This Week in Startups - Slow Ventures' Creators Fund and Helion Energy's Fusion Reactors | 2084

Episode Date: February 12, 2025

This Week in Startups is brought to you by…Vanta. Get $1000 off your SOC 2 athttps://www.vanta.com/twistNorthwest Registered Agent. Form your entire business identity in just 10 clicks and 10 minutes. Get more privacy, more options, and more done—visitnorthwestregisteredagent.com/twist today!LinkedIn Ads. Get a $100 LinkedIn ad credit athttp://www.linkedin.com/thisweekinstartupsToday’s show: Jason and Alex cover…TBDTimestamps:(0:00) Episode Teaser(1:16) Welcome Alex Wilhelm & Sam Lessin; Slow Ventures Creator Fund(3:49) Significance of the $60M creator-focused fund(5:47) Community-building and product development strategy(7:02) Mr. Beast's success as a case study(9:41) Vanta. Get $1000 off your SOC 2 athttps://www.vanta.com/twist(11:17) Financial structures and strategies in creator investments(14:14) Addressing challenges in creator-focused investing(19:46) Northwest Registered Agent. Form your entire business identity in just 10 clicks and 10 minutes. Get more privacy, more options, and more done—visitnorthwestregisteredagent.com/twist today!(22:24) Startups optimizing operations in challenging climates(24:50) Transition to fusion energy and Helion's technology(27:46) Helion's fusion technology and CEO David Kirtley(29:16) Commercial fusion power timeline(29:43) LinkedIn Ads. Get a $100 LinkedIn ad credit athttp://www.linkedin.com/thisweekinstartups(32:12) Fusion energy's market potential and agreements(33:39) Comparing fusion to other energy sources(36:10) Safety and environmental considerations for fusion(43:00) Helion's upcoming milestones and team expansion(44:00) Lon Harris on AI, copyright, and fair use(47:00) Exploring fair use and IP infringement in AI(53:00) Market impact and licensing for AI technologies(56:00) Financial implications for AI and copyright(58:00) Distinguishing personal vs. commercial use of copyrighted materialSubscribe to the TWiST500 newsletter:https://ticker.thisweekinstartups.comCheck out the TWIST500:https://www.twist500.comSubscribe to This Week in Startups on Apple:https://rb.gy/v19fcpFollow Sam Lessin:X:https://x.com/lessinLinkedIn:https://www.linkedin.com/in/wlessin/Follow Helion:X:https://x.com/Helion_EnergyLinkedIn:https://www.linkedin.com/company/helion-energy/Follow David Kirtley:X:https://x.com/dekirtleyLinkedIn:https://www.linkedin.com/in/david-kirtley-490b8230/Follow Alex:X:https://x.com/alexLinkedIn:⁠https://www.linkedin.com/in/alexwilhelmFollow Lon:X:https://x.com/LonsLinkedIn:https://www.linkedin.com/in/lonharris/Follow Jason:X:https://twitter.com/JasonLinkedIn:https://www.linkedin.com/in/jasoncalacanisThank you to our partners:(9:41) Vanta. Get $1000 off your SOC 2 athttps://www.vanta.com/twist(19:46) Northwest Registered Agent. Form your entire business identity in just 10 clicks and 10 minutes. Get more privacy, more options, and more done—visitnorthwestregisteredagent.com/twist today!(29:43) LinkedIn Ads. Get a $100 LinkedIn ad credit athttp://www.linkedin.com/thisweekinstartupsGreat TWIST interviews:Will Guidara,Eoghan McCabe,Steve Huffman,Brian Chesky,Bob Moesta,Aaron Levie,Sophia Amoruso,Reid Hoffman,Frank Slootman,Billy McFarlandCheck out Jason’s suite of newsletters:https://substack.com/@calacanisFollow TWiST:Twitter:https://twitter.com/TWiStartupsYouTube:https://www.youtube.com/thisweekinInstagram:https://www.instagram.com/thisweekinstartupsTikTok:https://www.tiktok.com/@thisweekinstartupsSubstack:https://twistartups.substack.comSubscribe to the Founder University Podcast:https://www.youtube.com/@founderuniversity1916

How do you get Microsoft on board? Because this is a paid partnership. They're giving you money. This isn't like a letter of intent. My understanding is they wrote a check. They put a deposit down. Am I correct? So no, this is a power purchase agreement is that they pay as the electrons come off the grid.

Okay. But if we can't meet those timelines, we pay them. And so there's penalties. Oh, so you have a dagger over your head dangling. I love it. That's going to, your sleep's going to suck. Yeah. There it is.

Yeah, sleep score is going to go, sleep score is going to be variable. 0.91. This weekend startups is brought to you by Vanta. Compliance and security shouldn't be a deal breaker for startups to win new business. Vanta makes it easy for companies to get a SOC2 report fast. Twist listeners can get $1,000 off for a limited time at vanta.com slash twist.

Northwest Registered Agent. Starting your business should be simple. With Northwest Registered Agent, you can form your entire business identity in just 10 clicks and 10 minutes. from LLCs to trademarks, domains to custom websites, they've got you covered. Get more privacy, more options, and more done. Visit northwest registered agent.com slash twist today. And LinkedIn ads.

To redeem $100 LinkedIn ad credit and launch your first campaign, go to LinkedIn.com slash This Weekend Startups. All right, everybody, welcome back to This Weekend Startups. I am the host of this show for, gosh, over 14 years. This is one of the first five podcasts in the world, back when the first five podcasts in the world, back when the pod in podcasting was the iPod. iPod, for those of you don't know, was a device that played music only.

It had no internet connection. And they were awesome with me, my co-host, Alex Wilhelm, who 15 years ago was in college, I believe. More or less, I graduated in 12. So yeah, I was. Damn, okay. Two years out of school.

And he's worked at all kinds of places, TechCrunch, Crunch Base. Matter Mark, gosh, so many different gigs. He's got a great newsletter. you can subscribe to cautious optimism. Thank you, Jason. And it does very well. And he's one of my new subscribers,

now that I have a thousand subscribers on X, X.com slash Jason. He's X.com slash Alex, first name club. We have a crazy docket today. We've got to get right to work. Yes. I have a lot of strong feelings,

a lot about a lot of different things. But I say we save all the stuff I've been talking about online. And we just get to our first guess. Is that a good idea, Alex? I love it. We have some of the strongest guests. we've ever had on the show today.

So let's start with the first one. Sam Lesson is on. He's from Slow Ventures. Everyone's probably heard of him who listens to this show, but they have a new effort called the Slow Ventures Creator Fund, Jason. It's a $60 million vehicle, and we have questions. So let's get Sam brought up. There he is.

Thanks for having me, guys. Good to see you. Nice to see you, Sam. He's been on the program a couple of times. He's got his own great podcast with his wife and my good friend, Dave Morin, and Britt. It's called... More or less.

It's the more and less, right, based on your lessons. Yeah. And it's great pod, great pod. I got a shout on it once every 20 episodes or so. I get, I get, I get a mention. You should come on sometime. You want to come on sometime? I will absolutely come on.

