Meta’s Muse launch, Josh is wrong on Netflix, internals weaken, 10-year bonds at 5%, Buffett steps away

The Compound and Friends

The episode explores the massive market impact of Meta's Muse launch and its implications for the broader AI trade. The hosts analyze the sh

Key takeaways

• Meta's Muse launch added approximately $450 billion in market cap in just three weeks.
• The AI trade is expanding beyond GPUs to include 'agentic' workflows that benefit CPU-heavy stocks like AMD, Arm, and Qualcomm.

Main topics

• Meta's Muse and the AI trade
• Semiconductor rally (AMD, Qualcomm, Arm)

Notable quotes

Meta finally launched something that people want in the AI arena, and it has literally ignited an incredible rally.

Conclusion

Investors should focus on identifying momentum in winners rather than

Transcript preview

Speaker 2 (0:15) Yup, we're back. I haven't done this show with you in, I think it's three weeks, right? Because last week we were in Future Proof. The week before, I was with a back injury, and you had Sharkade and Matt and Sean. And then, so the week before that, so it's August. Speaker 1 (0:36) Nothing's happened since then, really. Speaker 2 (0:39) Nothing at all. Everything has happened. All right. Well, I miss the show. This is like one of the highlights of my week every week. So, guys, we're pre-taping. It's Tuesday morning. I am doing a thing at the University of Miami later today, which prevents me from being live. But those of you in the live chat, thank you for coming for the premiere. And we appreciate seeing you guys do that every week. But we're live. How Speaker 1 (1:07) are Speaker 2 (1:07) you feeling? We're live. You and I are live together. It's good. Literally, Speaker 1 (1:10) we're doing it. I feel great. Well, actually, I don't feel great. I did not like seeing Jackie's knee go sideways last night. That sucked. But it sounds like we avoided catastrophe. And you know what? Who cares? The Knicks are champions. I'm still running high. I don't care. Speaker 2 (1:22) I was going to say I'm already pivoting to basketball. Yeah, Speaker 1 (1:25) it's okay. Speaker 2 (1:26) I sent my Giants friends a text last night. I might not be watching for the next two or three weeks, so don't even bother. Yeah, Jameis was a Speaker 1 (1:34) tough watch. Yeah, I don't think I can watch that again. Speaker 2 (1:37) All right. This is What Are Your Thoughts? For those of you tuning in for the first time, this is America's favorite investing and markets live stream. We try to get to the biggest stories or the ones that we feel we have the most interest and take on. We cover stocks, bonds, the economy, venture capital, technology, anything that touches. the investment portfolio, we're going to chime in on. That's what we do. We appreciate all of you guys for listening wherever you are and watching. We have a sponsor for the show. Michael, tell us who's sponsoring. What are your thoughts today? Speaker 1 (2:12) This podcast is brought to you by DBMF, the world's largest managed futures ETF. Feel like the world's changing fast? Imagine a strategy built to move when everyone else is standing still. Going long or short across global markets as the trends shift, DBMF is made to move differently, a single. low-cost ETF offering genuine diversification, even when stocks and bonds move together. Discover why DBMF's liquid uncorrelated managed future strategy could be what your alts allocation is missing at www.dbmf.com slash W-A-Y-T. DBMF may move differently. The IMGP DBI managed future strategy ETF's investment objectives, risk charges and expenses must be considered carefully before investing. The statutory and summary prospectuses contain this and other important information about this investment company. It may be obtained by visiting www.imgp.com. The fund is distributed by Alps Distributors, Inc. DBMF is the world's largest managed futures ETF as of July 31st, 2026, with $4.16 billion in assets under management. Speaker 2 (3:13) What kind of animal is that? That's a sloth. That's a sloth. Okay. You know, I was a big... It moves different. Speaker 1 (3:21) I was a big animal child. I enjoyed spending time in the library as a six-year-old. Couldn't get enough books about animals. You Speaker 2 (3:28) were big on Speaker 1 (3:29) animals? Yes. Speaker 2 (3:30) All right. Well, let me show you this wildebeest. Meta finally launched something that people want in the AI arena, and it has literally ignited an incredible rally for lots and lots of tech stocks. Speaker 2 (3:48) I don't know, man. I think this is pretty cool. And there are some big takeaways here. I'm going to share mine and I want to hear yours. But let's just set this up for people. On September 8th, which I think was a Friday, Meta launched Muse and immediately incorporated it into WhatsApp, put it all over Instagram and all over Facebook and got it in front of a lot of people. Speaker 2 (4:14) They're advertising it on NFL games, which is not cheap. Like they're going all in on this thing. And I think it could be the most important company event since they came out with Reels, where they said, nope, TikTok is not going to win. We're going to win. Here's our product. So the stock was $578 a share to open up September. It's now like $740, $750. So that is $163 per share in all three weeks. It's up 28 % month to date. This is basically meta adding between $400 and $450 billion in market cap in three weeks. This is the single biggest or the biggest single month value creation in the company's history. And just to put that number in perspective, adding $450 billion is like