it, then it's going to be ubiquitous. But more on the regulation point, I want to pin you down on this. I'm curious what you think is the right way to decide what counts as a, a model that we might want to have some oversight of, or if we should even have that at all.

• 1:01:25Speaker 00

  Because I think we've seen China possibly restrict open way models from release and there's rumblings of an executive order on open source AI here in the States. All this has to be pretty worried. So where do you stand?

• 1:01:36Speaker 02

  Yeah, I mean, I'm, I'm pretty anti-regulation across the board. Mainly I don't trust that the people who have decision-making over which model, model should be available to the general public, you know, at least going forward. How do you, how do you keep continuity kind of like over multiple administrations with different political beliefs when you have committees who are determining which AI models should, should exist? And if you meet, if you go to Washington DC and you spend time with the people making these decisions, they often are pretty far from the metal in terms of what's actually

• 1:02:21Speaker 02

  happening. And so not to get overly political, but I think the last administration.

• 1:02:29Speaker 02

  Was having to deal with a lot of, a lot of new innovation and that oversight actually cost us a lot of time in terms of the amount of time that, that went into educating that group and, and ultimately the policies that they were trying to enact were, were, you know, by their nature, very anti-AI, anti-AI in many cases.

• 1:02:52Speaker 00

  Haven't we ended up though, exactly where everyone thought Biden was going to take us with the federal government having unwritten rules about restrictions of AI, opaque standards, a lot of having to go kiss the ring to get stuff out the door, OpenAI being told they can't release certain things, Anthropic fighting with the government. It seems like everyone's like, if we have Biden 2.0, I'm not trying to say they're right or wrong, then we're going to end up here. And I feel like we've ended up here anyways. And so I, what, what could the current administration have done differently? Because the only thing that I can see is they'd have to say like, we're not going to have control over mythos level models. And I don't think any administration would say that. So to me, I wonder if it's less partisan and more just the technologists are too far from the government, Jeff.

• 1:03:35Speaker 02

  I think that's totally fair. And, you know, if you do go to DC, you see the, you know, Anthropic and OpenAI have huge, huge teams in DC now.

• 1:03:44Speaker 00

  Yeah.

• 1:03:45Speaker 02

  Like massive teams. I don't think this is discussed often enough, but you have, you know, what would look like big headquarters in DC now, trying to work with, with both parties to, to create new policy. And it's one of those, actually, Alex, like, pretty impossible questions to answer, sitting where I am in, in Menlo Park today, just as a venture capitalist. But the, you know, I think this should be a bipartisan topic where we all come to the table and have some, you know, shared point of view on doing what's right for the country. And I hope that happens.

• 1:04:24Speaker 00

  I, I hope so too, because it would be really unfortunate if tech's recent shift to the right, kind of since the, the last presidential election a little before, ends up turning AI into something that is inherently partisan, because then we're just going to make very little progress as a country. And, you know, we're going to have President AOC, and then we're going to have President, I, I don't know.

• 1:04:44Speaker 02

  You'll just keep going back.

• 1:04:46Speaker 00

  Yeah. Ping pong, ping pong. And that's not good. Neither extreme is probably the correct place to be. All right. Um, why don't we end on something a little bit more fun than making Jeff explain his politics live to the entire internet? Why don't we give him, why don't we give him a slight break?

• 1:05:00Speaker 01

  Yeah. Thanks for doing that, Jeff.

• 1:05:01Speaker 00

  Yeah.

• 1:05:02Speaker 01

  Not me.

• 1:05:03Speaker 00

  So Eric, I have your, your criminal record pulled up here and I wanted to go, kidding, kidding. Um, so some, some fun things to wrap up with. One, Eric Charter Space. This is a startup, as far as I can tell, that is providing insurance for launches and in-orbit activities. I don't know if this is one of your deals per se, but I thought it was incredibly cool company. So I'm just curious, what is the Hustle Fund thesis on backing space-related companies? And do they fit into your pre-seed formula?