Let's do it. I don't know that it's all in inspired, but it is the all-in format for really smart people talking for an hour, chopping up the week's news. But what's interesting about it, Alex, is these are power couples. They are couples who are both in the industry. industry. So Jessica Lesson, who's been on the show many times, from the information you know well. Of course. And then Britt Morin, who did Britt & Co. She was kind of like pinned as the Martha Stewart of the, or the next gen of Martha Stewart, but she's an investor now. So you got four and three investors and a journalist walk into a bar. It's a great show. Go check it out. But, you know, I saw your announcement today and you and I have a thesis around creators. Why don't you explain, uh,

what your thesis is. Sure. And then why this is so significant, significant, and I think it's very significant, that $60 million in LP money has come into this fund. This isn't just a bunch of Hollywood money. It isn't Kevin Hart or Kevin Durant or some celebrity.

It's mostly none of those guys. Believe it or not, I mean, I got permission to say it's people like MIT, Michigan, Rudkers. It's like the real institutions. And some of them actually are even, amazingly, some of those real institutions aren't even in our seed fund. They say we have plenty of

seed exposure. We have plenty. We like this strategy. So there's a real change going on here. And look, here, I mean, you know this better than anyone because you've been doing this 14 plus years, right? Which is there's something that something has happened in the last many years, which is the strategy for building companies and communities has actually inverted in some cases. It used to be you're two Stanford kids. You start with a product. You build community around that product. but you start with product first and technology first. As technology has gotten commodified, not everywhere, but in a lot of places, and as the internet has kind of exploded and we've gotten into these niche cultures from this mass culture,

there's this new pattern where you have entrepreneurs coming out and saying, hey, I'm going to build in a vertical I love, trust, I'm going to build community first, right? And I'm actually going to, in fact, if you put it in business terms, have a zero-cac sustainable audience, right? And then I'm going to layer products on top of that, right, in high LTV segments and places where they're really valuable. And we think this is a really big deal. This is kind of what happens.

This is the natural conclusion that I think some of us have seen for a long time of what it means to have a global internet population and a move away from mass culture to places where people are trying to find community and interests that are deeper and more niche and more specific. So this is a natural conclusion. Now, the interesting thing, Jason, as you know, is people have been trying to fund the VC community, the creator economy for years. This is not an unknown thing. It hasn't worked that well, right? Creator platform financing is meh at best. It sometimes works. It sometimes doesn't.

People have tried to finance individual products from creators who have their own problems. So the real innovation that we've been working on now for years is how do you properly seed fund creators in these communities for deep alignment, right, for cheap cost of capital to make something where we really can back these up-and-coming obvious business builders and entrepreneurs that have honestly been underfunded to date is the way we look at it. So to translate that, Alex,

I think customer acquisition cost, always the biggest piece of the puzzle when it comes to everything from a game, an app, a consumer package good or SaaS. You have some acquisition cost. That's KAC, as you mentioned. And then if you invert, as you're saying, Sam, the building of community first, and you spend a year or two or 10 building a community,

and you have trust with them, then you can launch a product. And we saw this, I think, the greatest example of the last decade's been Mr. Beast. Mr. Beast built this phenomenal audience. He builds really interesting videos every week or two. And then feastables, their chocolate bar is, I think, makes more money than, I don't want to speak out of school here, I have overheard some inside information. I believe taking a guess might have some inside information. I'm old and can't remember, Alex. But I think that makes more than maybe the advertising revenue.

And then, of course, he launched a Netflix show. Yeah. And Jason, if you don't mind me jumping in on that for a second. And here's the interesting thing is, East has been incredibly good for people understanding this world. Because I said, oh, my God, this can be way bigger than we realized. And I think that's actually been very helpful, popular. For two reasons.

One is it's helped LPs, the industry understand. This is not some small thing. It's not just niche business. But also, really importantly, it's gotten creators or people who have started a billion communities to dream much bigger, right? Which I think is great. Now, I will say from a slow ventures perspective, we actually probably would not underwrite a beast right from the early days.

We'd probably miss that one. You don't get everyone, right? And the reason is we think general entertainment is extremely difficult because it's so competitive. We're much more interested in people who, like, they love lawn care. They love chess. There's like a vertical, which is surprisingly big and deep, but not universal. The second is, you know, there's two elements. One is the pack, which is how cheap is it to acquire?

Like, how much trust do you have and how big an audience? The other is how valuable is the audience, right? And the problem with beast, and again, this is with respect for what he's done, is it's such a broad general audience. There aren't that many products you can sell, we believe, right, that are that high value. And so we're, it's not that we don't believe that there's things to do. there. We just think from a venture capital seed perspective, being hyper focused on niches with very valuable customers, right? And very clear products is the way we're going to run our

fun. Is that because the more niche you get, the easier it is to have a community that's incredibly aligned with the creative themselves versus a more general audience is going to be harder to kind of tie to one person most of the time? That's part of it. It's also the business building opportunities, right? Which is, you know, if you are, for instance, the god of aviation, You know, like for people who are really into flying serious planes, you are their god. And you have, you've built that trust over years. You have the community because you were authentic. You actually suggested the right things.

You've really built that personality. There are so many products to build and such a deep market for that, that you can then be that person. Whereas, again, general entertainment, the Kardashians, where it's not that it doesn't happen, it will happen. It's just very hard to underwrite. Nailing product market fit is every founder's top priority. But once you've got your product dialed in, you need to focus on selling it, especially to big customers. You know, we call them lighthouse customers. Some people call them whales. To land those deals, your security compliance has to be rock solid. Certificates like SOC2 or ISO, 2701 are the keys to building trust and unlocking these opportunities,

but they take time and energy, pulling you away from building and shipping a beautiful, product, and that's where Vanta comes in. They're your partner. Vanta is the all-in-one compliance solution, helping startups like yours and mine get audit ready and build a strong security foundation quickly and painlessly. How do they do it? Well, Vanta automates the manual security tasks that slow you down, helping you streamline your audit. And the platform connects you with trusted VCsos to build your program, auditors to get you through your audits quickly, and a marketplace for essentials like pen testing. So here's your call to action, whether you're closing your first deal or gearing up for growth, Vanta makes compliance easy.

Join over 8,000 companies, including many Y Combinator tech stars and launch startups who trust Vanta. Simplify compliance and get $1,000 off at vanta.com slash twist. That's V-A-N-T-A dot com slash T-W-I-S-T. So on the financial side of this, you guys wrote on the new website that you're going to invest between $1 and $3 million for an equity stake of about 10% on average. of the creator's cash profits. This, to me, sounds a little bit more like RevShare versus traditional venture investing, Sam. So I'm kind of curious, like, if you could kind of bridge that gap for me.

Yeah, what we care about is the company building. Like, when we're looking for entrepreneurs, like, we're not looking to invest in the cash streams from advertising, et cetera. The number one thing we care about, big picture, is alignment with creators. We learn this from our seed practice. You know, when we do our seed investing in traditional companies, it's usually around 10%, it's usually one to three million dollars, right? And we don't do board seats.

We're looking for entrepreneurs that we can be their first text message. We can be deeply aligned with them. And so the thing with creators is we want them to have the freedom to explore the best business models, period. If it happens, it's a highly cash generative business, but it isn't a company. We don't want to get crossed swords with them. We don't want to be like, well, don't spend your time on that because that's outside of what we invested in. You promised us X. We want to say, we're just investing in you.

