New Apple Products, Anthropic's Strategy, Why AI Costs Don’t Hurt Apple | Dean Ball, Scott Kupor & Jared Isaacman, Adam Bry, Matteo Franceschetti, Dillon Rolnick

10:24Matteo Franceschetti:We're going to serve Gemini.

10:26Dillon Rolnick:We're going to serve Gemini. And we'll pay them, what, a billion dollars a year or something like that? Yeah. It's like.

10:30Matteo Franceschetti:Yeah, I mean, this year will be really interesting. It'll be fascinating to see what the new Siri actually looks like, what it can really do. We'll start to see that, I think, in Q2.

10:40Dillon Rolnick:Yes.

10:40Matteo Franceschetti:And then separately, we're expecting new devices from OpenAI, right? Oh, yeah. Teasing the puck, time, whatever you want to call it. Yeah. It'll be interesting to see if they can kind of break the AI hardware curse that I feel like is...

10:57Dillon Rolnick:Well, I wanted to think about that because I wonder if the story's over. for some of those like cursed bad launches product. Like we saw Friend apparently has found a little niche, very controversial, but Humane has sold to HP. It seems like they're not working on that product anymore, the pin, the AI pin. But the Rabbit R1, I think his name's Jesse, that product, there's something there where with enough iterations, I feel like that might sell well. I was playing with the board the other day. that hardware it's like a huge screen touch screen that you can play games on like there's still these niche applications that i think you could potentially if you have a hardware team you have the ability to sell something it's it wasn't polished the ai wasn't good enough well the ai got better you still have the supply chain maybe there's a way to cut through but it probably needs to be a little bit more niche yeah like everyone's gonna have one tyler what do you

11:53Scott Kupor:think about hardware in the age of ai i mean i i think the mac mini is like a good example of like AI hardware right now.

11:59Dillon Rolnick:Yeah, no, that's a great example. Yeah. I wonder what, like how exposed the Gemini APIs will be in iOS. Because if you fully expose like Gemini kit, like, you know, in iOS, just like you can call like a weather API or a maps API, like Uber exists because there was like a GPS API that you could just call within an iOS application. with a couple lines of code. And you didn't have to like figure out where the user was because the iPhone just told you that as long as you clicked like, yeah, share my location with Uber. If you now get an app that has not just Apple's local LLMs that are sort of, you know, mediocre, but it can go and call Gemini and do a whole bunch of crazy stuff, you start running into like, will someone build an open cloth app or something?

12:58Dillon Rolnick:like how much of Apple's AI strategy will wind up being vended into the third-party ecosystem, which has served them so well for so long, they've sort of turned their back on it. They locked down Apple Vision Pro a little bit more. There was way less uptake there. And there's a ton of risk because if I develop like a free iOS app that then incurs like a ton of Gemini tokens and ton of charges, how do those get passed through? Is Apple just biting the bullet on it? But there's something interesting there where we're turning app developers loose to To build software that takes advantage not just of local LLM capabilities, but Gemini under the hood And and sort of wrap it up in a neat bow that is, you know, available to the consumer could be could be very cool anyway, the big question that I had and I will get into the essay was So Apple's unaffected by capex costs, but they've also with this release been unaffected by memory costs and the Ram-mageddon.

14:00Dillon Rolnick:So I will take you through how I dug into the question of memory pricing at Apple. But first I will tell you about LabelBox. RL environments, voice, robotics, evals, and expert human data. LabelBox is the data factory behind the world's leading AI teams. And I will also tell you about console. Console builds AI agents that automate 70 % of IT, HR, and finance support, giving employees instant resolution for access requests and password resets. That's a good sound effect. I like that one. I have one too. Five! Surprise.

14:32Matteo Franceschetti:I can't believe you haven't used that yet.

14:34Dillon Rolnick:No, no, no. Because I was so burned by the last time I got access to the soundboard, and I went too crazy. So there was one show where me and Jordy both had full soundboard access, all of the same sound cues.

14:46Matteo Franceschetti:One single show.

14:47Dillon Rolnick:And it was truly... A disaster. What's the metaphor from Greek mythology? It was Pandora's box for me. Once I opened it, I could not resist. And I became addicted. And I was pushing the thing way too much. It is very addicting. And it just didn't work. So we were like, no, let's have Jordy play the soundboard. I'll do other stuff. We found a great balance.

15:08Matteo Franceschetti:The issue for me as a soundboard artist is when I'm not doing the show, when I'm hanging out on the weekend, Yes. I'm just thinking, if somebody says something, like we're at dinner, I'm thinking about the sound effect. Yeah.

15:25Dillon Rolnick:I was hanging out with you and you said something crazy and I wanted to just be like. And so I opened up my phone, I went to YouTube and I pulled up that sound and played it. But then I had to turn on my volume and it took me way too long and the joke had passed. But we have a solution. Tyler Cosgrove over there has been working on a soundboard app that can go on your phone and you can have, we should do a widget too, so that it's just there on the home screen. Boop, your favorite soundboard cues. So expect that soon. Anyway, let me tell you about Apple 11. Profitable advertising made easy with Axon.ai.

15:58Dillon Rolnick:Get access to over 1 billion daily active users and grow your business today. So the question, like I get the strategy of Apple saying, look, we're good with Google handling the AI inference stuff. They have GCP, they have the TPU, they got Demis, who's got a book written about him by Sebastian Maliby. He, like, DeepMind is great, we're working with them. So, CapEx unaffected, I understand that. How are they not affected by memory prices? Ramageddon has come for us all. And Ben Thompson of Stratechery and John Gruber of Daring Fireball actually had a bet about this. And I hope I didn't misspell John Gruber's name.

16:42Dillon Rolnick:They had a bet about this. They said, whoever wins this bet gets a steak dinner. A steak dinner was at stake and Ben Thompson lost. John Gruber won because John Gruber said, Apple always overcharges for RAM. Getting a, if you went to build a computer or a phone or laptop or, you know, iPad equivalent, and you just went to retail, getting an extra 128 gigs of memory or a slight memory upgrade might cost you 30 bucks. In the Apple ecosystem, that's like$200. So Apple has built in this pricing buffer for a very, very long time. And that buffer really saved them this year. But there's a lot more to it.

17:21Dillon Rolnick:So many companies are dealing with Ramageddon this year. And some stats that are interesting. DRAM prices rose 172 % throughout 2025. DDR5 spot prices have quadrupled since September of 2025. TrendForce expects PC DRAM to roughly double in price just in Q1 of 2026 with LPDDR pricing seeing the steepest increases in history. So every wafer that's allocated to an HBM stack, a high bandwidth memory stack for an NVIDIA GPU, is literally a wafer denied to the LPDDR5X module of a mid-range smartphone or consumer laptop SSD. So they are truly commodities. There's only so many of them, and they can only go to one place or the other.

18:06Dillon Rolnick:Apparently, OpenAI's Stargate initiative alone could consume up to 40 % of global DRAM output as it stands today. And this is the problem because there's only three major companies in the memory category. It's Samsung, SK Hynix, and Micron. Together, they control 95 % of the global DRAM production.

18:26Matteo Franceschetti:Remember, Micron has been trying to build a new facility in New York, and six people basically blocked it.

18:34Dillon Rolnick:Oh, we talked about that, but I'm missing it.

18:35Matteo Franceschetti:And the rumor is that it's like foreign funding actually trying to hold Micron up.

18:43Dillon Rolnick:Yeah, yeah. I mean, Micron has been very – they've been a huge beneficiary of AI margins and higher prices. and so much so that they've actually reallocated all of their capacity to AI and they pulled out of their consumer brand, which is called Crucial. So if you ever tried to build a gaming PC, you probably found Crucial Ballistics, which was like this very aggressively X Games branded, like PC gamer DDR memory, which was, you know - It's like the monster energy of RAM. Exactly. Couldn't have said it better myself. No, it really is. It's like the branding on Crucial Ballistics is like insane.

19:21Dillon Rolnick:Anyway, gamers will be rising up soon with monster energies in hand to take the RAM by force if they need to. I already see memes about this all the time. I see these like the worst AI slop video you've ever seen, just absolute junk. And then the caption will be like, so this is why I can't afford RAM? Or this is why RAM is$800 a stick or whatever. And then on the flip side, people will say, you know, occasionally there'll be a good AI slot video. People will be like, okay, it's worth paying$800 for RAM because this is so good, right? It goes both ways. It goes both ways. But clearly people are feeling the pinch.

20:02Dillon Rolnick:But Apple fans will be sitting out the fight thanks to a few key moves from Apple that help them avoid disaster this year. So, one, Apple has long-term supply agreements that allow them to secure memory 12 to 24 months in advance. They also benefit from vertical integration. and custom silicon. So instead of buying that consumer-grade commodity DRAM modules that you see go into either gaming PCs or NVIDIA GPUs, for example, they negotiate multi-year contracts directly with Samsung and SK Hynix for custom LPDDR packages that meet Apple's exact specifications. And so I think...

20:37Matteo Franceschetti:Yeah, so this is like RAM's always been cyclical. Yep. And so it makes a lot of sense for Apple to say basically like lock-up pricing. Samsung's locking up the demand. I don't even think Samsung would have predicted this level of demand even three, four years ago. Otherwise, they may not have priced it in the way that they did. Of course. The other thing that maybe you get to is...

21:01Dillon Rolnick:And Samsung's still making their default margin. They're just like, the opportunity cost is so high. We could reallocate.

21:07Matteo Franceschetti:The other thing is they probably have so much margin in the upsell that they could have some margin compression and still be.

21:16Dillon Rolnick:So Apple at various times has had over$100 billion in cash. They currently have over$66 billion in cash. Like they can take the hit there. They can also just take a hit to their margins.

21:26Matteo Franceschetti:Every AI lab looking at that cash, being like, Tim, I could really put that to you right now.

21:32Dillon Rolnick:Yeah, they're the only hyperscaler that hasn't really shelled out to the tune of 10, 20, 30 billion into one of the frontier labs. Maybe that changes. I don't know. somebody was throwing out like Apple should buy Anthropic or Amazon should buy Anthropic, but I think the price is too high at this point. It's just getting away. So I'm not 100 % sure about how the manufacturing supply chain for memory works, but I believe that because Apple's specific requirements, it's harder for Samsung to just instantly reallocate a product that's been already manufactured in a way where crucial ballistics, if it's going into a gaming PC, it's easier to reallocate those wafers.

22:18Dillon Rolnick:So even if they didn't have those contracts, I think the supply chain's a little bit stickier with Apple because they're vertically integrated. And then, yeah, of course, they have the money to overpay. We talked about how the MacBook Neo is this pressure release valve. So even if the memory does increase, They have this they have a cash buffer. They also have that they're charging 200 bucks for something that cost 30 if it goes to 60 They can just take a hit on margins buffer then they have a contract buffer Then they also have this neo where if the MacBook Pro goes up in price The MacBook Pro buyer is probably less price sensitive.

22:53Dillon Rolnick:They're like, yeah, you know this year it was $4 ,000 instead of 3 ,000 I'm buying it for business purposes It's not that big of a squeeze and then the price-sensitive buyer says yeah, you know what? I'm not going to go for the$2 ,000 laptop. I'm going to go for the$600 laptop this year, and they're still in the Apple ecosystem. But my takeaway is like this is an incredible testament to the operational prowess of Apple to pull this off at this scale. The Sony PlayStation 6 is allegedly going to be delayed until 2028 or 2029 because of rising memory costs. They're going to wait it out. Because there are three players, it's not just TSMC in this case, it does feel like there'll be more pricing competition and there's more of a race to build capacity.

23:37Dillon Rolnick:Whereas the TSMC, if they don't build the next fab, they can just continually raise prices and increase margins once the next contracts go up for renewal. Whereas if Samsung has a bunch of capacity and Micron doesn't, like Samsung wins. And so there's more of this game theoretic, like they got to, they have to build. And so I think a lot of people, and Tyler was talking about this with me earlier, like we all expect this memory shortage to be a few years, not like five or ten years. So if you push out the Sony PS6 to 2028. You're not bullish, Tyler? He's bullish.

24:16Matteo Franceschetti:No, I'm very bullish, but I think, yeah, because there's more players, I think the shortage will be less important. Just say you're not AGI-pilled. What do you mean?

24:25Dillon Rolnick:I mean, there is a question about like if you truly go exponential forever and it's just like, okay, yeah, there's like, you know, trillions and trillions of dollars of data centers being built. It's like, can even Samsung, Micron, SK Hynix keep up forever? So we might be living in like a permanently higher price just because like it's a more valuable product. Like we get more economic out of it. Yeah, this time might be different. Might be different. Also, the Nintendo Switch 2 might go up in price next year. It's really bad for gamers. Gaming graphics cards are going up in price. NVIDIA is pulling out of gaming graphics cards or skipping a year.

24:58Dillon Rolnick:So it's mayhem for gamers. But long term, the competition between the memory manufacturers should stabilize prices. And but, you know, just watching all the big companies that are not Apple get caught flat footed is just crazy to watch because it's not like it's not like the folks at Nintendo or the folks at Sony are like, oh, yeah, first time manufacturing something like if you told me that like the team behind the Rabbit R1 was like suffering from higher memory costs, I'd be like, yeah, like it's a new company, new product. They haven't. Like Sony has been making PlayStations for decades, and they still got caught flat-footed on this, and a lot of that is a testament to the brand power that Apple has, the network effects, the scale they built, that they're in this position and still not getting squeezed, even when other hyperscalers are getting squeezed on all sorts of stuff.

25:45Matteo Franceschetti:Tim Cook, still underrated.

25:47Dillon Rolnick:He cooked. He has cooked. And, yeah, the operational side of Apple is always just habitually underrated, even in the face of slight misses on features and products and marketing and, oh, this ad didn't land. Well, the business is still thriving and dealing with just crisis after crisis. The tariff should have been a major crisis. They got through it. This Ramageddon should have been a crisis. They got through it. It's all been very, very good news for Apple. Let me tell you about 11 Labs. Build intelligent real-time conversational agents. reimagine human technology interaction with 11 labs and get that cha-ching ready because I'm gonna tell you about phantom cash fund your wallet without exchanges or middleman and spend with the phantom card Apple is exercising the coase conjecture they're waiting for open source models to get better applying pressure to frontier lab pricing eg rejecting anthropic overtures anthropic wanted more money from Apple they went with Google and Gemini.

26:48Dillon Rolnick:This is from Soren Larson who is an economist who I really respect. It may rather be that impatient buyers have adverse selection revealing actually poor business positioning. So he's identifying Apple's leverage in the system, they're this consumer aggregator and they've been doing very well. Kyle Harrison says that quote, we ran the math. OpenAI is buying three to four times more memory than it could possibly need in the short term. The most charitable explanation is aggressive forward positioning. The less charitable one is that they're cornering the supply to kill on-device AI before it starts.

27:32Dillon Rolnick:Well, if the on-device AI happens on Apple products, that doesn't seem like that real of a scenario, but it is an interesting thesis. I think that across the frontier labs, everyone is just very AGI-pilled. They're seeing the increase in capabilities, but also these sort of like stacked exponentials. Ben Thompson was explaining this today, that when you went from LLMs to reasoning, the number of tokens, the number of compute required just to get a single answer, when like, you know, 10X all of a sudden. And then we saw the same thing with coding agents where these long running tasks are sort of like squaring the amount of compute.

28:21Dillon Rolnick:And so you just keep stacking these things on and you just get more and more compute demand, but you get more economic output. So it winds up being all worth it. When will Apple release the iPhone?

28:34Matteo Franceschetti:The only other thing on the OpenAI memory side is they plan to sell a lot of consumer devices themselves.

28:44Dillon Rolnick:Oh, so they might be buying it for that, right?

28:46Matteo Franceschetti:Yeah, so we don't really know. But yeah, I think that obviously OpenAI has kind of rolled back some of that$1.2 trillion number. but seeing the growth in demand just in this first couple months of the year starts to look like, hey, maybe he wasn't so crazy after all.

29:09Dillon Rolnick:Yeah, yeah, yeah, totally. Like the models are justifying the compute. And I think that, we talked about this yesterday a little bit with the orchestrators, just this idea that, you know, where will the next 10x come from? Well, it's probably from just running 10 instances of Cloud Code or Codex and just firing off 10 times as much inference. And people are ready to do that. You see it all the time on Twitter, people talking about how much inference they're doing. When will Apple release the iPhone 18? Kalshi has before October at 35%, before 2027 at 34%. What is the chance that they don't release it this year?

