TWIST News: xAI's massive $6B raise, the startups building our drone future & TWIST500 | E1957

0:00All right, everybody. Welcome back to This Week in Startups. I am your co-host now, Jason Kelacanis. You can follow me on Twitter at Jason. And with me is Alex Wilhelm. Great job on This Week in Tech This Week. And I heard you on This Week in Tech. Twit. Thank you, sir. Leo Laporte Show, where I, if I'm being totally honest, got the idea for the name of this show and took a lot of notes from Leo, who is, in fact, the world's greatest moderator. Welcome back to the show, Alex. Yeah, actually, I think the first time you and I met was actually at This Week in Tech about 17 decades ago. But Leo's been getting people together for a long time.

0:35The good news is we have this week in startups, we are going to be digging into our core topic. And on the show today, we have the new kid on the AI block. We shall talk about XAI's$6 billion funding round. I think they're calling it a series B. That seems to stretch the definition slightly, but we'll get into it. Then Jason wants to talk about robot dogs with guns. And this is actually a conversation about drones, believe it or not. We'll get into defense tech, American dynamism, and what's good. And then I'm sure this will cause no one to get mad. North Korean tech talent taking American remote tech jobs.

1:08So nothing, nothing controversial about remote work coming up on this podcast today. No. This Week in Startups is brought to you by LinkedIn Ads. To redeem a$100 LinkedIn ad credit and launch your first campaign, go to linkedin.com slash thisweekinstartups. 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 sock to report fast. Twist listeners can get$1 ,000 off for a limited time at vanta.com slash twist. And DevSquad. Most dev agencies only offer developers. Why? Because product management is hard get an entire product team for the cost of one u.s developer plus 10 off at dev squad.com slash twist the biggest story i think of the last couple of days is just that xai has raised six billion dollars and some of money that i'm either thinking is is like brilliancy or absolute lunacy and i'm kind of torn between sure yeah so uh just give my quick disclaimer up here i am not involved in this raise i am not an investor um i am friends with a bunch of the folks who are involved i see sequoia and valor um two of the funds i'm very friendly with and involved with are also investors in this so congratulations there um you know six billion dollars seems like a lot of money but i think when we look at ai and large language models because there is a significant cost to the talent and an even greater cost to the hardware infrastructure to compete now table stakes is you know essentially having a billion two billion three billion in compute um and this is why we've seen nvidia stock stock you know skyrocket we were talking about just how blown away you and i were about this story that just is incomprehensible in some ways right that a company could have this much revenue well it's because there's so much at stake and we see it every day when we look at startups both as investors and on the show of just how impressed we are at the velocity at which technology and startups are embracing ai and just really making progress and making things that are useful and have use cases so um if you were going to place a bat on ai there's probably you know 10 people who could make some noise here ilan is obviously one of them what he's done in ai already with tesla you know obviously and he was the original co-founder of open ai open ai is up there as well uh meta aws and google i think those are your top five folks and i think it's a steep drop off but i do think the um falcon project in the uae also is pretty significant the open source one and i think we're going to be in a battle of let's call it six seven eight large language models consolidating down to three winners and i think those three winners and the prize here will be you know something in the neighborhood of the next trillion dollar company and it's a completely non-venture capital type investment here at this scale right this is a this goes like growth equity you know i this is kind of like an ipo in a way right this is kind of like what masa was doing with uber and doordash and other plays like that and so if you put it in that context you know betting on the team eulon set up it's kind of a no-brainer bet there's i'm sure in all these late stage rounds, devices to protect that 6 billion.

4:48In other words, you know, they get their money out first. So if this company became worth$6 billion, those investors would get their 6 billion back. So there's kind of like a floor in there, we call those protective provisions in the business. And I just love the fact that we're going to be in a dogfight for the next five years, between six or seven great companies. and whoever can make this be the cheapest possible technology that reaches the largest number of people and then it's going to be really interesting to see how cheap this can get because right now opening i i think they are on track consistently to lower their prices by 90 percent while it gets better so what that means is every two years alex i'm not like a perfect mathematician here But I think it means like 99 % of the cost plus are taken out of the system every two years.

5:41That's phenomenal. And like Uber and DoorDash and other companies that use capital as a weapon and then turned profitable, I predict a similar scenario here. Okay. What's your take? We're going to go ahead and take a lot of money and we're going to build new foundation models and we're going to train them up and we're going to have a big dogfight, to use your phrase, across major companies. to figure out who can make the best model with the best pricing, the best packaging, etc. Here's the thing. If you go back to the cost of training GPT-3, which is now several generations behind because there was GPT-3.5, GPT-4, and GPT-40 most recently, what's the value of GPT-3 today versus training the garbage?

6:26So the depreciation of models here is huge. So my concern when I look at this huge amount of money, and i get the upside yeah if you win this market yeah this bet will look like the cheapest money ever spent on technology companies right but they're gonna go ahead and take a billion two billion and they're gonna train models what's the shelf life on that investment is there a concern that they're going to essentially keep innovating so fast they're they're deprecating their old investment so quickly that six billion is chump change versus a king's ransom that's my concern yeah i i think you you you nailed it with the latter i think your assessment is correct it's jump change you know it's it's almost like you know what's the value of windows 3.1 windows 95 or like you know the pentium chip versus a chip each of those is a building block to the next right so an iphone 3g if you took it out right now might have some sentimental value but i'm certain if we went on ebay right now you could buy them for ten dollars yes but it was a stepping stone to iphone 15 and certainly to the glasses and you know all those sensors becoming commoditized and you know all the downstream impact of accelerometers batteries i mean what happened in smartphones impacted you know the ev story which is now impacting home batteries we just had a or we're having a company that is putting batteries in homes to get them off grid that's coming up actually i taped it already so i do think they're all building blocks i do think the velocity here is unlike things we've seen before it would be as if the iphone 3, 4, 5, 6, and 7 came out in a two-year span, right?