We want you to have business ideas, but let's just be aligned. And so there are lots, there are some extra nuances in that. Like, we always make sure on cash, for instance, that there's like, we'll say, hey, if you earn over a million dollars next year because you're focused on some cash generating business, then pay us, but less, don't worry about it. So there's some nuance. But that's the emotional and innovation in the direction, which is we just care about alignment holistically.

I think it's a really interesting model. And I've started to think about it ourselves. About two years ago, I saw this guy on TikTok, Chef. reactions. And I thought, my lord, this is hilarious. And it was like the only person I followed on TikTok. And I'll pull up a video of it or Alex, you can pull up a video of it. And we incubated it. He came to our founder university. Then he came to our accelerator. And I said, listen, let us just give you a small amount of money, talk about business with you and how businesses are formed. And if something comes out of it, great, if something doesn't, here's a video of him.

you know, he's getting these, Sam, these brand sponsorship deals. And I think this, what we'll play here, go ahead and play it, Alex, I'll talk over it. Headley and Ben is like the number one apron company. Like if you're a chef, this is what you wear. And they recognize him. He goes to French laundry because they see he's in Napa. They invite him to come.

They make him an entire meal based on his channel because they're so enamored with him. And I think the brilliance of what you're doing, Sam, is that you understand. endemic advertising, endemic products. What does that mean in my career, whether a Silicon Allie Reporter in the 90s, Engadget in the 2000s, AutoBlog,

This Week, and startups are all in. There's an endemic group that loves certain verticals. Then you know who the advertisers are. This Healy and Bennett, whatever these aprons are. I wouldn't know that, but he knows that. So there's just a clear path. And he's launching a new product that we're backing, and I think it's going to do great.

There's one issue that I've seen come up over and over and over again, which is, well, there's really two. There's the Martha Stewart issue. What happens if you get pinched and you go to jail and you have this dependency on one person? What if Mr. Beast, God forbid, got hit by a bus? It's the standard, you know, talent issue. But the second issue that's, I think, the real one, is what's included in the deal, what's not? Mr. Beast had a burger and there's some sort of lawsuit going on about Mr. Beast, Burger. Then there was another person who contacted me to invest in, like, you know, some projects he

was doing. And then, you know, there was just constantly like, are you investing in Mr. Beast and project three, four, and five? Or are you investing in projects two and seven and projects three and five hit? And you just made a mistake. And then where did they spend their attention? And then if they do speaking gigs or books, totally. Does that get included? Does not go. So how have you handled this? So 100% this is the problem. And for what it's worth, we actually use Mr. Beasberger as an example of what not to do, right? Because you don't, and the thing that is so different about these creator and community

driven businesses is the brand and there's a lot of equity and value that's held outside of feastables, right? It's actually Jimmy, right? It's not. And so it's like the question is, and I actually think in my experience that by far, the place that, this is with venture capital in general, that people get sideways is when they're misaligned, right? When I've invested in X, but you're doing Y or things like that. And so that's why I go back to this whole thing, which is our approach is extremely holistic. There are

obviously things that are outside of the deal, right? You know, if you say, I don't want to be a creator anymore, I'm going to go work at a law firm. That's not a deal. It's about the creative business, but it's very broad the way we look at it. And some creators don't like this. They say, well, we want you to invest in X and not Y. And I'm like, I get it. It's just not for us then. We look at it holistically. And the thing we like to, to say to people is two things. One is the reason we're able to give you unrestricted cash and trust you is because we're fully aligned. I'm not going to tell you what to do. I'm not even getting mad because I actually, I'm trusting you with the money to do the right things.

I think it's the right emotional thing. But the venture capital mindset, what we do as seed investors is we're looking at hundreds of deals to find one, right? And it's kind of a similar thing for now. Now, I do think, and Jason, I'm curious what you think about this, because you have experience here is, look, a lot of people also come and say, look, Sam, we like this idea, but you really need to pair this person with a business person. They're not business people. And I say, I heard this story before in 2003 about tech founders, right, where you'd find the kids who are great engineers and say, well, that's fine,

but they're just engineers. We need to find, and I'm like, that's not the way this is going to evolve. The best will rise and figure out the holistic picture. picture, but it will be a minority of people. It's not most. When Larry and Sergey were going public and they were going to make a run at it, they said, you know what? To engineer, Stanford, let's get Eric Schmidt. He did Novel. He did this. We can put him in charge. There's an adult in the room. You can do it. And it was unnecessary. It was performative. Eric did provide massive value. Then Zuck came. And, you know, Peter Thiel was like, you know what? We don't need somebody to tell

Zuck what to do for better or worse. And then he obviously had Cheryl Sandberg as a collaborator who taught him how to scale the ad network. And you were there, I think, for a lot of the Sam. So you probably have some thoughts on it. But people overindex on the business guy. And I specifically say guy, but it could be gal like Cheryl Sandberg was the business gal and some people will business guy, Eric Schmidt. You can learn business. There are business experts who are verticalized. You cannot learn to have a motivation and a sensibility about content. That is something innate in the creator. And I will include engineers like Zuckerberg in that creator class.

I'll include Elon in that creator class with SpaceX. There's something about the motivation and the drive. That's the magic. The business people can fill in on the margins. Well, I'll give you, Jason, I'll give you a funny one, which is this is actually, I'm stealing this from someone from actually. actually from Mark's sister, Ariel Zuckerberg, who's a great investor. I've met her, yeah, she's great.

She's wonderful. So she says, what we're looking for is Riz and Tiz. Right? You gotta have Riz, right? Which is like the, you know, this is the, they've got, in their community, swagger, charisma, etc. And then Tiz is autism. Like, you need to be a little bit autistic about, about the topic or area you're in, right?

Yeah, this is why I've never had, that's why I'm not a billionaire. I have the Riz, but I don't have the autism to focus long enough to be a billionaire. I mean, you and I might have that in common. But the point is only that I think we're here to support those types of people. And that's the thing that's special. We do care about people who have demonstrated commercial instincts, right? So you can't, if you're just a creator and you're creating video, that's

wonderful. You can build a great business doing that. This content oriented is fine. But, you know, the people who really are ambitious, you know, they've shown that they can monetize, they care to, right? That's an important signal. But, you know, you do the math on it. Let's pretend you're a successful creator making a million dollars a year top line on your content. You know, pay your, pay your team, chop that in half for taxes. You're doing fine, but it's not like you have a bunch of investable capital for business building, which means that you're much more likely to take the sponsorship deal rather than say, no, no, I can make a better product myself, right? And what we want to do is enable those seed creators to say, no, no, no, I have a better idea. I know the community. I have the

access. I want to build something special. And as you said, in an ideal scenario, it goes way beyond where they start. Hey founders, you want to build the next great billion dollar business, right? Well, you're going to need two things, a killer idea and a properly set up company. I can't help you pick the idea, but I can help you get your business started the right way with Northwest registered agent. With Northwest registered agent, you can form your business for just $39 plus state fees, one of the most affordable options out there.