29:53Dillon Rolnick:that would be probably zero and I think it is it is zero so stay tuned I'm sure Mark Gurman will have good coverage on on the next Nick let's get the Germinator back on yeah well I wanted to invite John Gruber to talk about these but Mark Gurman of course is welcome at any time but we've we have yet to talk to John Gruber and he has been covering Apple fantastically throughout his entire career. So he would be an interesting gentleman to talk to. Let me tell you about Gusto first. Gusto is the unified platform for payroll benefits and HR built to evolve with modern, small and medium sized businesses.

30:32Dillon Rolnick:And let me tell you about Restream. One live stream, 30 plus destinations. If you want to multi-stream, go to restream.com.

30:38Matteo Franceschetti:So TSMC submitted a new semiconductor fab construction plan for the Southern Taiwan Science Park last week. The site is located on Tynon's Onding District adjacent to the F-18 fab and covers 15.46 hectares.

30:56Dillon Rolnick:They're doing more two nanometer production, the leading edge, with eight hectares allocated for the production area and one hectare for administrative facilities. The project is expected for completion and production start in 2028, says Semi Analysis. So we'll see what happens in the actual CapEx guide from TSMC. Ben Thompson has been very disappointed with where TSMC has been guiding on CapEx, but maybe this is a shift. I wonder how much faster it will be to build another TSMC fab over there. If they ramp production of this new facility before Arizona, if they eclipse and there's a flippening and this gets to scale before Arizona, it will reveal some very serious problems with America's building and permitting system, I would imagine, or maybe talent pool.

31:53Dillon Rolnick:I would certainly hope that Taiwan, I mean, this is their crown jewel, they should get this up and running fairly quickly without many delays since they've done it before. and they are heavily invested in TSMC, of course.

32:06Matteo Franceschetti:More importantly, Wendy's U.S. president posts new video amid burger battle with McDonald's and Burger King. Let's see. Let's watch.

32:14Dillon Rolnick:Let's rate it. Does he put...

32:18Matteo Franceschetti:Okay, I like the way that he's holding it.

32:19Dillon Rolnick:Okay, seems normal. Wonderful. Kind of... That's a burger. This is good.

32:27Matteo Franceschetti:I like talking about food in his mouth. Okay, yeah, going back for a second. He's saying burger. He's not saying product.

32:39Matteo Franceschetti:There we go. Yeah, no hesitation. No hesitation. Wait, what did he dip in the Frosty? Was that bait?

32:46Dillon Rolnick:What did he do at the end there? He dips a french fry in the Frosty? Whoa, okay.

32:53Matteo Franceschetti:Okay, doesn't call it a product. No hesitation. Talking with the food in his mouth.

32:59Dillon Rolnick:Yeah, yeah, that's bold. That's bold.

33:01Matteo Franceschetti:Old. I'm giving this an 8 out of 10.

33:03Dillon Rolnick:If you scroll down, I like Colonel Sanders after seeing this.

33:07Matteo Franceschetti:Also, the way he kind of puts the fry in and quickly kind of pulls the hand back. It's got style points.

33:17Dillon Rolnick:Dude, here's a good job. If you're looking for a gig that will bring home the bacon, apply for the job of CEO at McDonald's. Apparently, the guy makes$18 million a year.

33:29Matteo Franceschetti:Wait, do they have an open job application?

33:31Dillon Rolnick:No, but after this, it's entirely possible. Oh, okay. That if you put on a masterful performance chomping down, you can make a run. I don't know. It is funny that there's just like one after another of CEOs. This is probably good for all of them. I don't know. This is fun. People are – I want no jump cuts. Somebody should be seen swallowing the food. Oh, people are analyzing.

33:57Matteo Franceschetti:That's somewhat fair.

33:59Dillon Rolnick:Analyzing the details. Let me tell you about Gemini 3.1 Pro. Gemini 3.1 Pro is here with a more capable baseline. It's great for super complex tasks like visualizing difficult concepts, synthesizing data into a single view, or bringing creative projects to life.

34:16Matteo Franceschetti:Berber Jin came in and shared, Altman defended OpenAI's Defense Department contract in all hands yesterday, calling the backlash, quote, really painful. This was an example of a complex but the right decision with extremely difficult brand consequences and very negative PR for us. In the short term, he told staff. Yeah, I still stand by that given the events that have transpired in the world since last week, like being there to meet the government feels like being on the right path. It was just odd that Google's strategy was to not say anything.

35:02Dillon Rolnick:Do nothing, win, basically, I would think. Or do nothing, don't go up or down.

35:08Matteo Franceschetti:Like Gemini just... But I don't think Gemini is involved in this contract. It's Grok. It's Grok? Grok was the other one that came in with the same term.

35:19Dillon Rolnick:So Gemini doesn't have any relation with the Department of War? I think they do. I think, I thought that there was an announcement.

35:26Matteo Franceschetti:I might be, they might. I just don't think they were involved in this contract negotiation.

35:31Dillon Rolnick:Because I've seen Palantir show that they have three or four models that they sort of mix together to get responses. I don't know. OpenAI announced its Defense Department deal on Friday, hours after the Defense Secretary Pete Hegseth designated rival Anthropica supply chain risk. There's still a question as to whether or not that's going to go through. So Kalshi has it at a 42.8 % chance, and it hasn't really moved. It's unclear if that's just, like, staying in tweet land and is, like, bluster, or will there be something there? I mean, I have to imagine that at this point both the Department of War and Anthropic are talking to each other.

36:08Dillon Rolnick:Do you have more information?

36:10Matteo Franceschetti:Yeah, Gemini does have deals with the DoD. So there was the 200 million one that was in July. Anthropic was also in that one. And then they did, there was the recent like genai.mil. Do you remember Seth was tweeting about this? Oh, interesting. But they're like involved in that too.

36:26Dillon Rolnick:Yeah, but what I'm so shocked by is how does AWS have an advantage in 2026 on FedRAMP and like federal classified networks? Like I get that it's like hard. You have to do a lot of cybersecurity. You have to really harden the servers. But you would think that Microsoft and Google would have figured out how to fully support the deepest depths of the U.S. government and get full approval before, like, maybe Amazon would have gone first. But I would have thought that Microsoft and Google would be like a year or two behind. And so you would see Gemini and OpenAI, like, checking those boxes really fast.

37:08Dillon Rolnick:but I don't know, I guess AWS does have a material advantage there still. Altman said that while he didn't regret signing a deal with the Defense Department, he wished he hadn't announced the decision so quickly telling staff that it looked opportunistic and not united with the field. And I think that's a good point. I agree with him on that. His remarks echoed a memo he shared with the staff and on X Monday in which he said the deal looked opportunistic and sloppy. A groundswell of AI researchers at the company and across Silicon Valley spent the past few days criticizing Altman and the company for what they saw as a capitulation to the Pentagon by essentially agreeing to a deal that allowed AI to be used in all lawful cases.

37:46Dillon Rolnick:And we will see where the laws go. I've been digging a little bit more into like the rules around surveillance. There's actually like a ton of them and it's a very nuanced thing. And so it feels like it's hard to bake it down into a single line, but a single line is what hits on social media for better or for worse. And so you can see a single line and be like, that's amazing. It's exactly what I think it means. And that could either mean good or bad. So very, very tricky.

38:17Matteo Franceschetti:Scott Bessett went on CNBC this morning and said, no private company will ever dictate the terms of our national security. Anthropics attempts to push use clauses into their contracts with the United States government are unacceptable and their products will no longer be utilized by the U.S. Treasury or any other government agency. So doubling down there. There was also a story in Semaphore this morning from Reed. He writes, Anthropic's investors don't have its back in its fight with the Pentagon.

38:45Dillon Rolnick:This is very interesting.

38:46Matteo Franceschetti:Anthropic may be standing its ground against the Pentagon, but the AI powerhouse is doing so with a noticeably quiet quarter, its own high-profile backers. Despite the company's defiance, Silicon Valley's biggest players have remained silent. In a recent meeting between Amazon CEO Andy Jassy and Pete Hegseth, the issue of Anthropic came up, according to two people who were briefed on the encounter. Amazon has invested billions in the startup, a crucial part of Amazon's custom chip strategy as the largest consumer of the company's Tranium AI chips. And Hegseth has been threatening to designate Anthropic a supply chain risk, which would make it impossible for many of the military suppliers to do business with the company.

39:25Matteo Franceschetti:Jassy demurred, declining to take Anthropic's side, the people said. he wasn't alone. Despite Anthropic CEO Daryamides' very vocal opposition to the Pentagon's demands and rival OpenAI Sam Altman's running commentary on the matter, a number of Anthropic investors have remained silent. One investor told Semaphore that speaking up might further inflame things with the admin and that they were still holding out hope the issue could be resolved. And others said that Anthropic requested they say nothing. Anthropic and Amazon declined to comment. The The Pentagon didn't immediately respond to a request for comment as well.

40:00Dillon Rolnick:The most important thing here is that we get Pete Hegseth on Twitch. Amazon, Andy Jassy is sitting down with Pete Hegseth. Hegseth likes Twitter. He likes X. Get him live streaming those press conferences. Put him on Twitch. That's the goal here.

40:15Matteo Franceschetti:Go direct. Go directly.

40:17Dillon Rolnick:We'll talk more about the ball about the back and forth. Noah Smith summed it up. Honestly, in the Ben Thompson versus Dean Ball debate, we have Dean Ball on the show at noon. I think Ben is right. Noah agrees with Ben. There was just no way, or in any nation state, there was no way America or any nation state was ever gonna let private companies remain in total control of the most powerful weapon ever invented. This is a big question about where are we? Are we there yet? Is this the most powerful weapon ever invented that feels like, this feels very much like a training run for something that's coming in the future, and it's an important conversation, but we're not necessarily there yet.

41:00Dillon Rolnick:So we'll see, Dean Ball replied, he said, also Dean versus Ben is wild and an honor for me as a guy who's been reading Ben since the days of Android will not commodify iOS and views Ben as a core inspiration for my own project today, even though I disagree with Ben here. You will hear much more from me on this soon on a certain podcast, I hope he's referring to TBPN, but the thing is, Ben is anti-regulation and does not own the consequences of state seizure of AI, neither do you. There was another post by Dean Ball. I want to get him to unpack a little bit more. He said, it is so clear that the important fissure in AI politics right now is not liberal versus conservative, Democrat versus Republican, EAC versus EA, or safety versus anti-safety, but instead takes advanced AI seriously as a concept versus does not take advanced AI seriously.

41:52Dillon Rolnick:And why this is so interesting to me is because it feels like both Dean and Ben are in the takes advanced AI seriously as a concept camp. So we'll see if he agrees with my understanding of both of their positions, because it seems like they both are in the same camp of takes advanced AI seriously, but then they disagree about the level of pressure that the government should place on a private corporation. And so very, very interested in digging into that. Let me tell you about Figma. No matter where your idea starts, Figma make, pod code, codex, or sketch. The Figma canvas is where ideas connect and products take shape.

42:38Dillon Rolnick:Build in the right direction with Figma. Let's move on to Will Manitis. He has also been chiming in on Darius's statements.

42:46Matteo Franceschetti:There's a set of strange little fictions that tech has chosen to believe as some sort of basis set for the perceived morality of their work. Chief amongst these is the idea that the legal system doesn't exist, and when it does, it's a precise mechanistic process. Tech is a positive sum in a community, which means you pretend that the standard weapons that exist to the modern corporation, coercive litigation, IP blocking and trolling, standard white shoe law firm stuff.

43:14Dillon Rolnick:Really taking shots at white shoe lawyers there.

43:17Matteo Franceschetti:Is negative some or something bad we don't do here? This leads to a weirdly formalist view of the legal process that we must avoid it at all costs. But if we must engage in it, then it's a fair and repeatable process that must necessarily yield, if not the truth, then at least the same outcome repeatedly and provably. it's this instinct that gives you Anthropics odd contract formalism this is how you end up thinking your best move is to show your cards at the forefront and thinking you've won the hand there's incredible potential for outperformance available to you for betraying any of these silly norms that tech uses to feel good about itself the modern corporation must avail itself of all possible edge and holy wars require holy weapons so get busy

44:01Dillon Rolnick:The idea that sticks with me here is that if we engage in it, then it's fair and a repeatable process. I talked to someone who is doing something with the legal system, and he said that on this particular issue, it was a true left-right issue. A liberal judge would always cite on one side of the case, and a conservative judge would always cite on the other side of the case. So no matter how well he argued, it came down to did his case get a liberal judge or a conservative judge? And I was like, oh, well, like that seems simple. Like get a judge that's aligned with your company and your position.

44:44Dillon Rolnick:And he was like, yeah, bad news. The way they decide which judge I get is they literally spin a physical wheel. And I was like, so there's nothing you can do? He's like, yeah, no, like it is secure. It's like a real process. And like that's by design. Like some things are better left in the democratic process to inject a little randomness so that you can't put your finger on the wheel and pick the right judge and get exactly what you want. Like that's not possible. But it is potentially like the least bad system, but still very stochastic and still very random. And so you wind up in this weird situation where things are not necessarily repeatable.

45:23Dillon Rolnick:repeatable is the interesting thing, that you can go into an engagement in a legal context and wind up with a different scenario just somewhat randomly, and that is almost by design. Let me tell you about Turbo Puffer, serverless vector and full text search, built for first principles on object storage, fast, 10x cheaper, and extremely scalable. Dario was at the Morgan Stanley the tech media and technology conference today and was dropping bombs. On defense, he said, we believe in defending America. Anthropic has been working with the national security community for two years. We are the most lean forward.

46:04Dillon Rolnick:On AI's acceleration, he says, we do not see hitting a wall. This year we'll have a radical acceleration that surprises everyone. Exponentials catch people off guard. We are at the precipice of something incredible. We need to manage it the right way. I'm very interested to hear like this idea that he thinks we're at the end of the exponential. I feel like most technological processes, Carpathia was saying this maybe. He was saying like AI is a continuation of previous technological revolutions.

46:37Matteo Franceschetti:Yeah, that was why he said we're not going to see like massive growth in GDP. I think that was his reasoning. Yeah. It's just kind of the same process. Same process. You saw computers and you saw internet and AI. It's the same curve. Yeah.

46:48Dillon Rolnick:It's very interesting. I mean there are people that like he's saying there like we don't see hitting a wall this year But it begs the question like do you think that there's gonna be a plateau? Because at a certain point like you you you plot out the AI revenues and you get to something that's larger than GDP And the GDP has to move in order to grow like like what happens at the end state? What does the end state look like? I feel like we got to talk about this Me and Tyler were talking about just the general AI safety discourse on both sides always ends with people just saying, we got to talk about this.

47:24Matteo Franceschetti:No one has any answer. We got to figure this out, guys. We got to talk to each other.

47:30Dillon Rolnick:Yeah, no one's like, I have a clear plan. I think it'll work. You might disagree with me, but this is my plan. And this is what we're going to do. And here are the steps. And I think this gets us the good outcome. Everyone's just like, I don't know. It's very, very weird. Just from a comms perspective, it's very weird.

47:47Matteo Franceschetti:I mean, I feel like that's how decisions get made in companies. Like if a decision is really obvious, you can handle it over text. If it's not, it's like, let's discuss live.

47:57Dillon Rolnick:Yeah, but I mean, you talk to a lot of companies about their plans. And I mean, the good companies can think in decades and think like, okay, we're going to do this. And then we're going to make a, I mean, the original Tesla plan. We're going to make a sports car. And then we're going to make a sedan. And then we're going to make a cheaper car. and like that's the plan and and there was no like oh we got to talk about this it was like it's like yeah we're going to make a bunch of cars and then we're going to make more factories to make more cars and like this is the plan uh we're like we're just not we're just not really seeing that from from from lab leaders at the moment um but still plenty of optimism uh there the the the square packing circle packing discourse has hit the timeline i absolutely love this first let me tell you about shopify shopify is the commerce platform that grows with your business and lets you sell in seconds online, in store, on mobile, on social, on marketplaces, and now with AI agents.

48:49Dillon Rolnick:So John Bidwell in 1997 found a more optimal way to fit 17 squares into a larger square. The smallest square that contain all 17 looks like an absolutely deranged solution. Everyone would expect that it's just a nice even grid. You add one. But if you pack them in this way, you get 17 squares of absolute chaos on the next slide. And someone made a waffle maker. Someone made a waffle maker that optimally packs 17 squares into the squares of the waffle. You will get mocked at breakfast. You will get mocked by anyone except a mathematician, I suppose. Yeah, but it's optimal. Kevin says, for years, society was limited to only 16 syrup squares per waffle, but with recent combinatorial optimization breakthroughs, our research department has achieved previously unheard of densities of waffle syrup.