7:59Right. Yeah. The difference, though, between the models and the iPhones, and I think, I can't tell if I'm splitting hairs here or actually noticing a difference. But the thing is, when the iPhone 3G came out, when the iPhone 4 came out, when the iPhone 4S came out, each one of those had a full year of commercial availability, and I'm sure it made back its development costs and also advanced the tech that could be used later on. My question is, can models like, I presume, Grok 2, when it comes out from XAI, can it earn back its full cost before it also passes the baton to Grok 3? And I don't know, but I will say that not every company has sorted out how to monetize these foundation models yet.

8:38OpenAI clearly is making a lot of money. Good for them. But I'm curious if that will pass along to Grok. Here's a way to look at TAM, Total Addressable Market. There's top-down, there's bottom-up. And then there is kind of like, what I'll call, you know, maybe the way Elon looks at the world, or, you know, like a visionary person might look at it, right? Or a major risk taker who doesn't, you know, have a fear and just assumes, hey, this is what the world will become. So yeah, it's gonna be messy getting there. But we know that it's going to be EVs, right? Or we know that we're going to have a base on Mars, or we know that all the world's information is going to be indexed or you're going to have a store where you can buy anything right so like a bezos mentality or a steve jobs mentality there'll be a computer or a bill gates will have a computer on every desk so let's do not the bottom up not the top down but the visionary risk takers tam this is a new tam that i just came up with in our conversation let me ask you a series of probing questions will every person on the planet be connected to the internet in our lifetime uh without eight people somewhere yes 99.9 yes perfect okay there might be people who opt out so we both agree everybody will be on the internet great and if everybody's on the internet then that means everybody has access to ai navigating the b2b maze can feel really tough huh you're trying to hit the mark with all those top tier executives you want them to pay attention to your enterprise product.

10:13But where can you find all those big fish, the whales, the ones who call the shots and make buying decisions for corporations, for startups and everybody in between? Well, here's where LinkedIn ads is going to solve that problem for you. And I've used this. It is one of my secret weapons. LinkedIn means business. Business equals LinkedIn in people's minds. When you're on LinkedIn, you're in the business mindset. So you're going to really be thinking about business products and services. You're open to those opportunities. And LinkedIn recently passed a billion users. 180 million of those billion are senior executives, 18%.

10:48But hey, we all know about the 1%. 10 million C-suite executives, that's your CFO, CTO, CIO. These are the people who are always looking for a new product or service to make their organization run better. But they are on LinkedIn. That's why LinkedIn's ad platform delivers two to five times greater return on investment compared to other social media platforms. So easy to understand why this is because this is where all the business people are and they're in that business mindset. Super easy call to action. Make your B2B marketing everything it can be and get$100 credit on your next campaign. Go to linkedin.com slash thisweekinstartups to claim your credit.

11:24That's linkedin.com slash thisweekinstartups. No spaces, no dashes. Terms and conditions apply because they're giving you a hundy. So now we say 7 billion people will have a device. smartphone is the most likely and then laptops tablets everything else and actually that's where it's trending right i know internet is probably five out of seven billion people and i think smartphones are probably four or five out of seven billion and something in that range is what my gut tells me you know i haven't looked at the stats in a couple years uh and the reason we haven't looked at the stats is because we are both convinced everybody will have a smartphone everybody will have internet access which means everybody has access to ai and then everybody will use AI and they'll use it more than they use the web or music or YouTube.

12:08AI will be the primary use case of a smart device is my belief. It'll be the primary interface. It'll be the primary use case is not you looking something at the web, but you interacting with an AI agent to solve your problem. So do you agree that the primary use of the internet will become communicating and interfacing with an internet AI agent. Do you believe that? It's the last step in your chain that I'm the least confident in because I'm not sure exactly what form the average person's interaction with AI will be. That said, small models at the edge, aka on your device, on your smartphone, do have the capability to bring this stuff to people's devices.

12:54Strong connectivity or not, like right now on this show, I have gigabit, which is great. but I don't think we're going to see even 10 % of the world on gigabit in the next 10 years, but everyone's going to have a smartphone, which is going to have 3g minimum, hopefully 4g, hopefully 5g. So my question is just what, what is the use case for, you know, a shop owner in rural India for AI? I think helping them access information certainly, but I don't know if it's going to be the principal thing they do with their device. So I want you to dig in more on that. Yeah. So my double click on that is I believe the chat interface or the talking interface will be the majority use case of what people use their smartphones for so you will in the near future and it will happen in the developed world and then move to the emerging and frontier markets very quickly and it turns out chat interfaces actually are low bandwidth right and what we'll see is that person in india will say you know hey um i need to um fix this uh refrigerator operator that's having a problem that's giving a error 72 code what does that mean or i'm going to take a picture of the broken fuse box and say how do i fix this right so just something simple and acute like that which you can do now in chat jpt i literally did it last christmas i was in the ski chalet and i had these bulbs that i couldn't find and i just took a picture of them and i said what am i looking at here and it like nailed it this is the exact bulb this is what it's called And then I just did a Google search for it.