No hidden fees, no upsells. Just a simple, straightforward process. And here's how it works. In 10 clicks and 10 minutes, your LLC is officially formed and ready to operate. Northwest handles all the paperwork fast, accurate, and hassle-free. So you can focus on building your business, not navigating all those crazy legal forms. Thousands of entrepreneurs trust Northwest registered agent because they make business formation affordable, efficient, and stress-free. And with their expert team standing by, you'll always have the support you need.

So here's your call to action. Don't let paperwork hold you back from your entrepreneurial dreams. Get started today at Northwest registered agent.com slash twist. That's Northwest registered agent. com slash twist for just $39 plus state fees. Your business can be up and running in no time at all. Get more value, more convenience, and more peace of mind only with Northwest registered agent. It's so great.

I love the idea. Have you made an investment yet out of the fund? Sometimes people have. So, yeah, sometimes you warehouse deals. No, we've actually made. about six or seven of these out of our opportunity fund. So we have, we have LPs that have allowed us some flexibility.

So we've learned by doing. Which one would you say you've had the best experience with, learn the most from? You know, all of them. You know, the first one we did was a woman named Marina who's crushed it with language learning school.

She was an entrepreneur by background. She's dramatically grown since we've invested. It started several businesses. You know, we've done one creator called Vin Wiki, who does some great stuff around cars. There's incredibly passionate cars. communities and, you know, things like that. So we've done a few. But we now, the reason we've

kind of switched this from our opportunity fund to his own special thing, which is the first time we've done this, you know, we did a lot of crypto investing. We kept it in the core fund. But this really is a different style of investing that we really believe deserves its own vehicle. And so we're here to double down on it. I love it. If anybody wants more information on the Twitter, you can interact with Sam. Are you lesson or Sam lesson? I'm just lesson. I'm not as good. you guys, I don't have Alex, but I've got a lesson. That's pretty good. I mean, it means you were, you were an OG, you were there in the first six months of Twitter slash X. There you go. DMs are open and, yeah, I ping him if you got an idea.

And Jason, let's find some deals to do together. I, you know, I'm going to introduce you to chef reactions because that's very early on. And I DMs you there. And then, yeah, let's find somebody to collaborate on. I love that. I think there's so many people, one of the things Alex and I have talking about was what is the impact of startups being able to do more with less? And you wrote an essay on it at some point, I think, and it's been like the trend of our careers, you know, used to take 20 people and $3 million to launch a product, then five people and $200,000. And now with AI, you can kind of fill in all these gaps. And I think that's what the creators are going to really take advantage of. And it seems like the niches are small. Dave Morin and I were actually talking about our love of a niche

app called Slopes, which tracks you when your speed. I was using it this weekend with Dave. Okay. So Slopes is awesome. You can get your speed. You keep track of it. I've been using it for four seasons. It's really interesting. We have another one, Go Polar, which is a app for people who do coal plunging and saunas. Now, you think these things are small niche app, but they're done by hundreds of thousands to low millions of people. And because they have, because the cost of building apps is now under a million dollars and two people. It just opens up a lot of doors. And it used to be whoever your language person was or the chess person, that might be a five to 10 million dollar effort to make a world-class app. Now it's a 500,000 to a million. So I think the timing is

particularly good for you, Sam. Well, I appreciate it. I appreciate you get in it. And I'm excited to work with you guys on it. It's a new thing. Can't wait. Cheers. That's Sam lesson, everybody. Well, done. Thank you. You know, we always like to have whoever the, Alex, main character of the week is a concept we use here on the program. And I just saw Sam, you know, and I'm like, you know what Sam, I think is the main character here. Also the data Republican one.

I want to have her on. We will work on that. But here's from my and how this, why this show works is sometimes you push a little bit on like what we can pull off. Yes. But it's always better because the last couple of guests you've thrown in politely at the last minute have been clutch. So I like this. And on the same lesson point, I didn't get to put this in when he was here. He had a hard out. But I have like six names that I want to send him that are possibilities for his fun just because I loved him already. So like I think he's really on to something.

But let's do a hard right turn. Let's talk about fusion, yeah? Yeah. And I just want to tell Jesus in the comments, Jesus is here, you know, if you're going to troll me on my height, 5.3, understand. I was 5.9 at one point. I've shrunk into 5.8 and a half. I'm in the best shape of my life. I'm rich. I've got more friends than you can imagine. I have an incredible family. I got a 91 sleep score and I don't ever have to work and I ski 40 days a year. Jesus, if I was 5'3, I'd be the happiest short king in the world. You have to come at me with something more like the Knicks haven't won a title since 1973. If you want to come at, J-Cal, that's where the pain is. I'm not fat anymore. You could have gone after that. I mean,

the pain in my life is that the Knicks haven't won a title since 1973, but we got the best team since Patrick Ewing. So you got to try a little harder on the hard insults. The sleep score is amazing. 91 sleep score, Ray from Rivers in the in the chat. She knows. When we break 90 in sleep scores, you're good. I was on time today because of that sleep score. But let's make our right turn because we went from creators and now we're going to. We're going to Fusion, ladies and gentlemen. This week in startups, killing it. This week in startups. Shout out to Jesus.

We are working on a project called the Twist 500, finding the 500, most exciting, most potentially lucrative private market companies in the world. One of them is a company called Helion. You may have heard about them in the news. Sam Malman has backed them. They've raised about a billion dollars. They're working on a very, very interesting piece of fusion technology. We have a clip we're going to play.

But first, let's welcome co-founder and CEO David Kurtly to Twist. David. Nice. There is. There is. Nice to see you, David. Ninety-one sleep score. My goodness. Yeah. It's pretty good. And you know what? I just want to say two nights in a row. So I'm looking at it right now. 91 last night. You want to know my deep sleep? One hour, 32 minutes of deep sleep, 21%.

Is that good? And I don't mean to flex here. But last night, two nights ago, 92 sleep score an hour and 13 minute, 15% deep sleep. when you get that level of deep sleep with a bulldog in your bed that's significant with a snoring bulldog it's significant than two nights before were 74 and 56 because the Super Bowl

and then the night before I was out late and I had a McAllen 18 Alex close your ears I can still smell McCallin 18 in my nostrils but David we did not have you on to discuss your A-sleep score although we could we want to talk about Helian

I think thought it'd be best to start with just a really compressed version of what the technology is behind Helion and why it's different than other fusion projects out there because you're not building a tocomac, you're not building a Stellarator, you're taking two plasmas, slamming them together, and then crunching them and then pulling energy out of it. So can you just give us the quick rundown of how it works? Yep. So we build fusion at Helion. Fusion is how the sun works. It's how the vast majority of energy in the universe is generated. And right now, we don't on Earth, we don't make electricity from fusion.

as humans. We need to be doing that, and we can talk about all the reasons why. Fundamentally, what you're doing is taking lightweight isotopes, helium and hydrogen, and then at extreme pressures, 100 million degrees, a thousand atmospheres. You force those together, they form heavier atoms and release a lot of energy. What most people do in fusion is do that, do those reactions to generate heat, to boil water, to run a steam turbine. Our pioneering technology at Helion is actually take that energy, both electricity and magnetism and the fusion particles themselves and extract them directly as electricity.