49:53Dillon Rolnick:It's great, great news. Very, very funny that this came through. And then someone in the previous post was posing, the universe is not real, dude. This is ridiculous. I'm going to fight God because it looks very crazy to pack an arbitrary number of squares within another square. But that's – Someone should do this for cubicles in an office. Can you imagine if you're in the one cubicle that's slightly off kilter and everyone's like, look dude like you get you have the same amount of space as everybody else and you're like yeah but i'm at a 37 degree angle to the wall and it's driving me insane uh that would be very very

50:35Matteo Franceschetti:odd um did you see researchers at the university of southern denmark developed a limbless soft robot that crawls using air-filled chambers and kirigami inspired flaps you're gonna love this i

50:50Dillon Rolnick:hate it.

50:50Matteo Franceschetti:You're going to love it.

50:51Dillon Rolnick:I am the Indiana Jones of tech podcasting. I hate snakes. I hate, I absolutely hate snakes. Very, very weird. It's over boys says autism capital. Um, very weird, but, uh, you know, exciting that people are working on robotics and I'm sure there's some, some odd application. Yeah.

51:09Matteo Franceschetti:I'm sure there's no like ultra nefarious use cases or something like this.

51:14Dillon Rolnick:We're going subterranean. Anderle's already working on it. What do you got, Tyler?

51:17Matteo Franceschetti:These look like these would be good for like medical. For what? Medical. Make it really small. Yeah, maybe. Would you volunteer as a trial subject? I think you should go first. If you think they're so good, I think you should go first.

51:34Dillon Rolnick:Someone launched Ciggies.app, the most comprehensive archive of Chinese cigarettes to exist on the Western internet. Discover new packs in their history, track favorites and packs you already tried. Share your collection. the Chinese cigarette industry is absolutely fascinating. I think it makes like hundreds of billions of dollars in profit for the government. The main cigarette company is nationalized. Oh, wow. And so they make an absolute ton of money. They fund their military with it. It's a fascinating industry.

52:08Matteo Franceschetti:Yeah, really lovely website.

52:10Dillon Rolnick:We really got to go into the history of nationalization because the nationalization and privatization is interesting all over the place. I looked into nuclear weapons and how those were nationalized. Fascinating sort of like hybrid public-private partnerships with a lot of consultants in the private sector. And then obviously the defense primes making the missiles, the ICBMs, but not the warheads. But in the tobacco industry, there are some fascinating examples of countries that start cigarette manufacturing businesses, develop large sort of deeper supply chain operations and then privatize those.

52:51Dillon Rolnick:And then when they sell that asset to a Western tobacco firm, there's sort of like a cash windfall for the government that can be very beneficial. They can go build roads with the money or, I don't know, cut people's taxes. But there's the other side of, okay, a company is getting large and then it gets nationalized. But are there any other industries that stuck out to you, Tyler, as you were thinking about it?

53:15Matteo Franceschetti:I mean, I'm still kind of looking at this. There's so many weird ways to view nationalization, though. So at least in the U.S., historically, it's been mostly basically during wartime or if a really important company is facing a huge crisis. So 2008, I think GM at one point... Government motors. Yeah, the Treasury owned like 60 % of GM. So there's all... Fannie Mae and Freddie Mac.

53:40Dillon Rolnick:But even if you think about tobacco in the U.S.,

53:43Matteo Franceschetti:almost like having super strong taxes

53:45Scott Kupor:is almost kind of a form of nationalization, right? Where you're basically taking a ton of the profits away.

53:52Dillon Rolnick:Yeah, yeah, yeah. I mean, that was always the pushback of, if there's runaway profits by the AI companies, just instead of nationalizing them in that scenario, you can just tax them and then redistribute the profits. Yeah, there's like this real spectrum, right?

54:06Matteo Franceschetti:Because if you think of Intel, so the government now owns, it's like 10 % of Intel, right? But what does that actually mean for running Intel? Do they have a say? Right now, they're not supposed to. Maybe that changes. It's like this interesting spectrum.

54:20Dillon Rolnick:Yeah, totally.

54:24Dillon Rolnick:Anthropic is on track to generate annual revenue of almost$20 billion, a projection based on current performance, more than doubling its run rate. This is from Bloomberg. And Zero Hedge says, unprecedented damage control, leak after leak. Just how bad is it? And Matt Slotnick says, the leak is bad, sir. really bad. Now everyone knows we're the fastest growing software company of all time. We must put a stop to this. I think what Zero Hedge was getting at was that the reason you leak strong economic performance is to allay concerns around pressure from the government, supply chain risk, because really the revenue would not be growing if companies were pulling out because they're so worried about the supply chain risk that they got to move on so fast.

55:08Dillon Rolnick:But still.

55:09Matteo Franceschetti:Yeah, clearly adoption is outstripping any kind of pullback from the Department of Treasury and various other kind of organizations in the government.

55:21Dillon Rolnick:Yeah. R. Karazian over at Ramp has some extra data. He said, most people in tech know Anthropic commands the majority of API spend by U.S. businesses. As of January, Anthropic took 50 % of spend on enterprise AI subscriptions, too. OpenAI still leads on business count, but the biggest spenders go to Anthropic, all from Ramp Data. And Ben Thompson had some extra context here about the data, but I was actually DMing with Aura last night, and it is fascinating what's happening. The market is so, so big that everything is growing, and you see these charts and you're like, it's over for this company, it's over for Cursor, and then Cursor puts up$2 billion.

56:04Dillon Rolnick:and there's just more and more growth that's happening because of the diffusion problem and the fact that companies onboard and just more and more companies are onboarding to AI tools every day. Ben Thompson said there are some caveats to the RAMP data. It only includes OpenAI and Anthropic, not Azure Co-Pilot or Google Cloud Gemini. RAMP is also probably tilted more towards startups and smaller companies and larger enterprises. That noted, it certainly matches the vibes. Anthropic established that coding foothold and has been on an absolute tear over the last six months in particular, in terms of not just its model capabilities, but also cloud code and the overall sense that it is in the lead.

56:41Dillon Rolnick:And enterprise CTOs are a smaller and more easily reachable set of customers than consumers at large. Meanwhile, an AI subscription business model has always been more compelling for the enterprise than for the consumer space for the same reason that productivity apps always end up as enterprise products, not consumer ones. There's no consumer CRM, for example. Enterprises actually value productivity and are willing to pay for it. Consumers don't and won't, which is why the business model makes sense for the consumer space, at least at scale, is advertising. So he's, once again, ad-pilled. What else is in the news?

57:20Matteo Franceschetti:Eddie, Endowment Eddie put this into context. Even in Q4 of 2021, public SaaS quarterly net new ARR peaked at$2.6 billion. Anthropic has added nearly$11 billion in early 2026. Yeah, this is from Altimeter. That, of course, is like run rate, not actual ARR necessarily, but still absolutely shocking. Tyler said, for those who can track at home, Anthropic just added a Palantir in revenue in the month of February alone.

57:53Dillon Rolnick:Yeah.

57:54Matteo Franceschetti:Pretty unprecedented. in it.

57:56Dillon Rolnick:Howard Linzon said, Anthropic pulled off the rare Quadruple Lindy. Great product. People pay for it. People love it and promote it even though they pay. And the people like Dario's handling of government. Next up, the people tear Dario down. There is this interesting tall poppy syndrome in Silicon Valley a little bit where whoever's at the top gets triple glazed and then it needs to be torn down a little bit. So we'll see where all this goes.

58:23Matteo Franceschetti:But back in Putting AI growth in the context of the shoe market. Yes. Darren Rovell, former guest of the show, shares that New Balance is up almost 20 % year-over-year at 19%. Adidas, 13%. Lululemon, 9%. Lululemon's doing well. And then every other brand here in the negative. Puma at minus 8%. Under Armour, minus 9.4%. Nike, minus 10%. And Jordan Brand, down 16%. Really insane. insane to see such a huge decline. I feel like the sneaker era, the Jordan era is over for now.

59:05Dillon Rolnick:Is this more of like changing in taste that New Balance and Adidas are growing and Nike is not? Because if they were all, if all of these like major household name brands were trading down, I would say like, oh, this is like the Teemu effect. This is the white label effect. This is the the all birds the dfc like just more competition from newer brands uh challenger brands but it seems like there's more of just like a reallocation from nike to new balance and adidas like adidas might just be on a brand tear doing some good advertising but um new balance has been killing it with their positioning interesting to see total revenue as well uh nike is huge 46 billion.

59:44Dillon Rolnick:Adidas is down at 26. Lululemon's at 10. Lululemon has done really, really well. New Balance, they're bigger than New Balance. Anyway, my portfolio every day at the opening bell looks like D-Day says, Human. This is an absolute disaster. It really is so crazy. I was trying to explain yesterday to a friend how the market, I was like, you're probably getting destroyed, right? And he's like, yeah, it's terrible. And I was like, well, it might shock you to learn that the market overall is like flat for the year. It's an absolute tale. Korean stocks are crashing. The Kospi index, the Korea Stock Exchange is down 10%.

1:00:29Dillon Rolnick:Korean stocks are crashing, says Joe Weisenthal. The South Korean stock market fell over 8 % and triggered a circuit breaker. Jim Cramer said, beware of South Koreans spill over to our markets, WDC, STX, SNDK, MU, all still vulnerable. So there's lots of companies that are at risk. We will dig into the South Korean stock market collapse soon, but we have our next guest in the Restream waiting room. First, let me tell you about Sentry, Sentry shows developers, what's broken and helps them fix it fast. That's why 150 ,000 organizations use it to keep their apps working. And without further ado, let's bring in Dean Ball from the Rooster Room Waiting Room.

1:01:10Dillon Rolnick:Dean, how are you doing? Good to see you. Good. How are you all? We're fantastic. I've been enjoying your coverage. We had Ben Thompson on the show Monday. It's very funny that this debate is phrased as the Ball versus Thompson debate because it feels like you agree on way more than you disagree on. But I would love for you to sort of set the table for us and just kind of give us your run of like what actually happened because there are a lot of things that have been said but not enacted there's been bunches back and forth so what's the what's the what are the rules of the road what are the what are the the table of facts that you think

1:01:49Dean Ball:like are the most reliable here first of all there are very few people i respect more than ben thompson and who have inspired me more than ben thompson so yeah i think we agree on the vast majority of tech policy issues and other tech stuff. Yeah, I mean, I think basically what's going on here is that, you know, there's a dispute about a contract, right? And that's fine. Brownups can have disputes about contracts and they cannot come to terms of business and then they don't sign the contract or they cancel the contract and that's fine. The problem here is the punishment. The problem here is that the government is saying, if you don't do business with us, we'll destroy your company or we'll threaten to.

1:02:25Dean Ball:And that just can't be the way that America works. and I don't really there are people who say including I think this was an argument that Ben made that well what should Anthropic have expected? What should they have expected? The way they talk about AI being so powerful, how could they not expect the government to bully them and quasi-nationalize them in these ways? And my view is like well no, what we should do is have modest technocratic regulations so that we can avoid these sort of extreme outcomes of either like, yeah, total decentralization, total anarchy, or like, full Bolshevism, right?

1:03:01Dean Ball:Like, we can avoid that. It's called capitalism. It's called like the system that we have now. And I've spent the last two years being confused as to why we don't try to do that.

1:03:10Matteo Franceschetti:So in terms of - Okay, okay. So I guess - So far, Anthropic has not been labeled a supply chain risk. Yeah. Officially, is your belief that the Department of War is just really busy right now for obvious reasons and they haven't gotten around to it yet? Or do you think that it's coming? And that's why, you know, like, where do we actually go from here?

1:03:38Dean Ball:I truly don't know. I operate under the assumption. So War Secretary Pete Hegseth said last Friday that they're going to do it. And they're going to do an extremely expansive version of it. So I operate under the assumption that he's not lying about that. I basically take him at his word. It'll happen eventually. But I don't know.

1:03:55Dillon Rolnick:Do you think he, I mean, I know that there was some debate over the shape of supply chain risk, whether that means you have a partnership with Amazon, Amazon works with the government, you can't work with Amazon at all, or Amazon can't use your software specifically, your model specifically on government contracts. Do you have any idea of where the clarity might be found there?

1:04:20Dean Ball:Well, so in terms of what the supply chain risk designation actually allows, my reading of the law is that it would be only in the fulfillment of the government contract and not more broadly. But there's a lot of things. First of all, you could just write it more broadly and try to fight it in court and see what happens, right? And who knows, right? You could also jawbone, right? You could have the president call Amazon and say, hey, we don't think you should be working with these guys, right? Very hard to sue about that because that's not a policy action. And there's also a very, very large range of other things that the federal government could do if it wanted to.

1:05:00Dean Ball:if it wanted to really harass, which, you know, the irony of this is that we all know this because of what happened to Elon's companies during the Biden administration. I objected to that then, and I object to this now.

1:05:12Dillon Rolnick:Yeah. Unpack a little bit of that history. Take us through it.

1:05:16Dean Ball:Yeah. So, you know, basically Elon being kind of a political enemy of the Democrats, the Biden administration brought on a very large number of investigations, regulatory actions, FCC, DOJ, a lot of different other things. I think there's a chart somewhere and it's like 12 to 20 things. It's a huge number of different things that they brought. And each one of those is going to entail millions of dollars in compliance costs and legal costs for these companies. So it's non-trivial amounts of harassment that's going on. And I, again, I think we shouldn't do that kind of thing.

1:05:53Dillon Rolnick:Sure. Is this broadly under the umbrella of like lawfare? Yeah. Okay. Walk me through this idea of the fissure in AI politics right now is not liberal or conservative, Republican versus Democrat, EAC versus EA, safety versus anti-safety, but instead takes advanced AI seriously as a concept versus does not take advanced AI seriously as a concept. It feels like you and Ben Thompson both are taking AI seriously as a concept. Does that map with your reading on his arguments, your arguments? It feels like there's just a different downstream understanding of how this plays out in a world where advanced AI is a serious concept to grapple with, different than making pencils.

1:06:44Dean Ball:Yeah, right. I think the other way to put this would be, do you think that Dario Amadei and Anthropic are directionally right? And the other Frontier Labs, it's worth noting, are directionally right about where this is going, or do you think they're not? And also, not just do you think they are, but do you really take that seriously? Do you own the consequences of believing that these guys could really be correct? And I think a consistent theme of my writing over the last year especially has been kind of being confused that there's people who just do not take, they might say that like, oh, AI is progressing very quickly, but they don't actually like, their downstream assumptions and like conclusions about what we should do don't seem to sort of reflect that.

1:07:32Dean Ball:It's very hard. It's a very hard thing to do. But yeah, that would be like my broad take.

1:07:35Dillon Rolnick:So this is where I start to get confused, because if I take advanced AI seriously, then it feels like nuclear weapon level technology in a few years. Maybe it's five, maybe it's two. But that leads me more into the Ben Thompson camp, where I say, well, of course there's going to be a dramatic showdown between the U.S. government and the private company that has invented something akin to nuclear weapons. And so walk me through how you are thinking of, you know, if we're developing new, new, you know, powerful technology, what how should the government actually interact with that? Aside from like the lawfare in the short term, like what's the long term solution?

1:08:24Dean Ball:Yeah, I mean, the long term solution, I think, is probably going to be a hybrid system where there just is regulation of these frontier labs and there is government oversight of them. but it's just like JP Morgan is a private entity that lobbies against government policy all the time, right? They're a totally independent actor of the government and yet they're also deeply intertwined with the government I'm not saying that we should regulate AI companies like banks, I hope not but I'm saying that technocratic regulation exists in capitalist republics and it's not communism and we didn't nationalize the banks right?

1:09:01Dean Ball:Right. So it would be my proposal would be to sort of take step by step approaches to solving actual problems, not just doing regulation for its own sake, but really trying to shape this so that it isn't some sort of violent or not literally violent, but some sort of conflict. And instead is, you know, a gradual development of like, yeah, this is how the government and the public ultimately achieve oversight of these labs. Because I would agree that if we are going to this place of like near term super intelligence or something, that isn't going to just exist in an unregulated fashion. That's kind of been my point for, you know, for a year and a half now.

1:09:37Dillon Rolnick:Yeah. So what's the gap between regulation and nationalization? Because at a certain point, if we're in the nuke analogy, it feels like the government takes that and then they get to decide where they point those nukes. And as I was reflecting on the creation of the atomic bomb, I was sort of left optimistic about the American project and optimistic about the way that played out. I don't know if you have a different side, but I feel like the government did get control of nuclear weapons. There hasn't been a nuclear war. And the whole concept of the person that you're voting for, they'll have the nuclear football.

1:10:19Dillon Rolnick:They'll have their finger on the button. Do you trust them? That is an animating force in our electoral process, whether you like the current administration or the previous administration, whether you trust them or not. But like Americans vote on that question every four years. And I think that that's a good system. I like that.