adding a Costco, which is 430. Procter & Gamble, 400 billion. J &J, 400. Home Depot, Netflix, 375, 380. So this is an enormous swing in the fortunes of meta. I want to show you how fast the story has changed. Put up this 18-month chart. Speaker 2 (5:40) So basically, ever since this time last year when the share price peaked, it's been a series of lower lows and lower highs. Meta had been locked in this substantial downtrend, and it got caught up in this, you know, are they spending too much in CapEx? There's no ROI. Is this the metaverse part two? And I'm not mocking those comments because they were well-founded. But the... The bigger take, oh, here's Meta versus Mag7. Look how fast this changed. I think with Tesla, they were the two worst Mag7 stocks this summer. And now it's outperforming. And that happened like almost overnight. So before we even get into what Muse is, what are your thoughts on like what investors can take away from the rapidity with which this entire story has changed? Speaker 1 (6:37) To me, the biggest takeaway for a $1.4 trillion stock knocking on the door of $2 trillion in a three-week period shows how wide the range of outcomes for this entire AI trade build out. Because we were discussing this over the summer. The stock is a piece of shit. Technically, it looks completely busted. It's in no man's land. Speaker 1 (7:06) after the conference call, they're not making anybody feel any better about the direction of the company. It just seems like a lot of spend. And it seems like the declining PE is warranted. So you had the setup of really investors throwing in the towel. I think Meta was trading at 16 or 17 times forward earnings, which sounds insane. But guess what? I had that pitch. So did everybody. And I didn't want it at 16 times earnings because the fears, to your point, seemed completely reasonable. And I wasn't enticed by evaluation because of the overhang. And here we have this game-changing potentially app to hit the App Store. And I think it's been number one since it popped on or in the top five. And it looks like a game changer. And I don't know how many tens of billions of dollars of new potential revenue this is, but nobody does. And that's why the stock re-rates like that. Well, that's a really Speaker 2 (7:57) key point. I don't know what earnings estimates are doing in the wake of this. I know some analysts are taking their numbers up because they know how good Meta is at monetizing products. Like that might be the thing that it is the best in the world at. It's not just that they launch products. It's that they're really good at making money. And so I'm seeing some earnings estimates go up, but the multiples are going up. I saw one analyst this week. go from 20 to 25 times earnings in his forecast. And that may not sound like a lot, but we're basically saying this one product is a 20 % bump on the multiple of earnings that people are willing to pay. And some of that is narrative, but some of that is like, oh, here's a new leg to the business that nobody even thought was coming. or nobody was bullish on, and now all of a sudden everything has to change. To me, that's a big takeaway. The other one for me is when are we, all of us, the colloquial we, when are we going to stop betting against Mark Zuckerberg? Like what do we ever learn our lesson? How many times has the consensus said? Well, he's really done it this time Now he's in trouble how many like how many can you think of offhand the metaverse pivot? Being like the most obvious gigantic example, but I remember when the company was coming public. Yeah, they have no mobile app It's a it's a desktop It's a desktop website in a mobile world. They're totally fucked. They can't reinvent blah, blah, blah, blah, blah. Oh, now they're paying a billion dollars for Instagram. Speaker 1 (9:42) I love this. Speaker 2 (9:43) I Speaker 1 (9:43) love. When are we going to stop betting against this? Never, never. I love the narrative because you're right. Zuckerberg has outperformed literally every step of the way from the time he met Peter Thiel until today. But we are never going to stop betting against him because no company. no company gets the benefit of the doubt because the stock market is a cold place. And it doesn't matter what you've done. It matters what you're doing and what investors expect you to do. And I sort of reject the premise of when we're going to stop betting against them because six months ago, they were in a shitty place. And had we known that news was coming out, then we would have been excited, but we didn't because we couldn't have. And they looked like they were going left. Now they're going right and everybody's back in the boat. And I love it. This is what makes the stock market so much fun. Another key takeaway for me is I would not be psyched to be an investor in Instinct, which raised $350 million at a $2.5 billion valuation. Now that's a private company. It's obviously not our game, but this shit is hard. It's changing really fast. So the check that was just written back in August for that company, those investors must be sweating. No? Speaker 2 (10:57) Yeah. And OpenAI has a personal assistant that's going to drop in the fourth quarter. And that'll instantly have a billion people testing it out. And Speaker 1 (11:05) I feel like this particular product has no moat because the switching costs are zero. It's all plugged into your tech. Oh, it knows me. Who cares? It knows your Gmail. That's very easy. That's a commodity. Speaker 2 (11:16) Yeah. And this is a competitor with extremely deep pockets that is not afraid to keep investing