14:29Now, in the future, that would just go get me that bulb, right? And I've done this now countless times. I had an error code on the pool heater. Boom. Same thing. Instantly got me that search. So that will be what they'll be doing. Solving problems in their life, or it might be even, you know, they're learning something. How do I cook salmon? You know, whatever their, you know, standard things that we all do here in the West and that they'll just eat up. So now let's say... Jason, please. Hard pass. Fish is disgusting. Yeah. Okay. I mean, whatever it is. you know what's the what's the proper temperature you know to eat chicken safely right like sure i didn't actually know that and this weekend when i was i was actually cooking pork belly i was smoking pork belly and i needed to just understand how to do that normally i would do google youtube uh or probably my stack would be like youtube then google and this time i just asked chat gpt and then i looked for confirmation of that and it actually cited some sites i know for barbecue you okay got it so then you believe if we just follow this i think 99 of the planet will be using ai so then your total addressable market is everybody essentially like clean water and electricity everybody on the planet eventually it will reach in some form or another so then the tam becomes everybody doing almost everything this is a tam that is very hard to get the human mind around now the overwhelming majority of those things will be free so knowing the cooking temperature of that pork belly or the light bulb that i need to replace or the broken generator and how to solve this problem or tutoring my kids in math i think all that gets abstracted away to free but there's going to be the next set of things the next next set of things that will be monetizable and if you can make a dollar a year or for or two dollars a year off of you know billions of people and then$100 a year or$200 a year in the developed world for 2 billion people, man, the TAM here is, let's call it$1 a year for 4 billion people and$100 a year for 3 billion people.

16:353 billion people,$100 a year, like Netflix price, that's$300 billion in TAM. And then the other$5 billion at, but$1 a year is another$5 billion. There you go. I mean, there is a lot of TAM here. And that's just for basic, basic, basic AI. That's not your robot. That's not your self-driving car. It's not a factory. So this is what I think visionary TAM looks like. Okay. And going back to your request about charts. So in the US through 2022, smartphone penetration reached 86%. And we also have a chart coming up in just a second that will show global internet penetration, which is also rising very, very quickly.

17:15So the underlying element of your view about TAM holds up in that everyone's going to be connected. But if we can rephrase this as clean water, electricity, the internet, which is connectivity, and then AI is intelligence. And maybe what we're saying here is that the TAM for globally accessible intelligence on your mobile device is 100 % of people. Hard to agree with that. Quibble about the monetization, exactly what it looks like, but that is going to work out, I think. And this goes back to why$6 billion is going into XAI, because if you do follow Jason's logic that there are only so many people who can pull this off, you're basically putting a wager down with floor or downside protection to own a big chunk of that eventual future.

17:57I remain pretty AI bullish, so I can kind of see all that work together. I just worry about the shelf life of these models and the cost there. yeah and i think looking at them in the traditional if you look at them as like each individual products i think what you have to look at them is as just versions of the same thing right so it's just this little versioning and it's just weird that in this technology in this technology you build a model it's codified and then you build another one it's codified like i think they'll probably eventually figure out how these things are not codified they're just organic and growing and maybe there's like um you don't remember they sharded databases we had like one big database and like this database oracle database will like last forever and they're like yeah and then everybody started recording every click on the internet and every interaction i was like oh yeah that's too much stuff and oh by the way it's coming in in real time and we want to watch it in real time it's like okay we need to shard this database You know, oh, there's a billion people on Facebook sharing, you know, 10 billion items a day.

19:04That's a different use case that we didn't anticipate. I think they'll shard these things. Yeah. So here's internet penetration from 2005 to 2023. Going up to 5.4 billion, which going to our idea of there's some folks who are going to opt out. There are also babies and there are people who are very, very senior who may not be interested. You're getting pretty close to 90 % of eligible folks. i think i said five billion was my guess so i think i'm pretty close yeah i must have had that stored away somewhere and you know listen it it looks like from this chart you know hundreds of millions of people are still getting on each year but it will flatten out and so um they're getting if there's 5.4 billion people internet users worldwide that means there's that many computers and smartphones if you combine them and did some de-duping or whatever i mean at some point we'll just be counting the people who aren't online and i think you're right it's going to be old people babies and people who are hippie dippies who are just going and homesteading and they're getting off the grid and they don't want to be on here so those are the hippie dippies and the preppers which are oddly enough becoming besties because i grew up amongst the hippies and i now know some of the preppers and they share quite a lot here's here's where i'm thinking next though in terms of vetting the spec so we know that xai put together grok one uh november 23 grok 1.5 march of this year they open source the first grok i had i played with grok this morning in anticipation of talking to you it is still a very um chatty nascent nascent you might say assistant and so i'm curious like if you're an investor in xai you're part of the six billion dollar round let's say you put in a hundred million of it, how fast do you expect to see XAI start dropping things that push other AI companies to increase their own cadence, to make them dance as Satya?