1:10:40Dean Ball:100 i think that the nuclear bomb analogy starts to break down okay but one thing i'd say about about nuclear weapons is that it's true that nuclear weapons you know we got that and that went i think basically pretty okay as far as these things go but you know what we didn't really get is nuclear energy to the nearly to the extent that we could have completely agree with you on this yes continue right and so that created this single regulatory sort of point of failure Single points of failure are often problematic in complex system engineering. So what do you do? Well, maybe we shouldn't have done it.

1:11:16Dean Ball:Maybe the government effect on the equivalent of the consumer side, the economic benefits, we didn't really get those as much as we could have. So that would be one thing. But the other thing is that even nuclear energy, it's not directly useful to you. Nuclear energy is not like, you are not personally going to express your liberty exercise your liberty through nuclear energy or nuclear bombs, but you will with artificial superintelligence, and you certainly will with today's coding agents, right? And so the question is, like, should the government be able to own that? And my view is that without really substantial guardrails, that is just almost certain to devolve into a profound act of tyranny.

1:12:01Dean Ball:And so I think that having the companies separate and apart from the government, but still overseen by the public through regulation. That is kind of like my view as to what is ideal. But the problem is that saying that view, hey, it's really hard to get that right. Regulation is tough to get right. You want to get it exactly right. And it's possible that all my ideas are terrible and they're all going to go to shit, right? That's totally possible. But what you want to do is you want to, yeah, you want to at least try to get that right. The problem is that that view has been characterized very often as sort of supporting regulatory capture.

1:12:37Dean Ball:So it's like, you know, what did Anthropic expect the way they talked about these models? Well, maybe they expected for like California's SB53 to pass and for there to be like modest technocratic light touch regulation that only affects the frontier labs. And like, I don't think that's lobbying for regulatory capture. I think that's actually just the prudent thing to do. Even if I don't agree with every regulatory impulse that Anthropic has, I think that in broad strokes, that's the play.

1:13:00Dillon Rolnick:Yeah, how do you, oh, sorry, Jordi, you go.

1:13:02Matteo Franceschetti:Yeah, I wanted to ask, like, how much do you think information asymmetry played into the dynamic of the negotiation and the results? I mean, you have the Department of War, who presumably everyone on that side has some idea that we're about to enter a conflict. And Friday comes around, the Department of War knows they're hours away from a strike. Dario, according to Emil, like isn't kind of engaging with him. and out of that, maybe the admin says, hey, this is not a partner that we can rely on and we don't want other groups in the government to be relying on it as well. And it just kind of blows up.

1:13:44Dean Ball:Yeah, I mean, so I think some of this is about personality. Some of this seems like it's about personality clashes. Some of it's about politics, undoubtedly. I think some of it is also about principle though, right? Like, and I think the DOD is not crazy at all. Like my problem here has never been with the DOD's policy principles here. They're saying Anthropic can't put usage restrictions that look like public policy because we set public policy. Dario Amade doesn't decide when autonomous lethal weapons are ready for prime time. We do. And I think they're 100 % right about that. But the solution to that problem is you don't try to destroy the business of Anthropic.

1:14:20Dean Ball:You tell Dario, no thanks to the business. And you move on and you find somebody else who will do business on your terms.

1:14:27Matteo Franceschetti:Which is what Dario said in the CBS interview. He said, there's other providers we can agree not to work with each other. Sure.

1:14:34Dean Ball:Yes. Right. That's fine. That's fine. I mean, frankly, the tokens are probably higher margin for Anthropic, you know, sold to private entities, not to the government because Anthropic is subsidizing them for the government. So it's a bad business as we're seeing.

1:14:47Matteo Franceschetti:What's the good outcome here?

1:14:50Dean Ball:I think we cancel the contract. And look, the other thing is like the DOD says, well, we have all these other contractors. there's Palantir's of the world and Palantir might rely on Anthropic. And so we can't have, our contractors can't be relying upon Anthropic because if that's true, then we're still relying upon Anthropic. So it's fine for them to say, no one who has usage restrictions like this can do business with any of our contractors. I think that's fine in the fulfillment of DoD contracts. Like I think that is perfectly fine.

1:15:21Dillon Rolnick:Isn't that basically supply chain risk by any other name?

1:15:26Dean Ball:Well, no, because supply chain, supply chain risk would be specific, would be done at the company level as opposed to like at the YouTube. At the contract level.

1:15:36Dillon Rolnick:Yeah, yeah, yeah, yeah. I agree with you. Also, can you explain a little bit more about the logic of supply chain risk? Because I was shocked when like DJI is not on there. DeepSeek doesn't seem to be on there. Quinn doesn't seem to be on there. Like this feels like very bad if I'm just trying to think of like my short list for things that I might want to supply chain risk. Like, yeah, it's good Huawei's on there, but I got 12 other. Unitree. Yeah, Unitree I don't think is on there. Yeah, like when should the government apply this? Does it matter if it's international? Is the national versus international designation important?

1:16:17Dean Ball:Well, one thing I'll say is we don't know everyone on the supply chain risk list because sometimes those designations can occur in classified contexts. But it is the case, I think, that the supply chain risk designation, if you go and look at the statutory history, it's really intended for foreign adversary companies. This is about, it's really about China, frankly. It's really about China. And so it is kind of wild to be in a, whatever you think of the chip export policy of this administration of wanting to sell more NVIDIA GPUs to Chinese companies. It is kind of wild to have that policy and then also to say that we're going to treat American companies that we don't like like Chinese companies, like like enemies of the state.

1:17:01Dean Ball:I don't think we should be doing that.

1:17:03Dillon Rolnick:Can you talk to me about what your 2025 was like, what your goals were, what of those goals have been achieved? Like give me sort of a scorecard on like AI policy in D.C.

1:17:18Dean Ball:Well, you know, what I wanted to do is, I mean, first of all, you know, obviously getting the action plan done, that was a, you know, 18, 20 hour a day kind of thing for four months of my life. So that was an intense sprint. And getting it done is obviously a career highlight for me. I didn't get it done unilaterally, of course. A lot of people worked on it. But, you know, that was a big part of it. Sort of course correcting from what I thought were some of the excesses of the Biden policy. a lot of which was, by the way, this national security inflected control over the labs. So it does disappoint me to see us sort of regressing back into Biden era mentality, which is what this is.

1:17:56Dean Ball:So that is a little bit concerning to me. But yeah, like that. And then more broadly, I want to AGI pill people. You know, I want people to like feel the importance of what is going on and hopefully, you know, try to think more strategically about this issue. Because the thing is, is that it can't just be one or two people doing this. It has to be a community of people.

1:18:25Dillon Rolnick:So walk me through what AGI pilling means in March of 2026.

1:18:33Matteo Franceschetti:More Mac minis than people on Capitol Hill.

1:18:35Dillon Rolnick:Yeah, I mean like there's terms like software-only singularity, that robotics will be delayed, that there might be – Dario talks about like the end of the exponential. He doesn't seem necessarily super intelligence-pilled. He seems like we will get someone who's 150 IQ and you will be able to run it in a data center like the – what does he say? A million geniuses in a data center. Country. It's not one god – country of geniuses in a data center. It's not one God that's like so like time traveling and teleportation. Like it's not entirely all this crazy sci-fi. It's a little more narrow. So in terms of timelines, expectations, like what are you telling to elected officials these days?

1:19:21Dean Ball:Well, I think the very challenging part is that there's the trajectory of the technology and then there's the effect of the technology in the near term, right? So the technology that we have right now is already legitimately science fiction in my mind. but we don't live in a science the world is not obviously more science fiction-esque than it was three years ago we don't have flying cars we don't live on the moon we're not doing it and we don't have Dyson spheres assembled around the sun or whatever and the reason for that is that it takes a very long time to transform institutions and to transform the way that organizations are structured and so again that's the important thing to me it's about the actual winner of the AI competition is going to be the country, the civilization that is like most imaginative at like, we went from artisanal types of, you know, manufacturing to factories to eventually assembly lines, right?

1:20:20Dean Ball:We're going to do that, but for the next generation of things and inventing that, what is all that? What does it look like? And, you know, that's going to be where the real economic benefits come and the real things that change the world come from. But those take time and they're really conceptually difficult.

1:20:36Dillon Rolnick:How do you think about competition with China with the backdrop of nationalization? I heard this quote that was like, if you nationalized SpaceX a decade ago, you'd just get NASA. You don't get SpaceX if it's owned by the government. It needs to be independent. And so with that backdrop, What does a healthy American AI lab ecosystem look like that can actually go and win on a geopolitical scale?

1:21:03Dean Ball:Well, I think first of all, yes, it's got to be private companies because they can move faster and be more innovative. Also, very importantly, it's about trust, right? Like selling, if we're going to sell these services internationally, we need other governments and foreign companies to trust our companies, is to trust that they're not just ultimately linked to the US military, right? Because that's the problem is that doing business with China, you sort of know, well, at the end of the day, every company in China is ultimately a military asset. And that diminishes trust in Chinese companies. That's something really profound that we're eroding with this action right here.

1:21:42Dean Ball:And that worries me quite a bit. So it's about maintaining that trust.

1:21:46Dillon Rolnick:I've been thinking about that in the context of like sovereign AI throughout Europe. And I was just like, look, look, France didn't need its own Google. They could just use Google. Like it was fine. But when you actually take AI seriously, then you start to wonder about, okay, well, does their strategy as behind as they might be, does it actually make sense? Yeah.

1:22:10Dean Ball:One of the things I hope we invent is open source models, I think are useful and good. But I actually think what's more important right now is like open source infrastructure such that more people can sort of like really put their own imprint on models. And more people can feel that sovereign experience without having to like build, you know, multi-gigawatt data centers and whatnot. I hope we can do that. That's a really interesting thought.

1:22:42Dillon Rolnick:Would that be like a national compute reserve or something like that? Or like, yeah. That like academic institutions could access?

1:22:49Dean Ball:It could be like that, but also even more than just the compute itself, also like all the infrastructure that connects the compute and all the like, I think there's something really interesting there. And I'm excited to see more people do things like that.

1:23:02Dillon Rolnick:Yeah. What do you think about the concept of like an FDA for new models? You're releasing a new model, like it has to pass a battery of tests or benchmarks that are maintained by a like a federal institution in the same way that when you when you release a new cancer drug, you have to test it on mice for something like that.

1:23:20Dean Ball:Yeah, I think the problem with that is that there's too much to test for because the models can do too much. So we don't know what to test for. So we have very clear endpoints when we're testing a drug in a person, right? We have very clear things that we want to test for. It's much harder to do that in the context of a generalist AI model. So that's one of the things that worries me about licensing. In addition to licensing can slow things down and regulatory capture is very real. It's a very real thing. I have always thought of AI as being a little bit more like financial services, where the regulation is not actually targeted at the products per se.

1:24:01Dean Ball:The regulation is more generally targeted at the entities themselves. So we regulate banks. We do regulate loans, but we do that by way of regulating banks. There's not a government agency that approves every single loan that a bank makes. What we do is we look at the entity of the bank and we say, this is a sound institution. This is a soundly governed institution. That's the kind of thing that I think will be, that's, I don't know, again, I don't think we should regulate AI like banks per se, but I think that is a useful structural intuition for thinking about AI regulatory policy.

1:24:36Dillon Rolnick:Yeah. Jordy, please.

1:24:39Matteo Franceschetti:How much attention is DC paying to the news out of Square? the pushback from or block the pushback from Silicon Valley was, hey, you're blaming this on AI, but it doesn't feel like maybe that's a small part of the reason to do this riff, but certainly not the entire story. But then a bunch of groups are going to use this as, you know, example number one as to changes in the labor force due to AI.

1:25:09Dean Ball:Well, it's a bad combination because it's all these companies, all these tech companies that overhired in the sort of immediate post-COVID. And then now they're also the most exposed to AI, right? They're the ones that are adopting AI the most aggressively. They do the things that AI is best at. There's lots of software engineering. And so it's both things happening at once, I would say. And that creates an exaggerated sort of funhouse-like effect of the actual impact of AI on the labor market. I'm very skeptical that we're going to see things like that at like wide, you know, firm level anytime soon, you know, in the broader economy.

1:25:47Dean Ball:I think we could see it more in the software engineering discipline because there are a lot of big software engineering enterprises that hire too many people, right? So like, you know, I think that's a very live possibility. But DC is DC, people are going to interpret the news for whatever's convenient for them, you know? And if you, when you play the long game, you have to try to just build credibility over the long term with people. But yeah, I mean, everyone's going to say this is evidence for whatever I'm on about.

1:26:14Dillon Rolnick:Yeah. Uh, hypothetically, how do you think something would play out if, if like there was a battle on the moon and NASA says, Hey, we want to buy a bunch of rockets from blue origin or SpaceX. And we want to go to the moon to fight this war on the moon. And the private company CEOs say like, no, we're not interested in, in that we want to just launch more satellites. Like it feels like another analogy for like the nationalization conversation. And maybe, maybe it's just a weird time because I feel like in previous eras, uh, a lot of the, you know, you go back to like the dollar a year, man, a lot of, a lot of American industrialists sort of like stepped up to the charge and we're like fans of the government.

1:27:03Dillon Rolnick:And now there's like much more, it feels like we're more divided than ever. But what is the correct way to marshal a cornered resource as a government?

1:27:17Dean Ball:Well, I would think in practice that the companies, the SpaceX's and Blue Origins of the world would actually eagerly help the government there. But let's take your hypothetical and say that they didn't. There's an authority called the Defense Production Act, Title I, which is, there's something called priorities authority. And the government can quite literally just put itself at the front of the line for any launch of any rocket. And that's what I would suggest they do in that situation. Rather than commandeer the rockets themselves, that's kind of what I would suggest.

1:27:46Dillon Rolnick:That makes sense. So Defense Production Act, but it requires extraordinary circumstances.

1:27:52Dean Ball:The president can do it basically whenever he wants. The president can make a unilateral finding. The other thing I'll say that's funny about all this is that one of the things I was enthusiastic about putting in the action plan, a little notice provision of the action plan is basically saying, in the event of a national emergency, DOD needs to know with the hyperscalers how it's going to commandeer the data centers if it needs tons of compute in some sort of national emergency. That's in the action plan. So I'm not here saying total liberty, whatever, no government involvement. That's not what I'm saying.

1:28:27Dean Ball:I'm just saying that like, Anthropic has a right to exist. If they don't do business with the government, we shouldn't kill them. That's all I'm saying.

1:28:34Dillon Rolnick:Yeah, I think the Wall Street Journal called it like a fight about vibes. And I think most people are not at the point where they say, yes, AI is actually nukes. And yes, like one company should like, you know, like they should be destroyed if they don't give it all up. Like that's a very middle of the road position that I think a lot of people agree on.

1:28:54Dean Ball:Yeah, 100%.

1:28:54Dillon Rolnick:Yeah, that makes sense. Jordi, anything else? Not for now. Well, thanks so much for coming on. Yeah. Good to see you. Hey guys.

1:29:01Dean Ball:Thanks so much. Good to see you

1:29:02Dillon Rolnick:We'll talk to you soon. Cheers. Goodbye Let me tell you about vibe.co Where DTC brands B2B startups and AI companies advertise on streaming TV pick channels target audiences and measure sales just like on meta

1:29:15Matteo Franceschetti:Another Tyler Dean said that the reason we don't have Dyson spheres is because of institutions aren't quick to adapt. So you have no more excuses not to build a Dyson sphere.

1:29:28Dillon Rolnick:Yes.

1:29:29Matteo Franceschetti:Make it happen.

1:29:29Dillon Rolnick:This is a high agency organization. Yeah. We have very little institutional overhead.

1:29:34Matteo Franceschetti:Make it happen.

1:29:36Dillon Rolnick:No, no. In the limit, I believe it. I mean, it's like a, it's a long chain of events that require all of these different diffusions to happen. But...

1:29:46Matteo Franceschetti:We got to ask Jared his Dyson sphere timeline.

1:29:50Dillon Rolnick:Yes, yes. I mean, obviously the big question is Moon versus Mars, but I like getting up to speed on the plan for the Dyson Sphere. That should be firmly within the 100-year plan for NASA. I believe it. Let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Just do it. And let me tell you about Graphite, code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. And without further ado, we have Scott Kabor returning to the show. show, and Jared Isaacman for the first time. Welcome to the show.

1:30:25Dillon Rolnick:How are you two doing? What's happening? We're doing great. How are you? We're doing fantastically. Thank you so much for taking the time to talk to us. Please take us through the news today because I'm very excited to hear from you. I'm very excited to be joined by you today.

1:30:39Scott Kupor:Well, I'm just here for moral support from my good friend Jared. So Jared, you want to give him the news? Fantastic.

