Semiconductor analyst 'Bubble Boy' says advanced packaging is the new Moore's Law — and the market hasn't priced it in

Speaker 2: We have some important questions.

Speaker 15: Another one.

Speaker 1: He's coming. He has been on an absolute terror.

Speaker 13: There he is.

Speaker 2: Bubble boy. I like this golden retriever. Hello.

Speaker 13: Hello. Hello.

Speaker 2: Hello. Okay. So you said, yo TBPN put me on the show to talk about semis. I will make your listeners rich. Go. Make our listeners rich. Tell us about semis.

Speaker 1: Now let's start like, why why do you still have a normal job at this point? Let's start there.

Speaker 15: Yeah. So someone actually asked me this recently, and I think the answer I got was a lot of my ideas come from working on the same problems these engineers are actually trying to solve. So I really don't think I could have actually called Intel or called a lot of these other companies like Sandisk if I didn't have that experience and know how and expertise. So I actually think having a job is a bit of alpha, believe it or not. I think a lot of the engineers who are actually just making a killing in this market, I mean, these are just real experts in their field. And we look at the market every day and we just see, you know, big divergences. I don't think the finance bros really understand what advanced packaging is or what yields really mean. They understand it maybe from a financial aspect, but not from a competitive standpoint.

Speaker 1: Re rewind to a year ago, where did you think we were in the overall AI cycle? And where do you think we are today? Yeah.

Speaker 15: Probably the start. I I I think

Speaker 2: It's good.

Speaker 1: At the beginning of

Speaker 15: the year. At the beginning of the year Yeah. Every hyper scale basically said that they're gonna spend 680,000,000,000.

Speaker 2: Yep.

Speaker 15: So, I mean, we're just the start. I think we have at least three years of CapEx growth going crazy from my estimates. And from there, I mean, tokens are just looking like the new oil or electricity. Think at this point, I actually can't live my life without a token. So, yeah, I I think we're just at the start. I think right now is really when things are gonna be going crazy. Okay. If you actually look at the data center build out, nothing's actually been completed yet. We're just at the beginning.

Speaker 1: How do you how do you think everyone having access to LLMs, you know, these incredible research partners is changing the markets? Like, does that do do do things get priced in faster when everyone can can, is maybe asking the same questions?

Speaker 15: Absolutely. And that's a good question. I just need to shout out GPT 5.5, my preferred model for, you know, stock research. I think that I know for sure that the number one consumer of these tokens is these hedge funds and financial institutions. They are absolutely running through a bunch of research, and it allows you to get up to speed quickly. If you really think about it, like, five years ago, if I wanted to make a bet on, like, chemical epoxy for semis, that would have took in six months of my life and been a bit ridiculous.

Speaker 2: And then

Speaker 1: and then and then so so share share your post yesterday which went viral, turned into news. You were just you were basically just showing that you could just make up stuff on the timeline and and I think you moved I think you moved the market. Give us the backstory.

Speaker 2: Yeah. What happened?

Speaker 15: So I was actually just researching these chemical suppliers, and I just realized, you know, you could really you could really talk to an LLM and get it to say whatever you wanted to say. I So I I just picked a random company, honestly, in Japan, and I just said, hey. Try to really pitch this as an AI bottleneck. Sadly, it's not a bottleneck. Right? But I I think they're doing well either way. Yeah. They've been I mean, we're in a good place. You know? Data centers will be built out. We don't need to wait for SumiTimo Bakelite to get their epoxy resins to scale. But yeah, so I I guess that's a harm. People need to definitely do their own research. I've seen a lot of people just piling into things they don't understand. So very interesting place to be in.

Speaker 2: What's your intel thesis right now?

Speaker 15: Great question. I think up to this point, intel was a talk intel's thesis was really a fab turnaround. It was about eighteen a and '14 a, which is coming. I think really now, I've completely changed my logic. It's about advanced packaging. Advanced packaging to me is the new Moore's Law. You've kind of already seen this with bilirubin, where they can't actually do four dyes on one package. They actually had to change that last second. If you look out to Feynman, I mean, these packages the the packaging of the silicon is a huge bottleneck. It already was a bottleneck. In twenty in twenty twenty five, NVIDIA was 60% of TSMC's advanced packaging. So everyone else was out there to fend for themselves. So any kind of advanced packaging that comes to market, that allows people to scale, compute way more than I think just shrinking a node. I mean, shrinking a node's great, but the way NVIDIA's timeline, and I think all the hyperscalers as well, they want more HBM. They want more radical sized dies, and they all want it in one package.

Speaker 9: Mhmm.

Speaker 15: So that's really where things are headed towards, and I think the market's not pricing that in. I think this is a real sea change to the semi industry. This really is just happening over the past year or so.

Speaker 2: What's the story with the dog?

Speaker 15: Oh, well, this is just kind of like how I look on a day to day basis when I'm sitting at my desk, sipping on a diet. Well, Pepsi in this case, but I just like to, you know That's for. Spread joy. Yeah. No. I think this is kind of like how you look when Intel beats on earnings and goes up 30% in a day, and you've been proven right after a year of being told you were wrong. So Yeah. Yeah.

Speaker 1: That's good. What's any any any non financial advice predictions for the this tech earnings cycle? What are you expecting tomorrow?