20:55Yeah. I think, you know, maybe they're a year behind and they close, you know, three months every year. So in two or three years, they're at parity. And here's a way to play this. If we know AI is the greatest sea change and technological shift in our lifetime, and that now is the inflection point. the easiest way to play this if you were a sovereign wealth fund if you were a venture firm if you were a high net worth individual we need to place you know seven bets on the top seven potential players because if there'll be three players that are the majority of it'll probably wind up being just like technology you know with the fang and you know other variations there you know the four horsemen whatever you know jim cramer dubs them you know this year there's gonna be um you know like five winners let's say six winners seven winners and there'll be like two that are like big winners where else would you put your place your bets you're gonna place your bets on mcdonald's you're gonna place your bets on you know united airlines like you're gonna place your bet on whole foods well actually that's owned by amazon you're just gonna this is the place i think smart money is going to place its bets that the highest growth economically the highest economic value will be created in ai there's no other space that i think there'll be more economic value created in our lifetimes or certainly in the rest of our lifetimes in ai so i mean i'm almost putting a j trade in here like i think maybe when i rebalance my portfolio next year i'm going to kind of look at that with my family office i think you know i might just put you know 80 percent of our net worth in the top five players and just call it a day and then have 20 percent in like dividend paying whatever or maybe 50 in those five and 50 in dividend paying you know safe stuff because i do think there's no there is no precedent for how big these can get very similar in that there was no precedent for apple and the smartphone revolution like that was something nobody saw and i think like amazon nobody could kind of figure out that books would equal everything eventually so i think this inflection point means and this is not investment advice but i like to share my thinking on placing bets.

23:14I mean, what's going to outdo these things? Like the gap, you know, like LVMH, like Disney, like, I don't think so. I don't think any of those are going to provide as much value to humans. And we have a consumer driven economy, and a corporate driven economy, consumer to those corporate government, maybe one third, I think ballpark is how you break the economic value up in the world. You know, thinking a lot about what you just said, one of the cool things about, well, you can view this as concerning or cool, is that if you buy, you know, just Microsoft, Alphabet, et cetera, you're getting essentially options on their investments in Anthropoc, OpenAI, et cetera, et cetera, et cetera.

23:52And they pay dividends and they're dropping tens of billions of dollars a year to rebuy their shares. It's kind of hard to not do that. And if you want to know how much you invest in those companies, everybody, just go ahead and look at breakdown of your index funds in your fidelity account and you'll see that the top five holdings are well those companies so yes for a reason yeah for a reason and they've done well for a reason um last question then we're going to move on if you're xai six billion dollars in the bank you're going to go hire a lot of people do you invest in your own infrastructure hardware h100s nvidia stuff or are you going to go cut a deal with aws azure whatever to get a big bucket of gpu power i think the answer to that is yes you're gonna get whatever compute you can get right now and then i think the question you didn't ask was are you gonna make your own silicon and i think you got to make your own silicon i think you have to look at this and say um how do i not have an nvidia dependency and how so it's like how do i leverage nvidia today and how do i not have an nvidia um limitation in the future and if you're going to be in an arms race you better make your own goddamn bullets uh i think you need to have a bullet factory you need to have a munitions factory yeah because the person who owns bullet town if you saw mad max this weekend futurosa i loved it um i haven't yet because my child's too little but yes like i still was in dune too but i can't wait i took my 14 year old so you know so if i'm a bad parent let me know but she loved it and um you know there's just there's bullet town there's gasoline town and then there's uh the citadel where they have water and vegetables and these you know three people are kind of moving stuff around and that's the fury road between them to you know try to keep civilization working you know in this case i think you know you you don't want to be dependent on one of those towns you need to have your own water supply you need to have your own gasoline supply in this case, that's CPUs, GPUs.

25:55So why not just buy a startup in the space or start, you know, your five year plan is to have your own silicon roadmap, like Apple did. Listen, a strong sales team can make all the difference for a B2B startup. But if you're going to hire sharks, you need to let them hunt and you can't slow them down with compliance hurdles like SOC 2. What is SOC 2? Well, any company that stores customer data in the cloud needs to be SOC 2 compliant. If you don't have your SOC 2 tight, your sales team can't close major deals. It's that simple. But thankfully, Vanta makes it really easy to get and renew your SOC 2 compliance.

26:31On average, Vanta customers are compliant in just two to four weeks. Without Vanta, it takes three to five months. Vanta can save you hundreds of hours of work and up to 85 % on compliance costs. And Vanta does more than just SOC 2. They also automate up to 90 % compliance for GDPR, HIPAA, and more. So here's your call to action. Stop slowing your sales team down and use vanta get a thousand dollars off at vanta.com slash twist that's vanta.com slash twist for one thousand dollars off your sock too remember when apple kind of got off the intel uh you know train they just steve jobs realized like we're a computing company and we're giving these folks three hundred dollars per laptop for their chips five hundred dollars per desktop for chips what if we made it ourselves we took that margin and when you buy an apple computer the m2 the m3 that's like a massive competitive advantage they have there they design that chip based on what they studied in our laptops and what they studied in our laptops was people keep a lot of browser windows open which uses a lot of memory which destroys your battery so they optimize the m2 and m3 chips to last forever the battery and last forever it's optimized for browsing and video it's pretty slick.

27:44What people use them for, if they are going to make their own bullet factory, if they want to own their own munitions pipeline by making chips,$6 billion is a down payment on this company versus a conclusion of its fund rate. In that case, my next, I guess my question is, who's going to drop the$10 billion they need to build up their own... That's presuming they go fabless, and they just design their own chips. If they want to build a fab, I mean... There's an unlimited appetite to invest in these companies and to invest in that founder. So I don't think funding is the limitation here. I think talent and strategy are the most important things.

28:24You have to make very important strategic decisions right now. And, you know, even with Sam Altman and OpenAI, they've made some brilliant strategic decisions, and they've made some very stupid, you know, decisions that they've tried to unravel. And that just shows you and you know, how important strategy is at this level of the game. Zuckerberg's like, we're going to open source this. And we're going to put a search box at the top of everything. You know, a decade, a decade ago, Google was like, we're gonna buy DeepMind. And then open as like, this thing is just worth too much money for it to not be a for profit company.