1:30:45Jared Isaacman:Sure. And by the way, I appreciate giving me a 100-year time frame on the Dyson Sphere. because it's been, I mean, we're over a half century here just getting back to the moon. So we gotta take things in steps. But really, that's where the partnership with OPM comes in. And Scott has been a champion for us here with NASA Force. We got to rebuild some of our core competencies here at NASA. You know, we made some big announcements last week that said, look, if we want to get back to the moon, we want to achieve President Trump's vision within the timeframes that have been established. We're going to have to do things differently.

1:31:21Jared Isaacman:We're not going to launch moon rockets every three years. We're going to try and launch them in less than a year. And people are like, that's impossible. I was like, have we, you know, I literally went and got the data from the Mercury, Gemini, and Apollo era saying, you know, we used to do this every three and a half months. We literally, from the time Apollo 7 splashed down to Apollo 8 launched to fly around the moon, is measured in weeks. It's eight, nine weeks here. So what do you mean we can't do it? Well, they said, look, we don't have some of the same expertise we used to have during the space shuttle era of turning our mobile launchers around.

1:31:54Jared Isaacman:You know, if you're launching every three years, you build up all this experience and expertise. And then they then they go out after the mission is launched and they go into industry. So we got to we got to do some things differently right now. And one of which is we got to start bringing in folks back from industry that have the expertise to grow our young talent, because we have no shortage of people that want to grow up and work at NASA. I mean, you know, like from the time they're kids, they want to get there. So plenty of young talent coming in the system. But what Scott and OPM are giving us the ability to do is bring in term-based appointments from industry.

1:32:25Jared Isaacman:Some brilliant minds from SpaceX, Blue Origin, Rocket Lab, all our great tech companies who say, I want to serve my country. I want to stick around for a couple of years under a term-based appointment, elevate some of the talent. That's how we're going to get back into turning rockets around in months instead of years. That's how we're going to get back to the moon. And so there's amazing partnership.

1:32:44Scott Kupor:This is part of a broader thing that I think you and I talked about a little bit, which is there's kind of two pieces we're trying to solve on the government tech side. One is obviously can we get early career young people in, as Jared said, kind of NASA, unlike most of the federal government, actually has done a really good job there. The other piece is how do we leverage the private sector to help make sure that we have much better connectivity in the private sector and public government? And so the second part of tech force was always the private companies having people come in. And obviously, we want to basically double down on that now for NASA force and make sure that we've got kind of all those relevant industry partners who are saying, hey, look, we want to have like some of our best people come in, second themselves for a year or two, help make NASA, you know, as effective as possible and bring up the next generation of folks.

1:33:26Scott Kupor:And then if people want to go back to the private sector, like that's great. They can go do that.

1:33:29Dillon Rolnick:Yeah. So talk about the type of roles that you need at NASA. When I think NASA, I think frontier science, rocket science, but also the actual astronauts. And then when I think of industry, I think of like the manufacturing expertise, the person that's going to build the rockets repeatedly because there's this beautiful partnership between the private industry and NASA these days. But how does that actually, when the rubber meets the road, who do you need to come join NASA that might not, or, you know, they're perfectly suited to advance NASA's mission right now? Yeah.

1:34:07Jared Isaacman:Well, I'll tell you, there's two categories there that you mentioned that we don't need to worry about. So astronauts, we always get, you know, tens of thousands of applications there. So that's good. And I would even say on science, because when it comes to the planetary science and heliophysics that we do at NASA, we're the only game in town. Oh, yeah. So, you know, in 2028, we're going to launch a nuclear-powered octocopter to Titan to search for signs of life. You're going to want to come to NASA to do that job. But there are areas where we have, you know, some overlap with industry when it comes to building our heavy lift rocket, SLS, where, you know, industry is also doing very similar things, which, to be honest, is a problem.

1:34:49Jared Isaacman:NASA should only be focused on the near impossible, what no one else can do. But SLS is a component of a broader strategy that President Trump laid out with Artemis. We're starting with SLS. Eventually, we'll evolve to other vehicles so that there's an Artemis 100 and beyond someday. But while we have some of that overlap when it comes to our launch pad, turning launch operations, mechanical engineers there, we need propulsion engineers that can be helpful on our vehicles, those that have cryo fluid specializations so we can stop some of these hydrogen leaks, some of these helium flow issues we've had before.

1:35:23Jared Isaacman:Those are areas that industry is extremely well versed because we have the healthiest commercial launch industry that we've had in the history of America's space program. And to get people that want to serve their country, give us a couple of years over at NASA, elevate our talent. Like I said, it's going to make a huge difference in terms of turning our vehicles, launching with greater frequency and achieving the president's national space policy.

1:35:46Dillon Rolnick:Can you walk me through the current top few reasons why the moon is important just as a high level? I want to go back. I'm already on board. But for those people who say, I want to hang out on Earth, why are we going to the moon? Why is the moon important? Why is it valuable?

1:36:04Jared Isaacman:OK, so I'm really glad you asked this question. And there's quite a few answers that go along with it. First, the whole like, shouldn't there be other things that we should be doing? Yeah, and we do. A very small percentage of our resources are actually invested into a national imperative goal like returning the moon. This isn't the 1960s anymore, where it's 4.5 % of the discretionary budget. It's a quarter of a percentage. And you know who's making up the difference? Private industry. Folks like Elon Musk and Jeff Bezos are putting billions of their own resources in play to build capabilities for the good of the American people and really all of humankind.

1:36:39Jared Isaacman:So it's not, you know, the taxpayers footing the bill to do this like it was the 1960s. Now, what are some other reasons that we should do it? How about we made a promise to the American people and have for 35 years? Every president for 35 years has said we're going to return to the moon. It was only President Trump under his first term when he created the Artemis program. Now in his second term, my first day on the job where he signs a national space policy that says not only go back to the moon, but also build out a moon base. Are we taking steps in that direction? And after$100 billion, I think we owe it to the American people to do it.

1:37:10Jared Isaacman:I'll give you national security reasons to do it. 1969, July 20th, America lands, Neil Armstrong, Buzz Aldrin on the moon, brings him safely back to Earth. It sends a message to the world. What else is America capable of doing? Well, if you take 35 years and$100 billion and you fail to do it, well, it sends the exact opposite message. That's a real national security implication because then it says something's broken. and now I'm going to encroach on America's leadership across all the most important technological domains. Here's another good reason. We don't know what we're going to find. I mean, that's the best part of the greatest adventure in human history is going out and learning.

1:37:45Jared Isaacman:Scientific, economic reasons. We will go to the South Pole of the Moon where there's ice. We will use that as a proving ground for power, navigation, communication, in-situ resource manufacturing, the capabilities we're going to need when we someday send American astronauts to Mars. And then we've got to be able to bring them back. And the way you're going to do that is probably making propellant on Mars. I'd rather do that two and a half days from Earth than nine months away from Earth.

1:38:09Dillon Rolnick:I love it. Jordy? Those bars. Fired me up. Fired me up. How do people actually apply to NASA and how broadly are you casting the net? I think you identified the astronaut program being like the most elite of the most elite. We've all seen that one guy who was like a Navy SEAL and then a doctor, and then he became a - Oh, yeah, Johnny Kim. Yeah, yeah, that's right.

1:38:36Matteo Franceschetti:What's he doing now, by the way? Any idea? I don't know.

1:38:39Dillon Rolnick:Is he still an astronaut?

1:38:41Jared Isaacman:Oh, yeah, he's definitely an astronaut. I mean, look, Johnny Kim is a rock star, right? I mean, you know, Navy SEAL, you know, Navy SEAL sniper, then goes and gets his MD. I think he got it from Harvard, right? So then he becomes a doctor. Then he becomes a NASA astronaut. While he's waiting for his mission, he goes to Navy pilot training. becomes a naval aviator. That's amazing. There's nothing this guy can't do. He just came back from the space station. He was up there for nine months. He actually went up on a Russian Soyuz rocket. Wow. So now he's entitled to some time off, although I'm going to do everything I can to convince him to come to HQ and help us out a little bit because that kind of leadership and competence, like we need as much of that as we can get at the top of the organization.

1:39:21Jared Isaacman:Yeah, that's amazing. But I guess your question on where do people go?

1:39:24Dillon Rolnick:Like if I'm not a Navy SEAL sniper and I'm not, you know, I'm just a guy who's hardworking, I'm an American and I want a job at NASA. What are the different roles that are open? What are the skill sets that you're looking for and how do you actually apply?

1:39:38Jared Isaacman:Yeah, I mean, I tell you, we're looking for a lot of engineers and a lot of technicians. There's a lot of ways to learn about opportunities. You can go anywhere from NASA's website to USA Jobs. We certainly have positions open and we're hitting this in a couple fronts. Another area where Scott has been super helpful, one of the biggest surprises when I was on the job is 75 % of our workforce is contractors. Now, they get paid almost exactly the same thing as civil servants. They don't necessarily get all the same benefits, but they're working for companies that have a 40 % gross profit premium on top of it, which actually is about a billion a year, a little more than that, that we could be using for science and discovery.

1:40:16Jared Isaacman:But even more than that, they have different tools for collaboration, for HR systems. Just, you know, you take 75 % of your workforce as contractors and then you combine that with five prime contractors and hundreds of subcontractors building your rockets. Probably no wonder why some things take a little longer than they should, cost a little bit more than they should. So we're converting a lot of contractors back to civil servants. We have this great pipeline of young talent that comes in through our internship programs. We have over a thousand every, I think it's on like a trimester type basis, we bring in fresh talent.

1:40:45Jared Isaacman:But then really, how do you bring up that talent and make them, you know, make them lethal within the domains that we focus on here at NASA? You know, those critical engineering capabilities and science operations and technicians. That's where that's where NASA's tech force comes in.

1:41:04Scott Kupor:Scott, yeah, I was going to say just on the tactical perspective, if anybody's interested today, just go to U.S. Tech Force and we will obviously make sure we have it. And then as we roll out formally the NASA force side of it, we'll have a separate landing page for people and make sure that we can get folks, whether they're coming on the engineering side or whether they're coming from private sector. But easiest way to go right now is just go to U.S. Tech Force, and we will absolutely make sure that we get you in the queue here. Fantastic.

1:41:29Dillon Rolnick:Jared, can you reflect on going to space? How did that happen? What does it mean? Would you recommend it for other folks?

1:41:38Jared Isaacman:I think like a lot of things in my career, I got very lucky. So I certainly, I mean, I definitely can tell you when I was in kindergarten that I said I wanted to grow up and be an astronaut. The chances of that I thought would be, you know, near zero in life. But got very lucky, you know, kind of, you know, reached out at the right time. I mean, this goes back actually 2008. I went to Baikonur, Kazakhstan to see a Soyuz launch with a lot of the early pioneers of the commercial space industry. And I just kept knocking on the door until it was the right time. And then I got lucky to lead SpaceX's first mission to orbit Inspiration4, followed up with a developmental program in September of 2024.

1:42:17Jared Isaacman:We went farther into space. We tested a new spacesuit. We communicated over a beam of light. It was very cool. But I'll tell you that despite the views that I've seen, the opportunities that I've had in space, I've got the best job in the world right now.

1:42:32Dillon Rolnick:Yeah. Do you see part of your mission or part of your legacy being not only the economic opportunity on the moon, the national interest mission on the moon, but actually just allowing more humans to experience space? Is that important?

1:42:52Jared Isaacman:Yeah, well, and they go hand in hand with economic opportunity, right? So I do point out all the time that I actually think one of the most important KPIs is more people living and working in space. uh now for that to happen i mean because i i'm not a believer there's a huge like tourism pipeline in this because it still costs a lot and it'll never be the safety level of an airliner so you're still you know taking a controlled explosion you know 1.8 million pounds of thrust send somebody to 17 500 miles an hour that's you know and when they get there odds are really good they're not even gonna feel good so uh so we gotta we gotta crack the code on the orbital economy um that some days necessitates more people living in working space.

1:43:32Jared Isaacman:But if you're asking what I hope the greatest contribution at NASA, because it's not landing on the moon, that's just luck, honestly, to be here at a time where President Trump gives us the mandate, the resources from the one big, beautiful bill, congressional appropriations. You have all the tools to execute on that. Really, in my mind, it's concentrating our resources here at NASA on the needle movers. You know, sometimes you get, you know, mile wide, inch deep on all the things you're trying to do, but really focus on the things that no one else is capable of doing but NASA. And then empowering the workforce, like the best and brightest across the nation.

1:44:08Jared Isaacman:Get rid of as much needless bureaucracy and things that impede progress. And if you can do that, then things like returning to the moon and building a moon base will be pale in comparison to what we're capable of achieving in the years ahead.

1:44:20Dillon Rolnick:Last question. Bit of a funny one. And the chat asked, what's the best watch to bring to space?

1:44:28Jared Isaacman:I mean, I don't even know if I'm even allowed in this world to endorse a brand. You're going to get yourself in real trouble here. No, I'm not.

1:44:36Dillon Rolnick:I guess anyone that keeps the time accurately. Well, thank you so much for coming on the show. Have a great rest of your day. It's great to be here, Jared, and always good to have you on the show, Scott.

1:44:48Matteo Franceschetti:Yeah, we'll talk to you soon. Thanks, Scott.

1:44:50Dillon Rolnick:Goodbye. Let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it. And let me also tell you about Lambda. Lambda is the super intelligence cloud, building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands. It is, in the government role, there's so many things you can't do. But according to the chat, he has a fantastic watch collection. I was wondering if there was a memento that he brought with him because you know the story of the moon swatch? You know this whole thing?

1:45:25Dillon Rolnick:Yeah, yeah, yeah. The moon watch. Was it going to be a Rolex? Was it going to be an Omega? It wound up being an Omega. Very interesting bit of watch history. We can dive into it at some other time.

1:45:35Matteo Franceschetti:Breaking news from this morning. What happened? Tucker Carlson's shipment of nicotine was hijacked. Very traumatic.

1:45:44Dillon Rolnick:An attack on one nicotine pouch brand is an attack on us all.

1:45:48Matteo Franceschetti:I mean, Tyler asked me this morning. Tyler was like, hey, Josh.

1:45:51Dillon Rolnick:Oh, you think I'm guilty? I'm not guilty.

1:45:54Matteo Franceschetti:You were a little bit late for the gym today, and we thought maybe. Hijacking. A little hijacking.

1:46:00Dillon Rolnick:A little fast and furious. That's what Josh said.

1:46:01Matteo Franceschetti:Billinson says, stealing a truckload of Tucker Carlson branded Nicodine pouches is like the plot to a Zoomer reboot of The Fast and the Furious.

1:46:09Dillon Rolnick:Yeah, yeah, yeah. Have you seen the original Fast and Furious, Jordy?

1:46:13Matteo Franceschetti:No.

1:46:15Dillon Rolnick:It is so embarrassing.

1:46:16Matteo Franceschetti:I've seen I've certainly seen trailers and or seen any Fast and Furious movies any of them

1:46:23Dillon Rolnick:there's so many good ones anyway in and there's there's like 40 of them right yeah and they uh there's like 12 uh they they and they all have like puns as names so there's uh the original movie is called The Fast and The Furious then later they just came out with one that's just called Fast and Furious which is very confusing then they had Fast Five they had Too Fast Too furious which is hilarious uh i was really hoping that for the 10th movie they were going to do fast 10 your seat belts that would have been a good pun but they didn't and then they did the

1:46:55Matteo Franceschetti:fate of the furious f all right dad i got you anyway in the fast and furious they they're

1:47:00Dillon Rolnick:stealing i think dvd players and so they they hijack a truck because they love movies so much they love movies it's it's insider dealing because it's like the movie industry is like we got to promote the dvd players make them you know oh you want these dvd players right well What else you got, Jody?

1:47:15Matteo Franceschetti:Trey says the IWC that Jared wore to space, Jared auctioned off for charity.

1:47:21Dillon Rolnick:Okay, okay. Well, I should have asked a more pointed question about that. Anyway. Patriot. Let me tell you about Vanta. Automate compliance and security. Vanta is the leading AI trust management platform. We have Adam Brey from Skydio. He's the co-founder and CEO in the Restream Waiting Room. Welcome to the show, Adam. What's going on? Great to see you. How are you doing?

1:47:43Scott Kupor:Great to be here, guys.

1:47:44Dillon Rolnick:Thanks so much. Give us the backstory of Skydio. Get us up to speed on Skydio. I'm obviously familiar with the company, but for those who aren't, there's so many places we can go in drones and defense and DJI. There's a million places I want to go, but introduce the company and take us through the shape of the company right now.

1:48:05Scott Kupor:so we are the largest u.s drone manufacturer uh we sell drones to military we sell drones to public safety and law enforcement uh we sell drones to critical infrastructure inspectors uh we serve a number of our allies around the world yeah and the big bet that we made when we started the company was on ai and autonomy so we started in 2014 we've been at it for 12 years yeah um so we met made very big bets on ai and autonomy when they were uh not nearly as obvious as they are today. And we're at a very exciting moment in drones where they're really having their moment. And the way people use drones is transitioning from manually flown to dock-based and remote and autonomous.