Super high efficiency, which means the systems get smaller, cheaper, faster, build, all that stuff. And the way this works is when you collapse your two plasmas together from opposite sides and then crunch them, it pushes back against the magnetic field. And if you change the field, that creates a current.

So you essentially pull electricity directly out. You take away the whole steam power component entirely. Alex, you got it. There it is. I don't, you don't even need me. Let's do it. I do need you. I need you to build it.

I just prepped about how it works by watching a lot of fusion videos. So my thing with fusion is apart from the fact that I'm excited by it, and I understand the massive need we have for more power around the world is just how close we are to getting net positive fusion production into commercial use. So I know everyone probably has to do this. Like when do we think this arrives? Give us a range. The earliest we could be powering a data center with it is X.

maybe, you know, the midterm would be why. And, you know, if it's hard, you know, the hard case, the worst case is we get it this year. Give us the range because that's what your investors must ask when you raise a billion dollars. They want to know the timeline. All right, everybody, it's 2025. If your B2B marketing strategy for the new year doesn't include improving your ad targeting, well, your ads are going to get lost in all the noise. It's crazy out there right now.

We need you to target your ads properly. If your business is marketing to other businesses doing B2B advertising, LinkedIn ads, is amazing because it guarantees that you reach the right audience. I'm a venture capitalist. Somebody wanted to reach me with their new product or service being tied together, not targeted. I don't need an ad for HR tools as an example, right? There are other people who work in HR. I work in venture.

So you want the HR people to get the HR ads. you want the venture capitalists to get the fund management software and services ads. Makes total sense. And LinkedIn's advertising tools will connect you to decision makers based on their firm graphics. Ferma graphics. You may not have heard that term before. That's your job title, your industry, your company size, and more.

LinkedIn has two things that you need to grow this year. First, reach. LinkedIn has over a billion members, 130 million of them are decision makers, and of course 10 million are C-level executives. And number two, LinkedIn has impact. B2B marketers report two to five times higher return on ad spend compared to other social platforms so you don't waste your money. Plus, 79% of B2B marketers say LinkedIn is the best platform for paid media.

LinkedIn allows you to build the right relationships, drive results, and reach your customers in a respectful environment. They're not talking about politics. They're not doing yoga. People are focused on business, right? And that's what you want when you're trying to do. deliver a business to business message. So start converting your B2B audience into high-quality leads today. We'll even give you a hundy, a $100 credit on your next campaign. Go to LinkedIn.com

slash this weekend startups to claim your credit. That's LinkedIn.com slash this week in startups. Terms and conditions do apply. We just raised another $425 million, our CD-F, invested in manufacturing, and then push our seventh generation prototypes. So we build a lot of these systems that do fusion. That's one thing that sets Helion apart. We're builders. We built a lot of hardware up here in Everett outside of Seattle. And so that's what we do. So our seventh generation system came online and started the early operations last year.

We got a lot of work to do this year to get up full power, doing lots of fusion with it. Goal is that we can make first electrons in 2028. So that's a deal with Microsoft that we signed in 2023 to start building a first power plant for them, making electrons and then dialing that power up through 2030. So our goal is first electrons 2028. I'm going to need a sleep score 91 to get there because we got a lot of work to do. And then full power by 2030 and we start deploying power plants in the early 2030s. Can I double click on first electrons?

Does that mean that's the first time that in your manufacturing facility or lab, this system will generate power? Or is that when you have taken it out of the lab and have built an actual power plant out of the technology that then power is taken out of? Yeah. So the goal is that's the first revenue from electricity. Got it. Actually, somebody buys those electrons. So our seventh generation system we built now, it's a job.

The whole goal is to demonstrate not that we can do fusion and make lots of energy from fusion. We've done that. Lots of other folks have done that from fusion, but actually make electricity. And so that's the goal of what we built now and what we're testing today. How much technology risk is there left in your model compared to more conventional reactor designs? Yeah. So two parts of that. One, because you're skipping over that whole middle step of steam turbine cycles and all of the closed-loop fluids and all the engineering around that, you actually have less technology risk because you're just missing a whole step.

How do you get Microsoft on board? Because this is a paid partnership. They're giving you money. This isn't like a letter of intent. My understanding is they wrote a check. They put a deposit down. Am I correct? So no, this is a power purchase agreement, is that they pay as the electrons.

come off the grid. Okay. But if we can't meet those timelines, we pay them. And so there's penalties. You have a dagger over your head, dangling. I love it. That's going to, your sleep score is going to suck. Yeah, there's, there it is.

Yeah, sleep score is going to go, sleep score is going to be variable. 0.91. What is the story with fusion versus nuclear? Because, correct me if I'm wrong, a piece of this was that nuclear was D-O-A, in America, you couldn't do it. Germany was turning them off. But then all of a sudden, France is hosting an AI conference the last couple of days. They are getting a bunch of people

like UAE, Mubata, I think, investing in their country because they have so much nuclear. Germany and the hippie-dippies over there are considering turning their nuclear back on after the whole undoggle with Putin and Nord Stream. And here in the United States, looks like we're going to be ramping stuff up. So now, I think you were, not in a race with nuclear, but maybe you are, you're not fighting this war for energy in a vacuum anymore. There's, I think, two really big competitors, new nuclear and solar plus batteries. So maybe you can compare and contrast those two. I mean, I think when we founded Helion on the order of 10 years ago, we were at a place where we were looking at small power growth in the

United States. And our focus was how do we replace fossil fuels and maybe more expensive technologies with cheaper fusion baseload, clean power technologies? What we're seeing is explosive growth of power demand in the U.S., right? I mean, AI data centers, you know, why we're working with Microsoft and others is that the 2% growth in power that we had seen for year over year for maybe decades, that's gone. Like, we're seeing massive power growth that needs it. We're going need it, it's got to be on all day long. So base load power. And it's got to be low cost. And customers aren't going to buy expensive power these days. And so that's our focus. I think that there's room for everybody. That's the general PC CEO talking that there's room for everybody in

the market. It's a huge market. I do think we focus on things like safety that we know Fusion when you turn the key off. Fusion turns off. And so that's one of the focuses that we focus on on heleneers, which means you can build them much faster. The nuclear regulatory agencies passed law last year that said fusion is not regulated like nuclear reactors. Fusion generators are regulated like hospitals. And so it's still highly regulated. It's really important to do it safely, but you can go build quick and build much faster. And that's one thing that and they unlocked Microsoft is that we had a path that didn't take 10 years of regulatory approvals. We had a path that was much faster. And so we think that. And then in terms of solar and battery,

We live up here in Seattle that yeah, when where the sun is shining and bright and low cost solar, you should put it in. Absolutely. There are places that I see that every day where that's just, that's going to be hard. Tell me about safety in these things because obviously, you know,

that is a concern with nuclear power. What are the safety concerns here? Are these things going to be too dangerous to put near people? How far out do they have to be from a city? and how concerned are you about that? Because I remember hearing people being concerned, and then I've never heard anybody bring it up again. So what's the state of safety when it comes to fusion? It's super critical if you build a technology, but then either it takes a ton of time and money to regulate one of that technology or the customer doesn't want it and the community doesn't want it, then you haven't actually solved the mission. And so at Helion, we focus a lot on that, both the technical case and the operational case for safety,

Fusion fundamentally, the way fusion works is you put in your fuel, hydrogens and heliums, these lightweight common materials. You then, when you don't want to run it or if something happens, you turn the key off, fuel stops flowing in, it turns off the same way as every other natural gas or other system you might have. So fundamentally, you don't have the big safety concerns, like the evacuation or those types of things. Those all go away with fusion.