28:56So let's have our cake and eat it too. We'll be able to virtue signal that we're doing this for, you know, great reason for humanity while we're selling 2 billion shares in the secondary market at 90 billion 80 or 90 billion dollar valuation like the most like incongruous behavior you've ever seen and you know they figured out a way to thread the needle so far with that i have a feeling the irs and other folks are going to come down on open ai like a ton of bricks that is not based on inside information but based on what we've seen with the drama you may have seen this new uh ai open general hospital what we did all in the bit like i mean all that drama i think the sec is already looking at it and i think irs is at some point going to be like wait a second under where where where are the taxes being paid under what shell yeah you know the game with the shells oh no here's the value and then also the revenue round tripping from major tech companies investing in these ai companies that are then spending it back on azure gcp etc there's a lot of yes it's essentially that crazy guy with the whiteboard and the and the red strings you know that's what these companies look like it's it's a mess all right but let's move on and talk about the twist 500 jason this is a new brainchild from you that we're going to be working on here on the show also in the twist ticker over on substack and i think it's gonna be a lot of fun but i want to just tell people what we're doing by starting with the genesis point from your mind yeah so you know um i have uh you know been in journalism for a long time like you and in my first magazine silicon hour reporter we did an exercise every year the silicon alley 100 and it was pretty famous first year i could only find 60 people who worked in silicon alley so i just asked those people working in new york on the internet like do you know anybody else who has the internet and they were like my lawyer or my headhunter and We literally include lawyers, ad earners, and PR people on the list just to get to 100.

30:55The next year, we had 1 ,000 people actively fighting to be on that list. And then the one thing that keeps happening over and over again in our industry is trying to make sense of the private markets, right? And you spent all this time at TechCrunch and at Crunchbase trying to get a handle on private markets. And we quote data from Carta, PitchBook, Crunchbase all the time. All of that's great, but there's too much noise in the data. And the data also does not have a curator. And it doesn't have an Alex and a Jason saying, okay, we know there's 50 ,000 startups in Crunchbase that are active in some way, right?

31:34What is the Crunchbase database up to in terms of entities in there that are private? It is now enormous. The question is how many are still active, to your point. I don't know the ratio, but it's a vast ocean of data. And so it is not helpful. crunch base is not helpful in that way to understand like what who are the most likely winners in each category right yes and so what i want to do is for my own understanding and for the topic of this show for you and i each week to have a database of companies that we think are notable enough to track now we also have what i used to call the launch ticker which is now the twist ticker it's an email newsletter people are addicted to it it's got subscribers it makes a little bit of money um but news has become commodified as you know in many ways you can get the headlines a tech meme you can get them on social media ai companies are now abstracting them i don't think the headlines are valuable anymore i sadly don't think people read the feature stories anymore they just get the bullet point summary they see people debating it on social media whether it's thread master don or twitter x and they're done right i think the majority of people are done and they don't click through to the original sources anymore you can debate all that tragedy but we are in a unique position you and i with our finger on the pulse of this you as somebody who worked at crunch base and tech crunch and then me as somebody who's an investor in this space and been doing 2 000 episodes of this to every week say um let's pick a topic um military tech and weapons right yeah now everybody off the top of their head would say andrel okay great we know andrel's valuation great there's one of the 500 and if we were to say who are the top 25 defense tech companies you and i wouldn't be able to go that far and then if you went to crunch base you might get too much information so what i want to do is you and i say you know what this month we're going to work on the defense tech 25 the construction 25 or by stage the you know y combinator 50 the tech stars 50 or maybe the mina middle east north africa region these are the mina 50 these are the japan 25 these are the australia 25 and they would obviously overlap so yes if we said hey these are the companies in ar that are important we're going to keep 25 on the list in order for when we have those 25 fully done for the ar companies in order to put another company on one has to come up yes you and i then have that dialogue why is this one coming off why is this one going on now i'm not doing this to tweak founders or investors i'm doing this so we have a deeper understanding of who's doing the best work in the space so we're going to have a website twist 500.com the reason i picked 500 100 ,000 is too much, and 100 is not enough.

34:32I think that's absolutely right. Now, we're going to have this list be public, so people can go and see this, and they can take a look at it if they want to, and that's going to hopefully invite some debate with readers, listeners, and subscribers, and critiques. The thing is, this is a great way to just get through, as you mentioned, the noise. The struggle that I had at TechWorks, the struggle that I had at Crunchbase News, was just the sheer number of companies out there right now. Every time YC does a class, there's a couple of hundred companies. You know, you want to love all your children the same, but when there's so many of them, you can't even shake a stick at them.

35:06It gets tough. And if you're external and you don't get to do this all day, like you and I do, you probably want a little bit of a guy. So I think this is a very good idea. And also it made me think differently this morning when I was writing for myself, I was like, what's the coolest company I'm seeing this morning? What company would I bring to the Swiss 500? So I think it's already changing how I'm thinking about things. And I think that's a great way to kind of keep an eye on startups, because then it's not just news items and headlines. It's, is this the next standout company that we need to bring to the This Big in Startups audience around the world?