• 1:05:28Speaker 01

  Yeah. Well, first of all, thank goodness for the American laws that juvenile records get expunged at the age of 18. So there's nothing for you to look up at this point. So Charter Space, right? This is a insurance underwriting for space missions, right? I like to think of this company as Lloyds of London. So back in like the 17th century, this company formed for maritime insurance because there's these ships that are going out in these crazy excursions of discovery and trade. And if cargo was lost or the ship got lost, you know, there's a way for these businesses not to lose all of their money through this insurance product, right? And, you know, our, our bet is that more people are going to want to do more things in space, that there's going to be more payload that

• 1:06:10Speaker 01

  sent to space, or maybe habitats or something, you know, like data centers, whatever it's going to be. It seems like the, the chart is moving up and to the right in an exponential fashion in terms of like, what is being launched into space every day. And also, we have seen so many catastrophic videos on, on Twitter, on Reddit, whatever, of payloads being lost from like wonderfully huge explosions, like on the launchpad or, or something like that.

• 1:06:36Speaker 00

  You can say Blue Origin out loud. Jeff Bezos won't come by and slap you in the head. It's okay.

• 1:06:40Speaker 01

  Yeah. But, you know, I, I applaud like this, this early days of like brave explorers trying to do this kind of stuff.

• 1:06:46Speaker 00

  Hell yeah.

• 1:06:47Speaker 01

  I also, I also want them to continue to be in business to make these kinds of risks. So, you know, we're really excited about Charter Space and that we think that this is the right time for a company like this to be formed. And we're, well, it's yet to be seen whether we're too, too early or Charter Space is too early in this bet.

• 1:07:02Speaker 00

  Wasn't Lloyds originally a coffee shop?

• 1:07:05Speaker 01

  Lloyds of London?

• 1:07:06Speaker 00

  Yeah. When there were people aggregated and met to talk about this stuff. And then I think it became a business after that.

• 1:07:10Speaker 01

  Yeah. So my understanding of the history was that it did, like, there was meetings taking place at the coffee shop. I didn't know if it started as a coffee shop, but all the underwriting was taking place, just like a startup would, you know, like starting their company, drinking coffee at like Pete's Coffee or whatever it is, right? So I know that was where originally the work was getting done. And then, you know, off into the races.

• 1:07:31Speaker 00

  Yeah. I was thinking if it was like, you know, old school cafes to Lloyds, maybe it's like group chats today to.

• 1:07:37Speaker 01

  Yeah. It's the original Koopa Cafe, let's say, to the Bay Area reference.

• 1:07:42Speaker 00

  10 points for that. All right. And then Jeff, I was going through your portfolio. And one thing I noticed is that the number of crypto infra and usefulness bets you have is, is pretty cool. So the graph Starkware, Moonpay, Lighter, Layer Zero and others. You've had a pretty good foot in the crypto world. And I think that as we've talked about AI and just ad nauseam for the last couple of years, it's really fallen off people's radars. So what's the state of crypto today? What are people building that's cool? And when is the next crypto boomlet going to come to make us all feel guilty about not loading up on Bitcoin now that it's about 60K per coin?

• 1:08:18Speaker 02

  Yeah. I think crypto is one of those really misunderstood industries where you have, you actually do have real great founders still building within the category, but the markets, you know, on the liquid side are depressed. And frankly, I think there's just been a lot of founders who have found other places to spend time as well with AI and deep tech and everything else. You know, if you look at, if you look at the roadmap to most big fintechs, stable coins have been for the past three years, a huge part of the, of the focus. And that's going to keep, keep happening going forward. The things that are most interesting right now are bringing, you know, there's been a lot of,

• 1:08:59Speaker 02

  it's called RWA, but real world assets. So bringing new equities and new forms of assets on chain, which enables global access, 24/7 trading, gives you the ability to lend and borrow against those assets, which is, is pretty cool. If you, and that's been a big focus for Robinhood, if you, if you've seen the roadmap as well. What, what hasn't happened in crypto, what's not happening today are like these big, you know, like society-changing projects. If you, if you look at things like DAOs or, you know, like the, the ownership thesis, I think a lot of that has, has changed quite a bit. And so there's a, a move towards

• 1:09:42Speaker 02

  more institutional use cases, which for a lot of people building crypto, especially younger people, is frankly, like really boring.

• 1:09:50Speaker 00

  Yeah.