That's really powerful. We've been able to make that case to all the regulatory agencies, so we're regular by the state of Washington, the Department of Health, been working with them for a lot of years on regulating all our previous machines. And to make that case. And I think that's really important that we've able to demonstrate that. However, just to be crystal clear, fusion is still a big industrial power system. This isn't a thing we imagine is going to be on the back of a DeLorean just based off the physics.

And so what that means is that we build industrial safety systems and industrial operating equipment. So this is regulated by the Department of Health, like a particle accelerator, a hospital. And so you do have trained operators. You do have shielding systems. And we do have some radioactive materials. And so we have to make sure that just like a hospital, we're regulating these, we're controlling these. And we spend a lot of time and money thinking about how to do that. What are the elements you're using besides the obvious ones like hydrogen or whatever that are common? What are the uncommon ones you're using? And is there an issue with those being available? Where do they come from and who controls them?

Yeah, we spend a lot of time thinking about the manufacturing of these systems. We spend a lot of time on the generators. We know that the cost of manufacturing, if you do a really good job of manufacturing on your product, comes down to the materials cost. So if you're using ultra-rare materials, that's going to drive your product cost and then your electricity cost. And there's almost no way around that, as good of manufacturing as you want to do. So we use common aluminums and steels and coppers and some more complex alloys of those, but not rare materials.

And we spend a lot of time on that to make sure that as we build these systems, it can scale to big scales. The other part to that is the fuel. And we use in Helion, we use a different fusion fuel than other folks. We use a helium three, which is called light helium. It's actually quite rare. And one of the things we pioneered very early in the company is actually manufacturing that. So we make that from hydrogen, from heavy water, from things that are very common and found in all water on earth.

And then we've demonstrated that we can make that helium three. Fundamentally, you make it from fusion. itself is that from the fusion process, it's one of the actual things that comes out of and we measure that on our sixth generation system that we could do that. We can make this helium three. And then it's a matter of filtering it. Why other folks don't make helium three is that you have to actually be able to do it to what you said efficiently and at low cost.

And so what you need is efficient fusion. And that bootsteps right back to that very first question. Because we can efficiently directly recover electricity, guess what that means? we can actually make the fusion fuel efficiently too. That's such a bummer because I heard that there's a lot of helium three on the moon. And I was really excited about lunar mining operations to power our fusion future. It's much lamer if we're going to do it in the same warehouse we're doing the power production. There's a couple of companies out there talking about how do you go to the moon to mine helium three to sell to helium.

But yeah, I think it's, yes, it's lamer, but our goal is to make electricity. And if it's lame, great. That means it's low cost. If it's sexy, it's probably expensive. So let's focus on how do we get it as lame as possible. Launch costs are coming down. Get excited. So what is the next major milestone from Helion that we should have our eyes out for?

Because we love watching technology products mature and come to market. So, David, what's the next big bang? Yeah. So the big ones from last year was get our seventh generation machine belt and get started operation on it. Big technology goal internally is get that machine running a full power, making electricity, showing you can do that from fusion. And in parallel to that, building out U.S. manufacturing of the fusion industry. industry, which is a brand new thing. So that's mass producing electrical components,

ones we call capacitors, building more R&D and advanced semiconductors here in the U.S. And then we haven't talked too much about all the hard manufacturing we do in terms of machining and electromagnets. And we've got a lot of those to build too. So we'll be doing that over the next year and then starting to work on that power plant. So one of the bigger announcements we'll have coming out later is where that first power plant for Microsoft will be. I'm going to guess Redmond, and then that means that I can come up and see you guys and Zapp at the same time for a field trip and see all the American fusion. David, thank you very much. And we love to ask, is there a particular role that you're looking to hire for that you're having a hard time feeling that we can give a shout out for while you're here?

Oh, I love it. Yeah. So we are a company of builders, hardware building. I think right now over 75% of our team are technicians and assemblers that are building. And so we're hiring anyone who is great at building complex technology with their hands. a lot of building out in the engineering team, the engineering side. So I'm a voltage power electronics.

Come on by. Yeah. Go to the website, check out the careers page, and go save humanity by making energy free. I mean, that is going to be a wonderful future when energy becomes,

you know, 90% cheaper or whatever it's going to become. Just great job and appreciate the hard work you're doing here in America. Make America even more extraordinary again. Amazing. All right. Thank you guys. All right. We'll talk to you soon. Bye, David. Thanks, man. Take care of David. You know, we have one other major story that's a little bit underground. I thought we bring Lon Harris on, new editorial director here,

back from the future when he was the original news reader on the program, gosh, 14 years ago. But I really wanted him or you to tee up this story around large language models and copyright and IP. So the headline here is that Thompson Reuter, has won its case against a company, a former legal AI startup called Ross Intelligence. Now, if my understanding is correct, Jason, this is not actually about generative AI, but instead about training a different type of AI model,

kind of in the pre-LLLM era, but people are really applying this decision by the judge to the current moment. So it is a very important point. And what happened is the judge ruled in Reuters' favor, essentially saying that the information that Ross intelligence had procured, and we can talk about that process, did constitute a copyright violation and did not fall under the protection of fair use. Now, there's four things that come up.

The four-part test. The four-part test of fair use, I've been through so many times because, you know, as a, when I was doing blogging or we would cover something in our magazine, sometimes we would want to use somebody else's IP to, you know, make a point, right? And the purpose of what you're doing, what you're using the content is important. Is it for commercial purposes or nonprofit uses? Is it an educational, right? In an educational setting. Are you doing commentary?

So there was a famous case, Alex, where somebody did a voiceover, producer will know all about this. Actually, Jason, let's bring Lon on right now. Oh, okay, great. Hop on and help us talk through this because you are our media guy. Let me walk you through this. If you, there's also how much of the original work you're using. Is it commercial, non-commercial, educational?

In other words, it's a context under which you're using it. And the percentage that you're using, right? And then the nature of the work also comes into fact. So if the work is the real world, like a documentary or something, that can play into it versus something completely fictional. And then how does this affect, and this is the most important one, the potential market for the original work, right? And so what does that mean, practically speaking? If you're Napster and you say, hey, we're just providing these MP3s or people are sharing them online, you sell CDs, you sell live concerts, this isn't, you know, impacting you.

Well, that's the wrong way to look at fair use because the potential market for the original work is what's critical. Now, when you look at something like the Internet, that was a potential market for all content creators. You can't just steal stuff. Now you look at the exemptions. If you were doing, Lonnie, you might remember this

or be able to look it up real quick, there was a guy who went over the prequels and he just did commentary on the entire movie. Put it on YouTube. And Lucas was like, hey, you can't do that. That's the entire goddamn film. He said, well, I'm doing commentary. I'm not actually making money from it.