1:48:43Scott Kupor:And we're just seeing incredible stuff happen now with our customers.

1:48:45Dillon Rolnick:I remember Chris Dixon, I believe this is the story, showed me a video of a Skydio drone flying around a house and doing sort of a roof inspection. Is that a real business now or insurance agency like walk me through some of the less everyone knows about ukraine what's going on there but like how are drones actually getting used in the american economy today in an industrial

1:49:08Scott Kupor:context uh so that is a real use case okay we have customers that are doing uh insurance claims adjustment with drones so rather than taking out a ladder and climbing up on a roof you just take a drone out and push a couple buttons it automatically digitizes inspects the whole thing the biggest and fastest growing use of our products today is in law enforcement and public safety where there's just been this huge paradigm shift. So law enforcement has been using drones for like the last 10, 15 years, but historically it was typically a pretty small team. They were manually flown, you know, they would get called out every once in a while if there was an event.

1:49:40Scott Kupor:But the paradigm shift now is the drone lives in a docking station on the roof of the police department. It can be flown remotely. And when a 911 call comes in, you click one button on a map and the drone will launch itself. It'll fly there. It'll give you real time awareness. and it just changes outcomes because you can see a crime in progress you can find a missing person we yeah we've got a good one here so this is uh this is close to home this is san francisco police department okay this is called drone as first responder so basically this was a report of a stolen vehicle they launched the drone they're following this guy from the air he doesn't he doesn't know they're there so they knew that people would do this they'd steal license plates after they'd stolen the vehicle whoa they never actually caught anybody in the act uh but with the drone it's it's easy he'll see so now he's got a stolen stolen plate oh he's putting it on the car yeah exactly so he he nicely holds it up so we can read it from the drone okay this is not

1:50:32Dillon Rolnick:an actor this is a real criminal no this is real this wow this happened uh this happened last year

1:50:37Scott Kupor:in san francisco so now he's tinting the window so this this is bad news right this is stolen vehicle cold plate in the vehicle tinting the windows he's going off to do a bunch of bad stuff but they know exactly where he is thanks to the drone they send out a place plain plain clothes unit, they roll a spike strip. Um, so he's got flat tires and that's basically into the story. And this is happening all the time. This is happening like every hour, almost every minute, somewhere across the country, uh, that cops are responding to calls with autonomous drones.

1:51:01Dillon Rolnick:So, uh, take me deeper into that example. Uh, that, that feels like a replacement for dispatching a whole helicopter. That feels like there's probably enough bandwidth at a police department to have someone sort of manually piloting that, even if there's some autonomous systems that are just keeping you from crashing into the walls. But what level of autonomy are we talking about right now for like the most functional use cases?

1:51:26Scott Kupor:So the way that our, our customers talk about it is force multiplier. Um, and it really is a force multiplier because you can have one person sitting in an ops center. They call them real-time information centers. Uh, and they can be flying up to four drones at once. Um, because there's a lot of automation built into the software. So you can just click a button on the map and the drone will fly there. or you could click on a car or a person of interest and the drone will start following it. So it's a highly, highly leveraged activity. And it's also just faster to get somewhere. So rather than taking half an hour for an officer to roll out on the ground and find the address, whatever, it's like 30 seconds for the drone to get there.

1:52:01Scott Kupor:It can very quickly determine what's happening. And yes, you're right. You do have sort of a remote officer who's flying and controlling this thing, but they're just much more highly leveraged. And so the impact is kind of two-sided. One of the ends is what you saw there in that video where it's like high-stakes scenario, prime in progress. We're going to use the drone to change an outcome. The other end of the spectrum is like the most mundane stuff where somebody reports that like there's a car parked where it shouldn't be or something. And it would waste an hour of somebody's time to go check on it.

1:52:29Scott Kupor:But you just send the drone and you can very quickly see what's going on. And so the law enforcement, I think default expectation in really probably like the next two or three years is if you call 911, a drone should show up in 60 seconds or less. And this is happening, you know, San Francisco, New York, Philadelphia, Oklahoma City, Miami, you know, the big cities across the country are all using Skydio. and when a city signs up as a customer,

1:53:00Matteo Franceschetti:deploys your product into the field, is there a lag time before you see, like I'm assuming crime statistics must start to change as criminals realize, hey, I don't have like 15 minutes to kind of get my act together after committing a crime. I have like 120 seconds and then I'm going to be being tracked and it just makes it a lot more complicated. So maybe you end up, you know, getting a real job or something like that. And instead of committing crime, but where do, what, what, what kind of shows up in the data?

1:53:36Dillon Rolnick:Economy booms as criminals get jobs.

1:53:40Scott Kupor:Crime goes down, employment goes up. So like San Francisco has been published this. I think that, I think that crime is down like 30%. Auto thefts are down close to 50%. Wow.

1:53:50Dillon Rolnick:Yeah.

1:53:51Scott Kupor:Since they've rolled this out. So I think, and there was actually like an unintentional testimonial from a guy who, you know, who had lived in this world. You might have seen this. I remember this. We should actually find that clip. Yeah. But he was basically saying like, Prime is over in San Francisco because you're screwed. You know, like you steal a car. The drone's going to show up in 30 seconds. There's nothing you can do. And it is, you know, it's satisfying for me. And I think it's also, you know, for us as a company and for our customers, there's just an asymmetric advantage. Like when there's a drone following you from above, like there's a lot you can do.

1:54:20Scott Kupor:And, you know, I was asking one of our customers, they were showing a video of a retail theft where somebody had lifted 50 ,000 bucks from a store or$50 ,000 worth of goods. And the drone got there. It ID'd the person just as they were getting into the getaway car. They followed the getaway car to the person's residence. Then he just arrested them. I was like, oh, man, this is incredible. Can you use this video in court? And the chief said, there's no way this is going to court. Right? Like, we've got this person pulled. We've got video evidence of the whole thing from the drone. And so it's also just way more efficient.

1:54:48Scott Kupor:Yeah, way more efficient for the system. You don't have to go through like a long trial.

1:54:52Dillon Rolnick:Because the defense attorney shows up and doesn't say, oh, there's a bunch of wiggle room. They just say, okay, they got your gender rights. Like you should just plead guilty, plead it out and just do the crime. You did the crime. Talk to me about what's on an exponential and what's on a linear trajectory. What I mean by that is AI progress, clearly on an exponential. when if you're trying to read that license plate, I believe that you're going to get better at reading that license plate twice as better. Like the quality of your AI systems is going to just double constantly. Probably same thing with your production capacity.

1:55:26Dillon Rolnick:You're probably exponentially growing the amount of drones that you can make. But battery technology, range, flight time, security, like what is still not on an exponential?

1:55:39Scott Kupor:It's a really interesting question. So the way that I kind of think about it is we have a similar kind of dynamic to the self-driving car companies where we're building this really hardcore real world technology that has to deal with the full complexity of what the real world can present. Different weather conditions, different operating paradigms, different visual lighting conditions. And a lot of these, you know, AI is on an exponential curve, but I also kind of think the real world is exponentially complex. And so there's this very, very hard real-world grind that you have to go through to make this stuff work reliably and scalably.

1:56:15Scott Kupor:I mean, you can kind of think of our drones like self-driving cars that fly. And that is the hardest piece. And we've been at it now for 12 years, and there's really, unfortunately, no shortcuts. I mean, it's taken us a long time, millions of flights, tens of thousands of drones we've built to learn a lot of these hard real-world lessons and then incorporate them into our hardware and software to make the system reliable. So we're benefiting now. We made big bets on computer vision. That turned out to be right. We're on a great exponential there. But the visual complexity of the real world is also exponentially high.

1:56:47Dillon Rolnick:Yeah, walk us through the importance of dock-based drones. When did you start building docks? Because I imagine you started with drones and then eventually realized that you need to actually sell the infrastructure, that there aren't just going to be sky ports with charging infrastructure that just pop up everywhere. So walk us through what we're seeing in that video and sort of the rollout here.

1:57:12Scott Kupor:Yeah, so that's a dock in action. So vertical integration, there's different ways you can approach it. I think I'm a big believer in vertical integration, like building the dock and the drone and writing all the software that runs on it and writing all the cloud interfaces to manage it. Like all these pieces need to work together to deliver kind of this solution. Like at the end of the day, we're kind of a boring enterprise solution to our customers. There's just a bunch of cutting edge robotics under the hood to make it possible. And it's been a journey to get to where we are. We were talking about docs when we started the company in 2014.

1:57:41Scott Kupor:One of the things I believe is you got to have for these really big technology plays, you kind of need you need an incremental business path along with the incremental technology. Like I think the biggest success is Tesla did this. SpaceX did this. You've got the grand vision, but you've got to find ways to make incremental technical progress and unlock incremental business value along the way. So building the drone without the dock is an obvious example of that. I think in 2019 is when we really like went all in and said, like, we think these enterprise markets are ready. We think the dock is the way to serve them.

1:58:08Scott Kupor:So it's been a six, seven year path from then to now have our products really being produced at scale, reliably working at scale. And, you know, there's a lot of little problems, little and big problems you got to solve along the way to make it work.

1:58:22Dillon Rolnick:Yeah. Talk about a little bit about privacy, surveillance, regulation. Can you spy on me? Can the government spy on me? I love the idea of finding my car if it gets stolen, but I'm a little bit hesitant about having a drone follow me around everywhere for no reason, just in case I go two miles an hour over the speed limit. Like, how do you grapple with those tradeoffs, the regulation, how you think the government uses this technology?

1:58:49Scott Kupor:So I think these are civilizational questions, and I think ultimately they are like policy and regulatory questions. this is a hot topic today what's the role of companies and how their government customers use it so we have a voice in this conversation we have a strong posture I think transparency is basically the key especially in law enforcement because if the police are transparent and if people see you guys hear me? I might just lock my earphones

1:59:17Dillon Rolnick:if you take them out you can probably just switch to the computer audio

1:59:21Matteo Franceschetti:we can hear the hum of your Your massive facility behind you.

1:59:26Dillon Rolnick:Workshop makes it sound like it's like Santa's workshop.

1:59:30Matteo Franceschetti:It is Santa's workshop.

1:59:31Dillon Rolnick:It's a manufacturing facility, Jordy.

1:59:34Matteo Franceschetti:They have elves.

1:59:35Scott Kupor:The biggest drone factory in the West.

1:59:37Dillon Rolnick:Yeah, we can hear you now.

1:59:38Scott Kupor:Okay, great. You're good. Yeah, so I think transparency is the key because, you know, if people see what law enforcement is doing with drones, they can form their own assessment, and there's natural accountability built in there. And really one of the things that I think our customers are doing an amazing job of is being maximally transparent on this stuff. When an agency launches their Growness First Reporter program, they're not trying to hide it in the back room. They're hosting a press event. They're explaining what they're doing. They're showing videos like the one that I just showed you.

2:00:07Scott Kupor:And we actually have – we do a bunch of transparency features in the product. So we have a transparency portal that basically makes it easy for an agency to publish all the flying that they're doing. And citizens use this. You can log on and you can see, like, oh, were they following me around? to see if I was going through miles of hours. No, they were like busting somebody who stole my car. And this is kind of a side, but one of the things that I've come to appreciate working with law enforcement is there's very high direct accountability there because these police chiefs and sheriffs, I mean, they are either directly elected themselves or they're downstream from an elected official.

2:00:40Scott Kupor:And if something goes wrong, they're on the nightly news explaining it and they don't want to be there. And so they take this stuff really seriously. So it's not to say that there aren't legitimate concerns and that everything's going to be perfect, hunky-dory. But I think the feedback mechanisms are in place to get us to a healthy place. And the safety benefits are just incredible. Like when people see, like the one that I showed, they tend to get it.

2:01:01Dillon Rolnick:Yeah, I've been pretty satisfied with the way local jurisdictions have been able to decide as a small community, whether or not they want license plate readers or whether or not they want, you know, doorbell cameras to be accessible. And it hasn't been something that's been forced upon every community or banned from every community. It's been on a city by city basis. And that feels like the correct formulation structure based on what we know we're entitled to as Americans.

2:01:33Scott Kupor:It's exactly right.

2:01:34Dillon Rolnick:Yeah.

2:01:35Scott Kupor:Like, sorry, just on that note, like when you're selling to public safety and law enforcement, it's basically a community sale. Like all this stuff goes before city council before it gets approved yeah and and they're the ultimate people that are carrying the purse string so that's agency knows that we know that and that's part of the test they're

2:01:50Matteo Franceschetti:they're our ultimate customer we gotta talk about dji dji got added to the fcc's covered list and the end of last year how well they just long time over just had a cool feature where i guess like

2:02:02Dillon Rolnick:anyone can access any of them so like if i have one like yeah we had the guy we had the guy that

2:02:08Matteo Franceschetti:hacked, that basically his way into being able to control and look at the video feed of 7 ,000 DJI vacuums.

2:02:14Dillon Rolnick:He found the back door.

2:02:16Matteo Franceschetti:Yeah, he found the back door.

2:02:17Dillon Rolnick:And then he hacked into a vacuum and drove it to the back door of the house that he was in.

2:02:21Scott Kupor:Anyway, you know, sometimes I say that Chinese drones come pre-hacked, but

2:02:24Dillon Rolnick:pre-hacked, yes.

2:02:25Scott Kupor:No hacking required.

2:02:27Dillon Rolnick:But yeah,

2:02:27Matteo Franceschetti:when you're expecting that. But how has the FCC's action impacted your business? I'm assuming that's something that you felt would have been an obvious decision for the government for many years prior, but give us a history.

2:02:42Scott Kupor:Yeah, look, I think that part of what's happened here is like when we started the company in 2014, drones seemed like toys. You know, they were like kids toys, people flew them. And they've been through this evolution from like toys to like useful tools for expert operators to now, I would argue, like critical infrastructure. And then the conflict in Ukraine, or in Ukraine, I think just demonstrated to everybody how important these small light drones are from a national security perspective. So I think there's been kind of just increasing awareness over the last 10 years that this is critical technology with very high national security stakes.

2:03:13Scott Kupor:And, you know, from our perspective as a company, like our goal is just to win on the strengths of the product and the tech. And I think the shift to dock and autonomy really plays to our strengths, both as a company and as a country. And you can see this in the market. Like historically, DJI had like 70, 80 % market share in law enforcement. And in the dock world, they're still our number one competitor. So like Flock Safety is effectively a DJI reseller. Like all of their drone is first responders work is done with DJI drones today. But the scoreboard is flipped. You know, we are winning the majority of these deals, even before the FCC action on the strengths of the product and the tech, because it's just sort of a different recipe it's similar but it's a different recipe to deliver the solution price people are a little bit price sensitive because the value is so high you know price is probably dji's greatest strength so look from our standpoint like our goal is just to have the best products and and tech in the world we certainly benefit um from policy and regulatory tailwinds and you know i'm i'm not neutral on this we have a business stake in it but i think there's a lot of reasons why we want a domestic drone industry um and why that's important for our national security.

2:04:27Scott Kupor:I mean, this stuff is critical for military, it's critical for infrastructure, it's critical for public safety. And so it's exciting to see like huge momentum now towards domestic industry and domestic manufacturing. But ultimately, the only stable long-term solution is to have the best products in tech designed and built right here in the US. And that's what we're doing. That's what the factory behind me is doing. I think this new chapter of drones as infrastructure is a huge opportunity for us as a country. And one that we have -

2:04:53Matteo Franceschetti:What are you thinking about scaling manufacturing capacity over the next decade?

2:05:00Scott Kupor:I mean, we're all in on it. It's a fun set of problems. I mean, it's all engineering of some flavor. It's a different set of engineering problems. So we're adding more and more automation to our factory. We're investing in the systems and processes. I think that the combination of AI and robotics changes the way manufacturing works. It's possible to automate more stuff. It's possible to have what happens on a factory floor be more software defined. I think that's going to be more to our advantage versus just pure low-cost labor. That's not our strength in the U.S. We have phenomenal workers playing the drones, but it's a different labor equation than it is in Asia.

2:05:38Scott Kupor:We're investing very heavily in the software aspects of manufacturing and the automation around it. It's part of our core identity as a company, and we're continuing to scale. So we're in an 80 ,000 square foot facility now. A year from now, we'll probably be in a 450 ,000 square foot facility.

2:06:01Dillon Rolnick:It's a huge facility. Huge facility. Congratulations. And thank you for taking the time to come chat with us. Good luck with the continued scale up.

2:06:11Matteo Franceschetti:Great to finally get that.

2:06:11Dillon Rolnick:And I mean, I think I met you over a decade ago. So a true overnight success. Overnight success. Yes. Come back on when you have news, Adam.