until they have the best product on the market and not afraid to spend on marketing. and advertising off their own platform. Like watching Facebook buy ads during a Monday night football game tells you how important this launch is to them. This is not them just dropping banner ads inside of Facebook and hoping for the best. Like they're spending big money to Speaker 1 (11:48) tell this story. Another very interesting development is that Amazon cut off Muse. Speaker 2 (11:55) Well, it's bundled with Spotify. So they have a Spotify partnership. And I don't know, maybe there was talk of an Amazon partnership and it couldn't come to terms. I don't really know the political side of it, but something tells me like people might look at that as a negative. And then all of a sudden they'll make a deal with Amazon and the stock could go up another 10%. But Speaker 1 (12:17) just another takeaway from this announcement is there are going to be knife fights everywhere. Speaker 2 (12:22) Oh, yeah. Because it's... Because it's the land grab. Yeah. So when you say land grab, people sometimes don't even know what that means. There's a great Tom Cruise, Nicole Kidman movie called Far and Away. Saw it in theaters. So they're both Irish immigrants. Wasn't he a boxer? He was boxing? He was like a bare knuckle, like, you know, what they used to do. Put up your dukes. Put up your dukes. Anyway, they take part in a land grab. Sorry to ruin the movie. The climax of the movie is... They're leaving Boston and they're going west like a lot of the Scots-Irish immigrants did. Bro, how many times have Speaker 1 (12:59) you seen this Speaker 2 (12:59) movie? Not a lot. But they have a stake and they're going to put a stake in the ground. That's literally stake your claim. It's a race. They have to get as far away as possible from everyone else and jam a stake in the ground to call – this is our future farm or ranch or whatever it is. That's the – The stage that we're at now, someone is going to own this personal agentic AI assistant and Mark wants to win. He ain't going to win the LLM wars, right? He's not going to necessarily lose if this works, but like he's not going to do the chat GPT war. This is something else. This is something that you talk to and it carries out tasks for you. And it's not just to ask questions. It's sick. Have you tried it? Speaker 1 (13:48) So I haven't used it yet. I haven't used it yet, but I'm seeing the use cases and I've been using Instinct. So I get it. It's the same thing. And it is incredibly powerful. Speaker 2 (13:55) All right. Muse became the number one free app on the Apple iOS app store and the Google Play store as of yesterday. It is available for people 18 years and older. So it doesn't like forget all the games downloads. For adults, it's number one. It was downloaded 902,000 times. in the first six days. How many times? 902,000 times in the first six days. All right, that's going to be like 10 million by next week. Well, I'm saying like it's an instant hit. Meta launched the Meta AI app and only did 773 downloads for that. And the capabilities were not the same as this. It was very different. Here's Evercore ISI. Meta has a hit on its hands with Muse. It amounts to a very tangible sign of successful product innovation. It's evidence that Meta's $200 billion in AI investments have not been in vain. This is Mahaney and his team. The user trend suggests, quote, a potentially very dramatic new driver of usage and engagement for Meta that could bring, quote, substantial new monetization opportunities. advertising subscriptions transaction revenue share yeah so Speaker 1 (15:15) it's it's a really Speaker 2 (15:16) big Speaker 1 (15:17) deal Speaker 2 (15:17) now this is important as quickly as the narrative became this it could fade i don't think it will i wouldn't bet that it will but we have seen this before also Speaker 1 (15:28) where Speaker 2 (15:29) an app takes you know catches fire and then everyone's over it we've seen this we've Speaker 1 (15:34) seen this a lot right Speaker 2 (15:36) um It did ignite a very interesting rally in CPU stocks. Did you see any of this stuff? I did. Give me my chart. I don't know. What's the connection? I'll explain in a second. The blue line is advanced micro devices, AMD, which just went absolutely wild. We'll talk about more in a second. Purple is Arm Holdings. Green is Qualcomm. And orange is, you can't really see the magnitude of the rally, but it's Speaker 1 (16:05) Intel. Speaker 2 (16:05) Intel Speaker 1 (16:05) was up Speaker 2 (16:05) 15%, I think, yesterday. Yeah, it Speaker 1 (16:07) was Speaker 2 (16:07) a big deal. It doesn't look like a big deal on the scale of this chart, but it was. Here's what's going on. Muse has a very unusual technical architecture that reshuffles the way people are thinking about the AI chip trade. Most AI apps obviously lean heavily on GPUs, which is what... NVIDIA is dominant in. Muse is keeping a secure virtual machine for every user. Okay? And that machine is handling browser sessions, what's called tool calls, where you prompt something and it has to call on a specific tool, sandboxing, all types of things that CPUs carry out in an AI workload. Speaker 2 (17:00) I think what's notable is that I mentioned Arm Holdings and Qualcomm, neither of which are making traditional CPUs, but they are very heavily betting on and invested in inference on device, which means they make specific chips for cell phones, for mobile phones, where people can do that inference work directly on the device rather than routing it to the AI data center. That's why those stocks,