I don't have it monetized. So these things act, you know, because they're one-offs, people aren't making money off them. There's kind of a special place for them in the fair use law. But the mouse house, Disney, the music industry has made this very clear.

You can't just take our stuff. And this is where stable diffusion, open Iowa chat, GPT, and in this case, they're making a legal document. They try, a legal service, they try to get Westlaw, which is a famous Westlaw or Lexis Nexas to give them a license to this. It doesn't matter that you tried to do it. It matters what you did. You can't just take other people's content without permission. And the judge here basically said as much. So in the question of what, what?

was the use because it was both commercial and competitive and not transformative enough. That was a strike against the infringing property, the Ross AI startup. The other two, the next two, so the copyright works nature and then how much of their work was used. Those actually went in favor of Ross. So going against Ross was, hey, this was a commercial use, right? You weren't transforming it. Transforming it would be like Andy Warhol taking the soup can and painting it. that's when you transform the original work significantly, and again, that's like up to a jury, it's up to a judge to interpret that. And it's up to the person whose artwork it is to decide if

what you did was transform enough. That went against Ross. What went in favor of Ross is Westlaw's material has more than a minimal spark of originality required for copyright validity, but the material is not that creative. In other words, what Westlaw has is a bunch of cases. that they have scanned in OCRed and cleaned up and put into their database. So if you're a lawyer and you want to search case law, that's what you use Westlaw for. There's LexisNexis is another. But part four is that this is the money quote, I guess. Even taking all the facts in favor of Ross, it meant to compete with Westlaw by developing

a market substitute. When I talked about this on All In, I showed, if you go on Chat, GPT, and you ask it to make a version of Darth Vader as a bulldog, it wouldn't do that. But I asked it to make a Jedi and it did or a Sith Lord and it did. So somewhere along the line, the folks at OpenAI who have a long history of stealing content, according to the lawsuit from the New York Times and stable diffusion, which got caught with like the Getty watermarks on their images. I mean, talk about red-handed.

They decided, hey, let's take out the mouse house's IP. In other words, somebody said anything that's a character name from Disney, they must have gone to, you know, Wikipedia or DBPD, like one of these sources, and say, give us the character names, put them into our rule set, not the original LM, but the rule set around it, that's like that little thin layer of rules put on top of the LLM. They put a thin layer on top saying, you cannot make IP from Disney characters. It's utterly meaningless, though, in stable diffusion, particularly, because the whole way it works is it's like a, it's a platform, and then people design their own models to go on top. So stable diffusion can say, we don't have Darth Vader,

we're not putting him in our library, but then I, as an end user, can just make my own Darth Vader model, use it in stable diffusion, and now I'm using their platform to make my own Darth Vader images anyway. Lon, is your point that all these models are inherently anti-copyright, and therefore they're all suspect? By the way, this is what Dolly, which is a open-AI thing, they have inside a chat GPT, said it would not create Darth Vader. It said it could try to get close. I said, sure.

And it made me a poodle. A poodle, Darth Boodle. But there's a tie fighter. Yeah. And the Kylo Ren lightsaber. So this is trademark infringing here using that IP, even though it's subtle in my mind. The tie fighter is Lucas's IP. You can't just use it.

And the lightsaber, that iconic one, you see, there isn't just a light sword. If you just did a light sword, that would be different. but it's actually Kylo Wrenz. So they kind of know what they're doing here. And yeah, the difference long to what you're talking about is the difference between personal use, that's non-commercial, and building a service. Chat ChaptiPT charges for their service.

Chad Chilipt has a vibrant, multi-billion dollar business that's based on subscriptions from consumers who want to make images. Now, if you on your own computer, or you at home, were to make your own home version of Star Wars. You literally hired actors, put them in your backyard on your ranch, and made a Star Wars film in your forest on your ranch.

You can do that. You can do all kinds of copyright-breaking stuff on your computer. You could download a New York Times story. You can edit it. You could republish it on your computer. You just can't publicly publish it. And that's the difference. So you will be able to make a Star Wars LLM for personal use.

You just can't use it for competing against Disney. And I believe Disney will make at some point as part of Disney Plus an LLM to do this, which, by the way, YouTube released today for premium subscribers. I do want to say, though, that producer Maddie did manage to get a very convincing Jedi Bulldog made. There it is. So it's funny how thin these barricades are around IP because if you look at that, that just screams I like dogs and I like Star Wars.

and half of that's copyright protected. And that's a licensing deal right there, by the way, Alex. You know, there are people who license Disney IP, Marvel IP, to make ashtrays and saucers and T-shirts. And that's a vibrant part of any IP owners, whether it's Elvis or, you know, with the Rolling Stones. They make very detailed decisions on who gets to license for cups, who gets to license for a fast food head.

happy meal. And they have a bidding war between all of the different fast food restaurants of who gets the happy meal characters for the next Marvel film. And there's so many legal requirements. Like, here's what you can have the characters doing. Here's what you can't. They can't be associated with anything like this. And you lose all that power now. People can put them next to whatever they want. Lon, I think Ferrari has a thing in all their licensing deals that if you show a Ferrari car, it has to be leading the race, for example, because they don't ever to be sure. shown in second place. Exactly.

People are particular about this. They're serious. But what I'm curious about is not just the implications amongst ourselves, but also for the industry itself. And according to Cornell University professor, James, I'm going to guess here, Grimmelman. Grimmelman again, Grimmelman. He's always got something to say. Grimman.

Mr. Grimbleman says, quote, if this decision is followed elsewhere, it's really bad for the generative AI companies. So we are talking about illegal database. as we are talking about Reuters. But the thing is, this does, I think, chip away a little bit at the foundation of the value of some of these companies.

Jason, we've talked about it on the show a bunch. What's Open AI worth? $157 billion, $300 billion. If their foundation is porous, it could be 30. And here's what's going to happen, I predict.

The New York Times is not going to settle, nor should they. New York Times has to take this one to the mat. And if they do take it to the mat, they're going to be able to get an injunction against chat GPT. And if they get an injunction, what that means is

Chad GPT has to turn its product off. They have to stop providing it. That would be cataclysmic. The damages that the New York Times could get here, I believe could be like nine figures, like $100 million, $500 million, which is pocket change for chat GPT,

except when 10 other people show up. And then it's like getting 10 speeding tickets or is having 10 cars impounded. Now you've got $5 billion in liability, $10 billion in liability. That's what sunk, Scour, Napster, and made all those companies become insolvent. Will ChatGPT become insolvent in this case? I don't know. Probably not.

Well, it's going to be, this could be cataclysmic. I think it's a 5% chance or less that this becomes existential for them. And a 95% chance, they write really large chance. And I do think this could balkanize everything where Reddit then says to folks, you know what? Pay up or take our stuff out. And then Cora, pay us, take our stuff out. Getty, pay us, take our stuff out. And it is going to cause absolutely, absolute chaos in the industry, I believe.

And then it's going to cause a really positive. of licensing for the New York Times. I believe the New York Times will make more money, more money from licensing their content to LLMs, then from advertising, and possibly more than subscriptions every year. But so many of the AI leaders have said they can't afford that. Like, if they have to start paying everybody,

but boo, who? Do you not believe them? I don't believe them. Okay. Spotify pays. You tell us, Alex, is it 70 cents on the dollar? 68 cents on the? Okay, that says why Alex is here.