2:06:22Matteo Franceschetti:Yeah, we'll talk to you soon. Really, really cool to get the update.

2:06:25Dillon Rolnick:Have a good one.

2:06:26Matteo Franceschetti:Cheers. Thanks, guys.

2:06:27Dillon Rolnick:Let me tell you about Cognition. They're the makers of Devin, the AI software engineer. Crush your backlog with your personal AI engineering team. And you know what? I got another ad read. Fin.ai, the number one AI agent for customer service. Did I mess it up? No, it didn't trigger. Fin.ai is the number one AI agent for customer service. If you want AI to handle your customer support, go to fin.ai. And without further ado, we have Mateo from 8th Sleep back on the stream his fifth, sixth time. We can't stop talking to this guy. How are you doing, Mateo? Looking rested.

2:06:59Adam Bry:Good to see you. Oh, good. Great to see you guys.

2:07:03Dillon Rolnick:I have absolutely loved my 8th Sleep. I continue to sleep on it. Negative 10 on my side, plus 10 on my wife's side. My five-year-old has a feature request. He says, where's my zone? I sleep in the middle. And we're like, you don't sleep in the middle. You sleep in your bed. But I'm sure you get that all the time. But how are you?

2:07:24Adam Bry:All good. How are you doing, guys? I'm really happy to be back here with some good news.

2:07:29Dillon Rolnick:Always great to have you. Give us the news. What happened?

2:07:34Adam Bry:We just announced a new round. We raised another$50 million,$1.5 billion.

2:07:41Matteo Franceschetti:Woo!

2:07:46Dillon Rolnick:Congratulations.

2:07:46Adam Bry:And who'd you raise it from? From Tether Investments. So the Tether company, they are doubling down on AI and health. And I have been friends with Paolo, the CEO, for a long time. He's a really big fan of us. And very quickly, he said, okay, how can I accelerate you guys even more? And we were able to close the deal very quickly.

2:08:14Matteo Franceschetti:Amazing. give us an update on the business overall, everything you're focused on this year, all that good stuff yeah, well the best

2:08:23Adam Bry:part is in between now in 2025 we raised twice right, including this round but the best part is we were profitable, so that's the part I love the most, we didn't need money, but you know when you have the right investors that they can't really push you further, it makes sense from a strategic standpoint So we were profitable. We launched three products last year. Now we sell in 34 countries. We're going to launch in China in a matter of a few months. So we'll be in China as well. And then we will launch Brazil as well this year. More hardware products coming. As I say, strong unit economics. A lot of people loving us.

2:09:04Adam Bry:We just need to keep shipping.

2:09:06Matteo Franceschetti:What actually goes into an international launch for 8sleep?

2:09:14Adam Bry:For 2026?

2:09:15Matteo Franceschetti:Yeah, I mean, like, you know, it's a heavy product. It's not the easiest thing to ship in the world. You know, some brands would have entered Brazil or China a long time ago, but you guys are going kind of country by country, clearly focused much more strategically, maybe.

2:09:34Adam Bry:Yeah, we almost wanted to prove the strategy even at the board level. So we started with Canada, and then from Canada we went to Australia, and then Europe, then UK, then we launched Middle East, which was super successful. Then we launched Mexico, which was super successful. And then Saudi Arabia, Singapore. Now the next big bet is China, which will happen in around April. And then Brazil. Brazil is quite complicated. The interesting thing is we sell everything direct to consumer. So we don't use distribution stores or we don't go into any retail store. So we control everything. That makes things faster.

2:10:10Adam Bry:But at the same time, There are still a lot of requirements also in terms of approval as a consumer electronics device. So usually that is the thing that requires the longest. And then the supply chain and the shipping. But that is honestly fairly tricky.

2:10:28Matteo Franceschetti:What key kind of health trends are you tracking? I have a question about that.

2:10:33Dillon Rolnick:Are peptides changing sleep at all?

2:10:36Matteo Franceschetti:Yeah. I think there is one peptide that specifically targets sleep. I'm just curious. Yeah, yeah, yeah.

2:10:41Adam Bry:So we are running some clinical studies, but yeah, there are a couple that seem promising for that. And then there are also some drugs about sleep apnea that seem really effective. They were not approved yet, but there are certain pills that it seems they can really help with the mitigation of sleep apnea. And then we filed for FDA approval, both for detection and mitigation of sleep apnea. We invented a new technology that mitigates sleep apnea without you wearing anything. I think it could be a massive home run for the business. Obviously, we need to get to the finish line and prove it and get the FDA approval.

2:11:22But if that happens, it will be a game changer for millions of people.

2:11:26Adam Bry:You know, 40 % of people with sleep apnea, they don't even know they have it. And this is a life-threatening disease for certain people. And I think we have a really nice solution that is extremely convenient, seamless. they will have millions or billions of people.

2:11:44Dillon Rolnick:Billions. 93 last night.

2:11:47Adam Bry:Oh, how'd I do?

2:11:48Dillon Rolnick:That's good. 95 % quality, 100 % consistency, seven hours and 18 minutes. Not too bad, not too bad.

2:11:55Matteo Franceschetti:91.

2:11:56Dillon Rolnick:Oh, let's go. But doesn't it adjust?

2:12:00Matteo Franceschetti:Isn't it possible that I got a better night's sleep than John last night, but he on average gets worse sleep. And so he put up an okay, a good night for him. John's best night's sleep is a nightmare for me.

2:12:15Dillon Rolnick:Yeah, yeah, yeah. But take me through. What does it take to get me to 100? I know that there are other pieces of the puzzle. I have the mattress cover, of course. But should I be taking supplements? What else do you have for me that could get me to 100 more regularly?

2:12:35Adam Bry:Yeah, so I would think in three-dimensional. On one side, there is consistency. So going to bed at the same time, waking up at the same time.

2:12:43Dillon Rolnick:I'm very consistent about drinking like a full bottle of wine before bed.

2:12:48Adam Bry:Perfect. That's not going to help with your school and your HRV.

2:12:53Dillon Rolnick:I just black out and then I wake up and I'm like, I fell asleep. It was effective. It made me sleepy. Do not drink before bed.

2:13:01Adam Bry:I give a presentation to a bunch of entrepreneurs, and some of them start challenging me and say, yeah, I know that you keep saying the thing of the alcohol, but every time I have a couple of drinks, I pass out. But then I believe alcohol is not bad for me. And I say, look, man, we can look at your data. You can look at your data.

2:13:21Dillon Rolnick:Yeah, I got the data.

2:13:24Adam Bry:But statistically, it's not good.

2:13:26Dillon Rolnick:Yes. But on the supplement side, what's interesting? Melatonin. And give me kind of sort of your thesis about the other things that impact sleep. Yeah.

2:13:36Adam Bry:Honestly, the one that works is well known is melatonin. The problem with melatonin is you cannot keep taking melatonin for too long. That's right. So usually we have run a bunch of studies. And what we have seen is you should keep it below 0.5 milligrams.

2:13:52Dillon Rolnick:Okay.

2:13:53Adam Bry:That otherwise it creates dependency. Then when you're traveling or you want to adjust for a leg, you can go up to 2 milligrams, something like that, but you should stop after 3-4 days. Not much. A lot of people take 3-5 milligrams every single night forever.

2:14:09Dillon Rolnick:Yeah, and then they build intolerance and they don't get anything out of it. You can't just drink caffeine every single day and expect to have the same results.

2:14:15Adam Bry:Exactly. Then some people report some benefit with magnesium, extravagandha, and some of that. but usually the benefits are fairly minimal. Right now, the only one that is very well proven to work is melatonin. Yeah.

2:14:31Dillon Rolnick:How many hours of sleep data do you have? Me? Yeah. At eight sleep, how much data are you studying these days?

2:14:42Adam Bry:Oh, we cross billions of nights.

2:14:45Dillon Rolnick:Billions of nights.

2:14:49Matteo Franceschetti:He just wanted an excuse.

2:14:53Adam Bry:The interesting part is what excites me is because we sell in 34 countries now, it's so interesting to really think that we are powering the sleep all across the globe, different genders, different everything.

2:15:08Dillon Rolnick:Can you tell us what country sleeps best?

2:15:10Adam Bry:Australia. Australia sleeps best. There is a specific place in Australia on the coast where they sleep like babies. They're like unbeatable. That's amazing. Americans don't sleep very well in the grand scheme of things. What's the thesis there?

2:15:30Matteo Franceschetti:They're just by, they're hanging out at the beach, they're super relaxed.

2:15:33Dillon Rolnick:They're lazy.

2:15:35Adam Bry:Yeah, I think so. is probably just a local culture where they're still active, but they're probably way more consistent and they have more time for sleep. Another interesting thing is sometimes you see weird spikes in a good or bad direction for sleep. We published a long time ago how sleep dipped during the election, election day, or there are all these studies about when you change the time forward or backward. Daily savings time or probably the Super Bowl.

2:16:10Dillon Rolnick:There's a whole bunch of different dates. Super Bowl, all that kind of thing.

2:16:14Adam Bry:So, yeah, it's pretty funny.

2:16:16Dillon Rolnick:And video earnings, probably people losing sleep. Yeah.

2:16:19Adam Bry:Even after COVID, a lot of things changed. People started sleeping way longer after COVID. Really? Particularly during COVID, they were sleeping like an hour longer on average per night, which was pretty impressive. And still today, you see the difference across even the different days of the week.

2:16:34Dillon Rolnick:Wow. Yeah, that's impressive. Well, thank you so much for taking the time to come chat with us. Congratulations on the round. On the new round. And we will talk to you soon, Mateo.

2:16:42Matteo Franceschetti:Yeah, it's great to see you as always.

2:16:44Dillon Rolnick:Thank you, guys. Goodbye. Let me tell you about CrowdStrike. Your business is AI. Their business is securing it. CrowdStrike secures AI and stops breaches. And let me also tell you about Cisco. Critical infrastructure for the AI era. Unlocks seamless real-time experiences. and new value with Cisco. And without further ado, we have Dylan Rolnick from Noose Research coming on the show. What's going on? Dylan, how you doing?

2:17:08Jared Isaacman:Yo, great to meet you guys. Big fan of your show.

2:17:10Matteo Franceschetti:Happy to be here.

2:17:11Dillon Rolnick:Thank you so much for taking the time. Since it is the first time on the show, please introduce yourself and what you're working on.

2:17:18Jared Isaacman:Yeah, I'm Dylan Rolnick. Like you said, I'm the COO. Broadly, I would say Noose Research is working on being a super internet native AI lab.

2:17:28Dillon Rolnick:Okay.

2:17:29Jared Isaacman:Essentially just focused on bringing open source to the forefront and all we can for open source AI.

2:17:34Dillon Rolnick:Yeah. Talk about agents, harnesses, conductors, orchestration. How do you envision this year playing out for how people, your customers, your users, will interface with AI? Yeah.

2:17:55Jared Isaacman:Yeah. Well, so I mean, I think this year has already been pretty heavy on agents. We did our, I mean, claw, right? Open claw hours. We just put out Hermes agent. So I think people are just starting to get used to a kind of different way to interact with AI that maybe they're not as used to that's like blowing people's minds. So if you're like super close to it, some of the stuff might not be that shocking. But I think you saw with how things became very popular, very fast. People getting a first glimpse of like putting an AI out there, kind of like letting it do its thing and coming back to something that's finished or done like a work product.

2:18:26Jared Isaacman:Baby Keem with the rap game.

2:18:27Matteo Franceschetti:Focus on open wall.

2:18:28Dillon Rolnick:Baby Keem is focused.

2:18:29Matteo Franceschetti:He's stacking Mac minis.

2:18:32Dillon Rolnick:I did, yeah. It's crazy. We got to convert him.

2:18:34Matteo Franceschetti:That'd be a big win for us.

2:18:35Dillon Rolnick:Yeah. Yeah. Humble the model. Ridiculous.

2:18:39Jared Isaacman:So anyways, yeah, I mean, like, we think it's super powerful. We put out our Hermes agent, like I said, sort of an alternative kind of, you know. We view it as sort of taking the best of what you would see in a coding agent and what you would see in something like OpenClaw. And then we added a bunch of capability that's important to our own work to advance the stuff that we're doing. Because, like, if anyone's going to use an agent to advance their work, like, it sure as hell is going to be us first.

2:19:02Dillon Rolnick:Okay, so when you say advance work, but it feels like there's still some consumer angle. Like what are the textbook use cases that you're seeing?

2:19:12Jared Isaacman:We're just starting to get some feedback on what people are doing. But I think, you know, from the professional standpoint, it's like people want a way to just have a task be put out there and they come back like they're messaging with it on Telegram.

2:19:25Scott Kupor:They're saying like, I'd like for you to do some research for me.

2:19:28Jared Isaacman:And then by the time I get back to my desk, I'm able to look over that kind of thing.

2:19:31Matteo Franceschetti:Sure.

2:19:31Jared Isaacman:But from our vantage, we put in like, you know, we have traditionally made our name basically doing post training. so all the tools that we've been using we've been trying to feed into that tool so that this thing could actually help us advance that goal as well.

2:19:45Dillon Rolnick:So are you building on top of open source pre-trains then? Yeah.

2:19:49Jared Isaacman:So what we did to initially gain some popularity was Lama was putting out base models we fine tuned those into something that we thought would be something the market would want which is why I'm excited about the agent stuff because we were kind of sitting there being like okay here's what we think is a gap between what models are doing presently and what we could add to the fine-tuned world. But once you actually have an agent out there that's touching people and doing economic work and helping, you get more feedback as to what it is that you actually need to post-strain. So for us, it's kind of like, now we have this thing out there, can we shape an open-source model into something that makes the best out of this harness and is cheaper or better across some vectors?

2:20:31Dillon Rolnick:How are you thinking about go-to-market? The open-source piece is interesting, but what are the other levers to get people to adopt the product?

2:20:39Jared Isaacman:Yeah, I think just bludgeoning people with how you can use this thing is important. I think there's a huge gap between what people know about AI and what AI can actually do. Like even some of the people that I consider some of the most savvy, non-technical users of AI are continually blown away by what Hermes Agent can do. And Hermes Agent, again, is just like a harness on other people's agent. It's not even like the compounding effect yet. So I think like making that super easy for people to dive into and get.

2:21:07Dillon Rolnick:So, yeah, what are you actually like pitching them? Because like, you know, ChatGPT comes out, everyone's like, this can do your homework. Like this will replace like Wikipedia. This will replace Google search, right? Like go here and instead of just typing, you know, tell me the history of the Roman Empire in Google, you just do that in ChatGPT and you get something that looks different and is better in a bunch of different ways. Like what's your pitch for like the first prompt that they should fire off?

2:21:35Jared Isaacman:Oh, that's a good one. I think the, well, so the first part of the pitch is like, I think open source is crucial for a lot of people and they might not realize it yet. But I think one of the things I hear from folks that use any agent is like, I'm willing to spend a couple bucks, but maybe not like, you know, a hundred bucks a day.

2:21:50Dillon Rolnick:Yeah.

2:21:51Jared Isaacman:And Claude, while extraordinarily capable is expensive because it's just like omni-capable. so I think the ability to like hone in on the agentic capability at a cheaper price point is definitely part of the pitch and I would say just like get in there figure out what you spend the most amount of your time doing that you don't want to do and ask Hermie's agent to solve that for you

2:22:12Dillon Rolnick:I think we need some more specificity if we want to pitch these people it's it's tough people need a killer use case they need their hands held uh and uh there's certainly there's certainly opportunity out there, but hopefully, you know, this will evolve. I'm always, I'm always taken by that example of mid journey, where if you gave David holes, mid journey, he said, if you give people like a blank box and just say prompt, whatever they want to prompt, they will just type dog and they'll just get like a picture of a dog. And they're like, I could have gotten a dog on Google images. Like this doesn't do anything special.

2:22:45Dillon Rolnick:But when they went into mid journey, they saw everyone else prompting, they say space dog on Mars with a, you know rocket ship blah blah blah and then they would they would remix and and enjoy how are you thinking about bringing the different community together the different hermes agent i i was on instagram and i saw someone who used open claw this is just like a family friend uh but used open claw to build a new uh photo album piece of software to collect photos from all of their different family resources so they have some in google drive they have the the grandparents use some Microsoft product, and they wanted something that would synthesize all this together into one central place, tag them all, do all of this.

2:23:28Dillon Rolnick:And I was like, that's an interesting use case. I like that she built that. And that's something that I could maybe say, hey, maybe I want that. And maybe that becomes a product one day. But how are you thinking about bringing the community together?