35:39And I think it's a fun intellectual exercise, and it'll force us all starting to cast a wider net. So I don't see any downsides other than, I mean, it's going to be some work, but like, whatever, that's going to be fun. i the way i look at large projects is i like building in public i like having the debate in public and i always you know just from my career you know if you told me you're going to be hitting 2 000 episodes of this podcast i'd be like are you nuts yeah but what happened over time was i i looked at and it's like you know there's 100 people every year i'd love to have a lunch with and i have to eat lunch 365 days a year so one out of you know 3.65 lunches i'll record that literally was my premise when i started this week in startups was it's not a bad one yeah you know i said i gotta have lunch and instead of lunch in that case it was 50 a year and then i was like you know what there's kind of more people i want to talk to and then i like the discipline of talking about the news every day what people don't see behind the behind the curtain of what you and i are doing here and what you did at these other publications is you know when you're preparing a docket i call it the docket when you know the people call it the show notes but the docket requires you to make editorial decisions you know i said i'm interested in this weapons story uh you said hey the xai story is the big story of the week i was like oh yeah thanks for reminding me and then we kind of negotiate back and forth and build a docket the audience tells us hey you missed this okay great you know uh we'll we'll get that on the next episode so this is if you think about it on a pace if we can put 10 companies into the database the twist 500 per week we'll be done in a year it's one year's worth of work and we can just go back and forth and ask each other because i want to do you're here for three days a week it was our deal so if we do three new show programs a week together which is what i'd like to get to by the fall you know just every show hey here's three or four companies i'm thinking about adding and then we just have a little debate about them or you know here's five that people are suggesting, okay, let's add them.

37:39And then the hard one's going to be keeping track of them and taking them away. But once the list is established, we will have somebody who says, hey, you've got these three companies on the list. And a venture capitalist will call me and say, listen, you have these on the energy list. You left off my portfolio company and my portfolio company has 10 times the revenue of that company and half the number of employees. Yeah. And then you and I will get smarter. Yes. And then we can share that with the crew. So if you want to follow along, the URL is ticker.thisweekandstartups.com. You can snag the newsletter there.

38:14It's going to be showing up there. It's going to be on the show. It's going to be on its own website. So expect to hear and see a lot about this in the upcoming weeks. We'll remind you. Hey, take a moment. Picture the ultimate all-star team for your startup. You got that image? Maybe you're thinking about the Avengers, huh? Maybe the X-Men, Wolverine, Cyclops, Iron Man. Well, let's be realistic here. Until you've raised your Series A or B, you're going to need help building this A team, right? You may not have all the resources to do it. Finding talent, managing all these timelines, and ensuring quality is a huge task for anybody building a product.

38:45And scouting for the perfect all-in-one 10x developer, it's going to take a lot of time. Well, there's somebody out there who does this for a living. Skip the hassle and power up with DevSquad, okay? They're like Professor X from the X-Men. They just know where all those great mutant developers are, those really powerful ones. DevSquad offers a complete product team loaded with top-tier talent straight from Latin America. Same time zone as us, right? Your team is going to have two to six full stack developers, depending on what you need, a technical product manager, keep everything tight, and a bunch of specialists.

39:15Maybe you need UI and UX design. Maybe you need DevOps. How about quality assurance? They're all going to be synced up in your time zone, and it's going to cost 75 % less than the equivalent team in the USA. We all know there's an incredible arbitrage. There's talent everywhere in the world. And now that we all work online, remote, hybrid, whatever it is, you know that you can integrate this amazing DevSquad team into your team. If you're looking for seamless collaboration and the flexibility of monthly payments without long-term commitments, it's time to take action. Head over to devsquad.com slash twist and snag an exclusive 10 % off your engagement.

39:46That's devsquad, D-E-V-S-Q-U-A-D.com slash twist. It's time to squad up and transform your startup with the all-star team you deserve. Let's talk about that defense story that you mentioned because I got to, while prepping for today's show, learn a hell of a lot about the world of defense tech. and it turns out it's uh way more interesting and monothematic than i thought but the the story jason is about a robot dog from china with a gun so start there uh all right so here is a robotic dog it looks like yeah what's that looks like cgi that looks like a machine gun like an assault rifle so it's an assault rifle it's going into like a green um you know training exercise building that is made of wire mesh and you see like this it goes in first basically and it's firing live rounds um and the person who's controlling it has an xbox controller so this is terrifying and this will be the next terrorist attack in america sadly or somewhere else so this is you know the thing of nightmares and china i guess is showing this and showing a drone whoa i didn't see the drone i the drone with the gun so there's a drone with a machine gun on it i mean it's super weird because you would think it would just build an integrated gun into these devices but instead of strapping one on which makes no sense to me if you're listening to the audio version when we say it has a gun on it we mean they literally have like a human sized and set up gun on board and it looks like a drone carrying a human's weapon versus a weapon made for the drone or the dog in this case which is a strange way to go about it but then again we have figured out as a species how do you make handheld weapons we are very good at that so you know i can kind of i can kind of see it you went down the rabbit hole here somebody showed us americans have this too i've been watching youtube videos of people putting paintball guns on drones for a long time so i think it's illegal in the united states for you to put a gun on a robot i don't you know we're kind of in a new area here where like you know should your kids be able to chat with chat gpt and should they have an unfiltered chat with gpt like this came up recently i was talking to some parents like there are some spicy you know ais out there and kids are talking to them should they be allowed to do that well we are some people think the model should be totally unfiltered and some people like anthropic think they should be much more controlled and i think we're having that debate right now but i i will say when i was growing up i grew up in rural-ish oregon gun safe shooting guns off the back porch paintball course in the field type life um i don't think we would have looked at the robot dog with the dog with the gun and worried about it much but after living in large cities for a while it it does feel much more dystopian and dangerous because i'm now around folks in much closer groupings and it makes me it makes me fret quite a lot actually yeah i i wouldn't fret too much but i wouldn't be naive to exactly how dangerous this could get and you know uh the whole concept of terrorism is to scare you built into the name and so this is released by china to be scary for americans um and one wonders if this is to amp up military spending or to be like a thread about taiwan who knows what the motivation of these being released at this moment in time but look here's a paintball gun you know when is this video from seven years ago somebody mounted a you know a paintball gun on a drone i mean this is not like it's not rocket science uh to do this and then if you look at what's happening in ukraine um you know i was watching a video came across my feed of them putting a grenade on a commercial drone pulling the pin and then they have this little plastic sleeve that goes over the grenade that kind of holds the pin in and then they just go over a person and then i guess they have a mechanism to open the sleeve up and it just drops the grenade do you know what's really ironic about that exact setup yeah go back to world war one and the advent of plane based reconnaissance in war zones we had you know uh hot air balloons before yeah but you know like small aircraft i think that was the way they started off doing bombing was just dropping things off the side and literally out the window yeah and they had handguns they would shoot at each other this is before we learned how to shoot through propellers and so forth yeah and now we are back to that just with a bunch of smaller aircraft drones if you will and that's why i think when i was going through the what are the hot defense tech companies out there they're almost all as i can tell focused on either drone warfare or drone warfare suppression so everyone's trying to figure out a way to like essentially um neutralize yeah exactly and this is not a small thing so like um allen control systems 12 million dollars craft ventures drone warfare shield ai zero eyes apyrus fordham technologies like there's a bunch of people who are doing lots of different ways to approach getting rid of drones that could have guns or grenades on them because i think everyone's thinking about you know it's hard to send humans they need to eat drones just need juice yeah and you don't need to recover the drone if it gets captured they just explode it's not like a navy you know if somebody one of our soldiers gets caught we you know we'll spend decades trying to get them released right and we have absolutely and so i guess my question to you is what's more dystopian sending young people into war or sending a bunch of robots to kill each other on a beach somewhere oh so as long as it's robots and robots totally less dystopian in fact at that point it becomes nearly a video game i it would be great if if international conflict came down to who's got the batter robots and we settle it far away from population centers I think people get a little worried about the dystopian element of it, that it allows warfare to be done against humans more cheaply without the loss of life on the attacking side in that construct.