He knows this stuff. You know pop culture, he knows numbers. It's 65 cents of every dollar. You know what? Napster said we can't possibly make this work. Somehow Spotify figured out how to give two-thirds of their revenue away. By the way, Apple gives away 70 cents of every dollar in the app store. Oh, okay, yes, but.

Okay. I mean, Sam outman literally said we need too many materials to train these models. It's not possible. We'd run out of money. Try harder. Before we could pay all of these creators. Very simple. So you just think it's made up.

Percentage. Okay. Percentage. He should take 65 cents of every dollar of consumer subscriptions and give them to content creators. Yeah, but they hoovered up like all the content. Okay. So I'm just saying this is a potential settlement. If you say you can't pay everybody, then you could say I'll pay a percentage.

Okay. And then you come along for the right, which, by the way, is what Spotify did. Spotify figured out a way to say, even though we can't pay you for every song in the system, we can pay you a percentage or your revenue. As it goes out, we pay you. But Spotify can measure that numerically. Here's how many people listen to Rihanna this month. Open AI can't do that with their whole system.

They call everything in. I call BS. They could say in the system, these are citations. When you get medical information, go ahead and do a medical search right now and ask, like, what is, you know, this statin drug used for? Please give me citations. You can ask these LLMs right now, Gemini, et cetera, please give me citations. And they somehow figure out how to get citations in there. So I do know that the model has been trained.

And yes, you're right, Lon. There's probably some general knowledge in there, guess the next word in. email that they can't, but they can also add citations. So I do think there is a way for them to re-engineer the models from first principles to say anything we ingest when it comes out the other side of the transistors, let's make our best effort to attribute it. I struggle with what to do here because I am a writer, right? We do this three times a week. I also love technology. I think startups are awesome. I use AI. And so I just feel conflicted. as hell and I don't even know which way to shoot.

If somebody republishes this and puts ads in it and reconstitutes it, I can't pay my salary and your two salaries. And I mean, that's the whole idea. Fair use isn't supposed to be a new business. It's supposed to be me doing some kind of analysis or some kind of parity or it's not just supposed to be me recreating your business for myself. Now, if you do a picture and picture reaction shot, that would be okay with me. So transformation.

That's addition. That's doing a lot more. A reaction is something new. You're making the content of you reacting. But in the economy of reactions on YouTube, quite often, if you sit down and you put on a new band's new single and you have yourself in the corner

and you're talking over it, often the label gets the money from that. And I think that's mostly okay. Sometimes they cut side deals with creators just to help encourage them to kind of like show off newer bands. But I mean, like, even in that case, YouTube won't mess around. Nintendo, this is always an issue.

with Nintendo. They don't like, let's play people reacting to the new Zelda game, even though, you know, that's how you get the word

out that there's a new Zelda game. Yeah. You can choose to do that, but if you were just taking our show without permission, yeah, okay. Yeah.

So short, that's not okay. Short meta, short Microsoft, because they're in trouble with open AI, long New York Times. No.

Dispicking startups. Investive advice. I think they're all going to win. It's just going to take some settlements. Yeah. But, you know, I just love this one, The Daily Doug. He's a composer.

He smokes weed on air and then talks about songs. I can relate to this. Like, I like this. And he does like my, and I gave him a hundred bucks. I said, do more dire straits. And he did a couple of dire straits for me. And I don't know how much he makes, but the Daily Doug is so great. Oh, this guy.

He's so awesome. Yeah, I've seen this guy before. And if you go look at, you know, it's a really great feature in YouTube is they, if you go on the time. timeline on the scrub. You can see which are like the key points that people will watch again. So pick whichever one is like a P. There you go. There you go. Boom. And he started playing along to dire straits and trying to figure out the chord progression. You can play there. I'm going to play only 30 seconds at most to respect the copyright.

It was fair use of a fair use. And here he's like doing, he is digging brothers in arms, which is one of my favorite songs of all time. Dio Shrates or otherwise. And just go check out the Daly Dug. And if we could book him as a guest, Daly Dug, I want to book as a guest. Okay. And then Blanco Lirio.

Pull up Blanco Lirio. I pay for both of these subsacts. This is bizarre stuff, Flan. But Blanco Lirio channel is so great. And Doug is so great. Delly Dug is so great. Ah, okay. Blanco Lirio I love because I fly planes,

not myself personally, but I fly in planes. I was going to say, like, I sit on planes. I don't want to die in a plane. I don't want to die in a plane. Blanc O'Learyo Channel does an amazing job. You can pull up any of his videos.

Here is a very short clip of him discussing the Azerbaijan Airlines preliminary report. It was, quote, shot down, and here we go. Okay. This guy is a pilot. He's somewhere around Lake Tahoe where I have my ski, and I have my ski house. And it's incredible.

He goes through every step. of every report line. And he actually did the Black Hawk one. And, you know, he will do three, four, five of these reports. Pulls up the NTSB report. I love this guy. I love both these YouTubers. I want to do a collab with both and talk to both of them about their businesses.

Do you know the, there's a TikTok guy who's an airplane mechanic? Do you watch his stuff? No, send it to me. He just shows you whatever he's working on. He just brings the cab. is like, now when you're changing a fuselage cover, you got to... You guys need to play more video games and consume less content. It's fascinating.

I mean, he's funny, but it's just interesting to see how planes work. I had no idea. All right, everybody, another amazing episode of this week in Startup. Thanks, producer Lon. Thanks, Maddie, Chris, Chris Court sales team, Hannah, Jamie, and Maddie Boy. And of course, Alex, Alex. I'm X.com slash Alex. I'm X.com slash Jason.

And the shortest, no, tied for shortest is X.com slash Lons, L-O-N-S. We all have great early year one Twitter handles. You went to South by Southwest and then came back to the Mahalo offices and stood up on a chair and said, everybody, stop what you're doing. Yes. Go download this new app called Twitter. We're all going to use it every day. Alex is like, and nothing's changed in 15 years. He still does that.

No, I was just thinking that I wish I had been there for a little bit because I've known Jason mostly in his like investor era. Kind of like, you know, like aging into the elder statesman role type, you know, I didn't know like Scrappy Jason startup Mahalo. I mean, I knew about Mahalo, but we didn't know what you were back then. I'm going back to Scrappy J-Cal mode. Oh, no. Well, I'll tell you. Well, Scrappy J-Cal with resources. This has been a horrible mistake.

I'm sorry. With resources. It's like, I like it better now. Because I have resources. It's a little bit easier. Everybody gets paid a little more money. There's a little bit more budget for food. We get a nice office.

Maybe there's some money to pay for the old relocation cost. So Lon is moving to Austin. And he will be with me, Jackie, and the team in our new studio. People don't know, but we're building the studio. So I am looking for a 10,000 square foot warehouse somewhere in Austin to put my team and maybe two or three sets of the pond. We keep working on the sets and maybe some new shows and ideas. And then just generally a good time.

We'll see you all next time on the Weekest Arms. Bye-bye.