2:23:40Jared Isaacman:Yeah, that's a good point. So here's what I'll say is when we put this out, we were kind of thinking of it as like a gift to the community. like we're very open source focused like i said like let's just see what people can make use of this thing and we're probably in that discovery phase of like yeah what is it that people are doing so we had a bunch of immediately cool stuff that i thought came out and then we're putting together a hackathon right now to try to get more of that like you want to see like you said you want to be inspired by what other people are doing so we're putting together a hackathon that's going to start tomorrow online let everyone do i think we're in play mode like do the coolest stuff that get in 100 and let's figure out what exactly hits yeah i mean it is like the best time to go to a

2:24:18Dillon Rolnick:hackathon to host a hackathon like we are truly at this like incredible moment where the tools i

2:24:23Matteo Franceschetti:would use the agent to figure out homes in my area that look like castles because i'm very inspired

2:24:30Dillon Rolnick:by our oh yeah somewhat of a castle this is a cool building yeah i'm in a castle but yeah i mean you

2:24:36Matteo Franceschetti:might not want to give that away because that could be a hit for the hackathon you're gonna

2:24:42Dillon Rolnick:Terraform, he operates his business out of a castle in Burbank. It's amazing.

2:24:48Jared Isaacman:It makes you feel very secure and strong.

2:24:50Dillon Rolnick:It has, like, true turrets. It was built. It's not really, like, a defensible castle. I think it was built more for maybe just a crazy person or something, maybe for the film industry or something. But it looks fantastic. And your office looks fantastic as well. Congratulations on all the success. Thank you for taking the time.

2:25:04Matteo Franceschetti:Shoot us a note on who ends up winning the hackathon, and we'll cover it.

2:25:09Dillon Rolnick:That would be awesome. I'm sure you'll be talking about it as well. Appreciate that. Anyway, Dylan, have a great rest of your day.

2:25:13Matteo Franceschetti:Great to meet you, Dylan.

2:25:14Jared Isaacman:We'll talk to you soon. Great to meet you guys. Have a great one.

2:25:16Dillon Rolnick:Goodbye. Let me tell you about public.com, investing for those that take it seriously. They've got stocks, options, bonds, crypto, treasuries, and more with great customer service. Did you know, according to Tyler Cowen, this is how people talk, according to Sardius, brutally sweater-mogged by boomer, foy, doom, slop, maxing on change-gooning your sweater is not ready for. That sentence, it's like it completely escaped velocity to the point where I don't even understand it and I follow most of this lingo pretty closely. But I appreciate that Tyler Cowen has just reflected on the fact that the vernacular is fully changing on the internet and it is in many ways indistinguishable or indecipherable.

2:26:00Dillon Rolnick:It's a mess. It is its own form of slop, the looks maxing lingo. So the A &W CEO has joined the burger wars. This just continues and continues. I mean, I guess it's free press if you're in the fast food industry. I didn't know that A &W was even a store. I didn't know that they sold burgers. I thought they just had root beer, A &W. Have you ever been to an A &W, Jordy?

2:26:28Matteo Franceschetti:Never heard of it. Never heard of it. Never, ever heard of it.

2:26:34Dillon Rolnick:You mentioned it, but here is the picture. with beautiful lighting. Baby Keem says three more to go, two sparks on the way. Is that the NVIDIA spark?

2:26:44Matteo Franceschetti:DJX spark.

2:26:44Dillon Rolnick:DJX spark. That's an expensive computer, right? He's really wiring everything up. 10 ,000 likes. He's having a great time. I appreciate that Baby Keem is having a good time in the open source AI agent world.

2:27:01Matteo Franceschetti:Sudo says, developers are cooked. I vibe coded a worse version of an existing product in four weeks with serious security costs.

2:27:09Dillon Rolnick:Yeah, it's a wild time. Be careful out there. The other interesting thing that's going on is like there's new data on serious, serious traffic declines across tech publications. This is from Danny Crichton. It says, no discussion of tech media can get past this basic traffic fact. In the AI world, Google and social no longer refer traffic. which means that the vast majority of readers just never find you in the first place. This is, of course, a reaction to Trey Stevens' proposal that maybe he should buy Wired. And it's unclear if he's just having fun. But there's been a call from the tech community like, bring back the Wired we read in the magazines as a kid.

2:27:56Dillon Rolnick:It used to be so techno-optimistic. Now it's this hypercritical, maybe semi-political outlet. But what Danny is pointing to is that these publications have seen just in the last two years since 2024, traffic declines of between 30 % and 97 % for digital trends. The interesting one there, I think it was, of course, Wired. In November of 2024, Wired was putting up 7.7 million views in traffic. Now it's down at 2.9. And there's another one on here. I had no idea that Mashable was still so big. Mashable has actually done pretty well. 16 million two years ago. Now it's down at 11 million, not too bad.

2:28:44Dillon Rolnick:But ZDNet, that was a website where I would go and The Verge is on here too, 5 million views down at 7 million.

2:28:52Matteo Franceschetti:Yeah, I wonder how much of this is due to AI overviews and Google search results now are more paid links than organic links. There's a lot of that. There's a lot of something happens and just gets screenshotted and shared on X. The screenshot, there's a hundred. For each one of these publications, there's a hundred or a thousand people that will take whatever news and just make it more socially native. So I would imagine that the stories these companies are telling are getting a similar amount of impressions overall. That's just not happening.

2:29:32Dillon Rolnick:Potentially more because a lot of the stories that they write about, at least the ones that do scoops, are more interesting than ever. The Verge has had some great reporting. Alex Heath, who formerly worked at The Verge and is now writing at sources.news, his substack, says, More viscerally, these traffic declines represent directionally aligned revenue declines. And so if you see a publication that's down 85 % in traffic, they might be down 85 % in revenue. And that is going to completely change the way they think about the business, both from a staffing perspective, but also from a content and editorial perspective.

2:30:12Dillon Rolnick:Because if you are seeing traffic decline in one niche or one category, you might need to go viral by doing a listicle or doing something that doesn't feel like your roots. Well, you know, the roots have been dug up in some ways on some publications.

2:30:27Matteo Franceschetti:We're not seeing a decline in VC podcast viewership. That's true. There are a number of VC podcasts that are racking up a million views for interviews with Series A or Series B founders. Yeah. Very, very impressive numbers. There are some allegations. It's a white pill.

2:30:46Dillon Rolnick:A lot of people think that, you know, technology is niche. Some Series A founder that you haven't heard on, they can't attract a million people to sit down and watch them on Zoom talk about their business for an hour. But apparently they can. And many venture capital firms have showed us that it's possible. You do have to pay the people to watch. And the views might be a little low quality. But if you pay the viewers. And people won't comment. Yeah, they won't comment. That's an extra bill you've got to pay. And it's very unclear why they aren't buying comments. Because if you're buying the views, just buy the comments too.

2:31:19Matteo Franceschetti:The reason is that they are running paid traffic on Meta at their own YouTube videos. Oh, is that what's going on? That is actually how they're getting the views. The Meta ads, as of a couple of days ago, were public. You can see them. They're just like running. If you're on Instagram, you'll get an ad for a podcast. Oh, that was on Meta? Because apparently - And Google.

2:31:40Dillon Rolnick:Because you can actually just run ads for your content on YouTube. And then you get people who are in YouTube to click on that content. Now, are they less likely to comment? Absolutely. Like the example on the right is Dwarakash. He got a million views and it had 4 ,700 comments. I went back and checked some of my videos just to kind of know like what is the correct ratio? And if I got a video with like a couple hundred thousand views, like there were probably a thousand comments on there for a variety of reasons. Also just people saying like, when the real fans show up, they say like first, second, oh yeah still watching in 2026 you know people will just do this um and uh um uh very very odd uh um just to go by views because it doesn't create a lot of value and it sort of sets you up for this type of dunk although matt turk is being nice and not calling out the particular podcast that uh did this uh max ferens who runs he's the general manager of the dorkesh patel podcast says deep down i know these these are pictures of screens but it really looks like you printed these out mark them up with a pen and highlighter and then took the photos max is so is so funny and uh he he had a bunch of things he says uh uh he was quoting this and said you have no idea how long it takes me to write 4 ,783 comments like he's you know responsible because of course he's not faking the numbers at that scale uh that's not what's happening what else

2:33:09Matteo Franceschetti:Kyle Corbett is highlighting some news. The New York legislature is rapidly pushing through a 2025 bill that would prohibit LLMs from providing substantive legal analysis or advice in New York.

2:33:27Matteo Franceschetti:It seems that LMs could still provide substantive help to lawyers under this bill as written. Consumers, however, would be stuck with the chatbot, refusing to answer their legal questions. And so, again, this just seems bad. The entire point of LLMs and AI is to give intelligence to anyone, right? Like the fact that an individual can get the equivalent of five hours of legal work in a single prompt by like uploading a contract is very good for consumers. I don't even know how bad it'll be for lawyers, especially given that at least right now, even if I could generate entire legal docs, I still want a lawyer to play a part in it.

2:34:11Matteo Franceschetti:But again, Kyle Corbett says, I sold my company last year. GPT-5 Pro was at least as valuable in that negotiation as our attorneys who billed us$450 ,000. This is a terrible law.

2:34:27Dillon Rolnick:The Primeagin is chiming in. I saw him on a podcast recently and I just really like the way this guy talks. I've been getting a bunch of his Instagram reels too. He's a great streamer. Responding to former guests of the show, Gergay Oroge, says, things that were not on my bingo card. One, Anthropic marked as a US supply chain risk by the government, same as Huawei. Again, that hasn't actually happened, but it's predicted. Two, cloud going viral across non-tech people as a result, number one on the App Store. Three, mass chat GPT cancellations, thanks to OpenAI becoming the government AI supplier.

2:35:02Dillon Rolnick:And PrimeEgen says, then they will release 5.4. Everyone will say AGI and forget all about the last week and this week. That's why it's so fun to cover AI, because everything changes every two weeks. It's not the something big is happening. Like I was reflecting, I was someone was asking me about that something big is happening essay. And I was like, well, actually they predicted that in March it would be like April of 2020 in regard to COVID. And the fact that we're just having this normal conversation and walking down the street normally and things are pretty much the same, sort of negates that, but it's still worth paying attention to.

2:35:43Dillon Rolnick:And I don't know, I'm excited for the next, model shift. There's some crazy...

2:35:48Matteo Franceschetti:Which I think is, our theory was, they said, GPT 5.4 sooner than you think. No, no, no.

2:35:56Dillon Rolnick:It's GPT 5.xx extreme. Citrini chimed in and said, the next AI model will have extreme reasoning. Probably trying to jump to the top of the meter chart, I would imagine. That's sort of the benchmark that a lot of people are tracking. So you got to do a lot of reasoning tokens, a lot of high, long context reasoning. But whatever they call it, it'll be fun to test drive and see what happens.

2:36:22Matteo Franceschetti:There is a new company selling a device called Spectre, the first smart device to stop unwanted audio recordings.

2:36:30Dillon Rolnick:This was so crazy, yeah.

2:36:31Matteo Franceschetti:And this post went super viral. People are pretty excited about it. someone else was highlighting that there's hundreds of products on Alibaba and spy shops that sell these sort of portable audio jammers.

2:36:46Dillon Rolnick:There's also a company that sells to doctors that's for transcription. Remember we talked about it and the company's printing like a hundred million dollars and they just like have a very niche sort of like boring use case.

2:36:55Matteo Franceschetti:That's for recording. This is for jamming. Oh this is for jamming. This is for TJ people like TJ Parker who don't want to be recorded.

2:37:02Dillon Rolnick:Because there's there's both of these Anthony Pompliano said We have to live in a simulation of both these companies launch on the same day. And it's one company launching a note taker for real world meetings. It records everything around you. And then on the same day, they launched a different company launched Spectre One, the first smart device to stop unwanted audio recording. So the battle for the sound waves is well, well underway. way. But yes, so there's other posts that you said is sort of putting Spectre 1 in the truth zone on the first smart device to stop unwanted audio recordings, talking about the idea that there are mic jammers elsewhere, and these are available if you want them.

2:37:49Dillon Rolnick:But they certainly haven't been marketed, and a lot of people were unaware of them until this product launched. So, you know, Is distribution important here? Is distribution always important? Yes, probably. And so good luck to them. Good luck to the ongoing cold war between the recording and the jamming. It's going to be just a crazy, crazy battle. Let me tell you about Plaid. Plaid powers the apps you use to spend, save, borrow, and invest, securely connecting bank accounts to move money, fight fraud, and improve lending now with AI.

2:38:24Matteo Franceschetti:Accenture is buying DownDetector. I saw this. $1.2 billion. Why? There must be somebody to this business.

2:38:31Dillon Rolnick:They said as part of a deal, a$1.2 billion deal. So is it possible that Down Detector is owned by another company and Down Detector just sort of came along for the ride? Because I can't imagine that Down Detector is worth$1.2 billion. I mean, respect to everyone who runs that site, it's certainly valuable. But that seems like a lot for what they do. But I don't know. I mean, I could imagine this having business value if you've implemented a system and you want to make sure that the system that you implemented as Accenture stays up. Having the infrastructure to track uptime is valuable. A lot of companies are struggling with uptime.

2:39:10Dillon Rolnick:Doug O 'Loughlin was talking about GitHub's uptime. There's been a lot of people posting about Claude having patchy response times because of the surge in demand. And this is clearly something businesses care about because every dollar, every minute of uptime is dollars.

2:39:26Matteo Franceschetti:Maybe in the age of vibe coding, every product is going to be going down so much. Accenture's helping companies, you know, unlock AI. And they're like, wow, this stuff, this is going to be pretty bad. Yeah. Bull market for down detector. uh julian weiser uh interviewed uh a founder named ben sarah he's building a company called pulsia uh they've gotten to one and a half million of ar in two weeks with zero human teammates uh very very uh cool progress but zarim says why is the company named ai slop backwards

2:40:06Dillon Rolnick:is that intentional is that a nod or do they just get no i think it's a maybe a happy

2:40:13Matteo Franceschetti:accident that's so funny really good uh donald trump uh is at a round table i believe on energy on the on on uh the the data center build out and donald trump according to mike isaac wistfully at the tech round table speaking about open ai's brad light cap brad so young look at how young

2:40:41Dillon Rolnick:that's so funny brad light cap does uh look very youthful he has aged very very very youthful business also an absolute dog who's been on a generational run and done uh fantastic things throughout his career uh so congrats to him but this is so so funny what a weird that's a total like record scratch freeze frame yep that's me here's how i wound up at a white house round table with the president of the United States saying, so young, look at how young. Thanks, Mike, for editing that up.

2:41:17Matteo Franceschetti:And a good place to close out the show, Brian Johnson walked at a fashion show at Paris Fashion Week. A Capital says, from Immortal Unk to Runway Unk.

2:41:29Dillon Rolnick:Let's give it up for Brian Johnson walking the runway. That's a bucket list item. I don't know if it's on my bucket list, but I'm certainly glad that he had the opportunity. Congrats, too.

2:41:39Matteo Franceschetti:The only thing is it kind of looks like a futuristic funeral. It does. Like a sci-fi. Like, it doesn't have...

2:41:47Dillon Rolnick:The clothes look fairly comfortable, though. I like the functionality. But he is looking absolutely jacked. He's looking jacked. That is true. He is looking very strong. This is some mass surveillance going on. They're surveilling his mass.

2:42:01Matteo Franceschetti:It really is mass surveillance.

2:42:02Dillon Rolnick:This is mass surveillance. And whether you like it or not, they're going to be taking a video of you and you're looking massive and putting it on the internet and going viral. Well, thank you for watching the show today. We will see you tomorrow at 11 a.m. Sharp Pacific time. Leave us five stars on Apple Podcasts and Spotify. Sign up for our newsletter at tbpi.com. Give us a follow on LinkedIn. I have been doing a magic trick recently where I will post on LinkedIn while I'm live. It's a certain hand signal. You have to figure out how it's happening. How can John possibly be live on the internet on this stream and simultaneously posting content, great content, some of the best content on LinkedIn at the same time?

2:42:42Dillon Rolnick:It's a mystery. No one understands how I can do it.

2:42:45Matteo Franceschetti:Right now over at LinkedIn, it's...

2:42:48Dillon Rolnick:No, no, no, that's not true. I got 100 ,000 impressions. It's working. It's starting to work. But we figured it out that on LinkedIn, we are going to be focused more on behind the scenes storytelling about how we think about media, how we think about the business of TVPN. That has been what's resonating. And so we're going to be focused a little bit more on that as opposed to some of the tech analysis that we do on X and here on the show. Some of the interview clips. The interview clips just aren't really flying there like they do on YouTube and X and Instagram. So we're starting to find our audience there.

2:43:22Dillon Rolnick:Give us a follow on LinkedIn. We'd love to hang out with you there. Anyway.

2:43:25Matteo Franceschetti:And we hope you have a wonderful evening. We'll be back tomorrow for a big show. Goodbye.