46:07but i i think looking at what andurl's doing with both autonomous vehicles in on the ground in the water and in the air we are moving towards a world in which carrier groups and you know amphibious landing squads of humans are passe they're the cavalry charge of the modern era and so basically it'll be it'll be battle bots like wasn't there a tv show where you put a bunch robots in a ring and they just had like chainsaws on them 20 years ago so basically modern warfare is going to be battle bots what was the one that craft invested in uh david sacks i think it was allen control systems i believe yeah i mean this is the question going around silicon valley like is it ethical moral noble okay writ large pick your word um to invest in defense stack and i'm like if you're an american and i mean there is there are personal decisions to it if you were a pacifist yeah i could see you saying i don't ever want to be involved with that i don't want to spend my time on it i can get like a personal decision like that but i would never say to another investor or entrepreneur building defense tech is immoral if i am benefiting from the security the united states has you know helped um establish globally and yeah and our borders and as an entrepreneur you are benefiting from democracy here and freedom which is tenuous so i don't want to be like all team america here but i don't understand the argument i understand the personal argument i don't choose to spend my time on defense tech i just it's not my thing or i wouldn't be able to sleep well at night but how do you look at that i wonder i'm curious you're allowed to give your opinion now because you're not a tech crunch and you're you're not you can really give your opinion here in fact i encourage you to get as based as you want yeah tech always had me on a on a relatively long leash oh yes now i have a leash nonetheless a couple extra links in there um my brother's in the army he's uh he's a major i'm really proud of him um he was a tank commander a couple tours in iraq and now he actually is a professor at west point and just finished his PhD in amazing what's his name Andrew Andrew thank you for your service on behalf of something physics I don't know what he does lasers who knows I'm not good at that sort of stuff fantastic uh but if someone came to me and said look there's two features ahead of your brother in the army and one of which he is better protected because we built some cool stuff and one of which he's not I know exactly which one I'm going to choose now okay that is not carte blanche though I do think that there are some companies that for example want to um compromise smartphones of dissidents and journalists and so forth sure so to me spy companies they okay so yeah there's a privacy thing where it's you're you would be okay hacking a terrorist phone but you don't want americans phones hacked and so that is something i think most people would agree with i think we need regulation on compromising devices remotely i would say even for the u.s government i i don't think just because i'm part of this country that means that i don't care about anyone else but if you want to look at how the technology is used look at mexico and saudi arabia and india and so forth and you'll see how china is the number one right yeah well china is almost a unique do they compromise phones in china or do they just like well yeah i mean i i think you're right i think the expectation in china is there was a really interesting story in the new york times this weekend about not just forget about the technological surveillance they've actually re-established like a stasi level so elders in the community like sit on the street corners and they visit apartment buildings and they ask people you know hey who's arguing or is anybody talking about these topics and then they go visit them and they're not police they're your neighbors yeah they're essentially yeah so china is a unique thing and i don't i think this is called the i'm trying to google this my chinese is so bad i'm not going to try um it's a system by which you have very much top-down government control i'm thinking about free societies democracies effectively where you would hope that people could be a dissident could be a reporter and so forth and uncover this and not have their privacy compromised china the chinese model the russian model a little bit different you know if you go to the uae i don't think you expect your phone not to be hacked for example yeah i mean i think uae is pretty advanced uh in these regards um because they want to be a safe having spent time there um they want to be a safe haven for international business so they've taken a more american approach to it you're totally free to do whatever you want um probably not a great idea to criticize the royal family or anything like that in those regions um and be respectful like it's almost like it's like a respect thing and i don't think they're hacking everybody's phones i think if you were doing something then they would do it and they wouldn't have a problem with it just like our government i i also think you know i give tim cook and apple a lot of credit for um you know saying hey listen we're going to build the phone for privacy yes and we are doing that for humanity and for people who buy the phone um but when you come to us and you ask us to break that we're sorry but we don't have the keys please and i have to talk to the person who owns the phone and you know also at the same time tim cook gets to limit the ability for arch rival android owner google's parent company alphabet to invest even more information about us so it is it is one of those twofers it's good for the people and it's also good for the apple shareholders and yeah last thing on this and then we'll talk about north korea i never got vice clauses for investment like i i mean why can't you invest in a craft booze company or a craft weed company like i just oh yeah so what it turns out is um a lot of the endowments and family offices are old school old school and so they don't ever want to be in a situation um where there is an embarrassment potential because you might have at the top of the hierarchy high net worth individuals let's say a family office so grandma and grandpa you know are like we invested in weed we invested in psilocybin you know therapies you know and their grandkids are like yeah we're doing psilocybin therapies like you know your your daughter and your cousin took it and they got rid of their ptsd doing some you know uh psychedelic therapy you know whatever and so what they have is just these blanket things that are meant to protect the downside embarrassment or chaos that could ensue from harvard's endowment invested in magic mushrooms would be the headline you know uh and then some great headline i would love to read pretty good yeah i mean i'm assuming in the 60s and 70s i think there's probably been more magic mushrooms consumed on the harvard campus than many other campuses and a lot of elite rich kids with a lot of time on them yeah there's no cocaine at harvard at all there's never been absolutely it's been a dry campus forever um but yeah they just don't want to have egg on their face is basically what it comes down to that that is limiting though to me because there has been so much money to be made in all beverages for example i mean both alcoholic and non-alcoholic and so it always seemed oddly prude to me you know we're going to be investing venture capital but only in these narrow areas and you can't go outside and know having fun yeah and the second piece that comes down is for vcs those areas typically haven't had the highest return so it makes it really easy for them to be like yeah we don't touch gambling we don't touch alcohol we don't touch drugs and we don't touch porn uh sorry we can't even take the meeting we're done they can just move on with their lives right but now when the gambling regulations have changed now you can start to look at that and now also remember some of the money might be coming from a sovereign wealth fund that has specific religious beliefs it could be coming from you know a family with specific religious beliefs so if there's a family that's catholic and they don't believe in certain reproductive sciences and you're we have an investment in a company control line which is doing um uh reversible vasectomy you shoot a gel into the vas deferens and for you know some period of time five seven ten years the swimmers can't get through and then if you decide hey i want the swimmers to get there and i want to have the joy of being a parent they just stick a little needle into your vas deferens uh you know that that is the hard part yeah i think they they numb it out uh sort of graphic and then they give you um another uh hydro gel that releases the blockage and allows the swimmers through so imagine you're a young person and you say you know what i'm not going to be making babies in my 20s i'm going to save that for i'm 30 and you could on the way to college your parents could get you a reversible vasectomy when you're 18 going off to college and then say you tell here's your doctor's information go here when you're 25 30 35 and you want to have kids and then you so you know that's one where if i had an lp who didn't want to be in reproductive science i wouldn't be able to take yeah i'm just trying to imagine that conversation with my parents on my way to college when i was 18 i think i would rather just cut my head that I've had that with them.