• 1:09:50Speaker 02

  Even though that's, that's probably where the most value will be created. But if you're, if you're like, hey, you're going to go build better, you know, financial technologies for the top 10 US banks, that's a very different pitch from, hey, we're going to change the way that society thinks about owning companies or kind of like the, so, so I think a lot of the enthusiasm for crypto has gone away, despite the fact that there's been a lot of progress being made.

• 1:10:22Speaker 00

  Can I just say that it's really, really funny that for a while DAOs were a venture backable category because DAOs are, and correct me if I'm wrong here, decentralized autonomous organizations, right?

• 1:10:33Speaker 02

  Yeah. The funniest part about DAOs, I don't know if you remember or sign this, but there were VCs, and this was like a totally real thing. You would go on like a call like this, and you'd have like 200 members of the DAO, and the VCs would have to pitch the entire DAO on why they should take the VCs money for the next funding round. And it was just like the, it was the funniest thing seeing VCs have to, have to like explain what they do to people who hate VCs. And, and it's like this, yeah, it's like a public, it's like this public forum. It was so funny. And.

• 1:11:11Speaker 01

  Proof of love.

• 1:11:11Speaker 02

  Obviously, proof of love. And you realize how inefficient DAOs are, or any organization when there's like 200 decision makers who have loud online opinions?

• 1:11:23Speaker 00

  Right, right. That's what I'm saying. Like what, what has been the pitch since Snap's IPO, when they went out and gave zero vote shares to the public? It's founder control, centralized decision-making. We call it founder mode now. But I mean, a DAO is the opposite of that. Like, if you think about it, it's literally like everyone gets a, like, oh, man, that's, that's, that's a historical quirk. The idea though of VCs pitching 200 furry avatars to get allocation though, and thus requiring them to have actual RIS, that's funny to me because not every capital allocator is funny.

• 1:11:53Speaker 01

  Alex has an excellent soundbite, by the way.

• 1:11:55Speaker 00

  I do what I can. Um, we, I've actually had a lot of fun today. You guys were great. Thanks for coming on. But let's give you guys some, some time to do some plugs. Eric, where can people find you and their firm online? And is there a startup you're looking for to back that you haven't got found?

• 1:12:10Speaker 01

  Yeah. So my website, hustlefund.vc. You can follow me at, at Eric Bond, E R I C B A H N. Companies that I'm hoping to see, you know, I think that we're generalists by design. So I think about founder archetypes. I love teams that hustle hard. We define hustle as great execution, means high velocity, you know, chasing after a large market. And they're actually really biased towards founders who are good friend of the house people who know how to sell. So, you know, moats are difficult, but I think the ability to sell and close is the best moat, especially at the early stages. So come check us out. Hustlefund.vc is a good place to begin.

• 1:12:50Speaker 00

  All right. Jeff, same thing over to you. Where can people find you online? And is there a company you're looking to back and haven't found it yet?

• 1:12:55Speaker 02

  Yeah. I'm JMJ on Twitter. Chapter1.com is our firm's name. I'd say companies we're looking to back would be, you know, we're obviously doing a lot within deep tech and, and that world. We still do a ton within fintech. We still do, do some crypto investing. Then AI tends to, to make its way into every pitch. So there's, there's AI in everything. But, you know, one company we talked about earlier that I'm very interested to find them back would be this like Agentic Finance Future Company, which Alex pointed to himself, maybe Alex is building it.

• 1:13:34Speaker 00

  Oh, no, no, I was quite literally itching my face. I.

• 1:13:37Speaker 02

  Oh, you were? I thought you were saying me. I thought you were saying me.

• 1:13:39Speaker 00

  No, no, that was not a signal. That was literally, I was trying to, I was trying to be demure about it and not bother you. Sorry.

• 1:13:45Speaker 02

  But yeah, I'd say, you know, on the whole, we want founders who are building serious companies, solving serious problems and, and would love to meet.

• 1:13:56Speaker 00

  Yeah. All right. Well, guys, this has been this week in the startups. My name is Alex. Wednesday's Beauty Venture Capital Roundtables. We're back on Friday. We'll see you then. Goodbye.