55:45Alright, let's move on to North Korea. Last story. So, essentially, the gist here is that there are people in the US who are using American identities and VPNs to help North Korean nationals work for remote tech jobs to make money for the North Korean missile program. One more time. That's a true statement. Okay. You gotta go slower on this one because this is crazy. There are people who've been charged in the United States. Charged? Yes, the government has said you. Federal prosecutors charge an Arizona woman and four other people, according to Axios, while helping North Korean nationals use American IDs and VPNs to pretend to be Americans working remote tech jobs.

56:34and they got 6.8 million dollars in revenue at a minimum from this program to help fund the north korean missile program so yes some fortune 500 money some big tech money has gone not defeating north koreans what to a third rate missile program from a 10-pot dictatorship and okay this is the power of the connection power of the internet all right so let's see who will be the most outraged as everybody i think i think yeah so there's so many ways to respond to the story i guess the first i have is how did they get the jobs without interviews or had to get through the interview process so they got somebody's credentials some american person found credentials for them and did that american person who spoke fluent you know not only english but also python or whatever the job they were going for then hand the job off take the interviews for them and then hand the job off so i came on and i'm assuming it would be a korean person sort of some korean americans said i am you know jason kim and uh yeah i went to mit and i'd love to have this job i interviewed really well i got the job and then i gave it to you know bill kim in north korea is that what happened here is that what they're saying in these allegations because that that to me seems like incredibly clever and insane yeah so axio says that generative ai tools have also made it easier for north korean it hires to craft believable resumes we also know that with generation of AI, you can do lots of stuff with video.

58:27And then I love this anecdote. In one case, a worker created a resume that claimed they had worked at Amazon and Meta. They aced the interview process and technical test. The only red flag was the person would take about 30 seconds to answer a question. So essentially, you can't solve latency, but you could just say, hey, I'm at the chalet. I'm at the ski get together. So sorry that the internet is not great. I'm at Coachella. Right, exactly. But I do think that remote work does depend a bit on trust. And we've talked about, you know, corporate culture at launch and so forth and how we handle that.