Speaker 15: Well, I think we have Sandisk reporting soon. I think that's kind of been telegraphed that Sandisk can just increase the price of Flash until the customers cry. So I think, yeah, they they will be fine. I think the other story is people think there's a CPU shortage.

Speaker 2: Yeah.

Speaker 15: I think that this is gonna be we're gonna get a yes or no definitely in this earning season.

Speaker 2: And

Speaker 15: people actually just price up CPUs 25, 30%. And will people take it?

Speaker 1: When when you when you'd post a picture of your a screenshot of your returns on X, do you ever get worried that that it's a sign that you should at least go to cash somewhat or does does it just

Speaker 15: not. That's never not. I've never thought of going to cash. I I actually took some money out last year for taxes, and it was the biggest mistake of my life because that money, who knows where how much it would be by now. I see this as we have a three year run at least, and now it's time to really pull the trigger. If anything, I'm trying to borrow money at this point to go bigger.

Speaker 1: And that's why they call him Bubble Boy.

Speaker 2: That's why they call him Bubble Boy.

Speaker 1: You used to pray for a bubble like this.

Speaker 15: Yeah. I I definitely did. Yeah. And it's come to fruition so

Speaker 1: What what can people what can people expect from you? You you say that having a job is alpha. I'm sure you're getting people are trying to pitch you on new job opportunities. But what what can people expect from you this year besides making a lot of calls Mhmm.

Speaker 15: All the

Speaker 1: time and to date often being quite accurate?

Speaker 15: I think you'll I think I'll give a hint. You'll see me in the flash business somewhere, somehow, maybe at a pretty established flash company in supply chain. So I think flash is one of these technologies that I think scales a lot higher than the market expects. And I always have this joke. Say, like, if you want to, like, change the world, you can cure cancer, or you can come up with new kinds of memory. So I can't work on the first one, so I'm gonna look at the number two for now.

Speaker 1: And would this be an early stage

Speaker 2: get exposure. Yeah.

Speaker 15: Potentially or it could be really partnering with a very very large company.

Speaker 2: Cool.

Speaker 15: So there's some advantages to both.

Speaker 2: Well, good luck wherever the future takes you

Speaker 1: and This was fun. I I've I've deeply I've deeply enjoyed your your posting. Always been fun. And and yeah. You're clearly having a lot of fun and it's fun to watch. Great to meet you.

Speaker 15: Thank you guys both and big fan of the show. So thanks for having me on.

Speaker 2: Thanks for having It's

Speaker 1: been an honor.

Speaker 2: We'll talk you soon.

Speaker 1: Cheers Bubble Boy. We'll talk you. Cheers.

Speaker 15: Bye.

Speaker 2: Should close with an example of why memory is so expensive. Star Wars meets Pawn Stars. Have you seen this video? Have you watched this, Jordy?

Speaker 1: I did. It sounds crazy.

Speaker 2: I saw the first second. I haven't watched the full thing. But I think it's

Speaker 1: probably It's

Speaker 12: pretty incredible.

Speaker 9: I have items to pawn.

Speaker 12: Alright. Let's see what you got. Okay. So these are lightsabers.

Speaker 9: Correct.

Speaker 12: So where'd you get them?

Speaker 9: Estate sale.

Speaker 12: Estate sale.

Speaker 1: It's so funny that you can clock it by the audio or the video.

Speaker 2: It's very weird. Collector for many years.

Speaker 9: It is a great passion passion of mine. Each lightsaber has a story, and he's telling the people. There is a certain

Speaker 2: I guess the character is CGI to begin with. So it's it's it's hard to clock. But even he's probably fully AI

Speaker 9: is a lung condition.

Speaker 2: Video or maybe it's set it together. I don't know. But, yeah, you're a 100% right. It's the audio that's off.

Speaker 9: Sell my collection.

Speaker 12: That's rough, man. How much are you looking

Speaker 1: to get for him?

Speaker 9: A 100,000 prints.

Speaker 12: Look. I'll give you $50 for him.

Speaker 2: You fool. It's interesting that the like, why that's popular is because it's this mashup between two intellectual pieces of property, and they're not being compensated for that. And so there's, like, thing that can only exist in the piracy world, basically. Like, if someone if you had recreated this without leveraging Star Wars IP or Pawnstar's IP, doesn't go viral. Right? And so the the business case for AI video is still a little bit more narrow and I think it will be tucked in the tool section. I was I was, yeah, looking a lot at like the the Aflac projects that he sold to Netflix and some of the background replacements, some of the VFX stuff. Like, it feels like the way even though this is like this oh, it's like one shotting entertainment, it feels like the next the next moment of, like, AI in Hollywood is very much, like, tool driven, you know, leveraging things like, you know, green screen removal and object replacement, VFX workflows, and just sort of speeding up remote or like rote tasks. But people will continue to have fun with these mashups. I'm sure we'll see many more of them on the timeline. Anyway.

Speaker 1: The other news.

Speaker 2: What

Speaker 1: else? Mike Isaac and his team over at the New York Times are in the courtroom live blogging.

Speaker 2: They're blogging.

Speaker 1: They went. They're blogging. Yeah. They're blogging. It is There's some crazy moments already.

Speaker 2: The judge is like doing bits.

Speaker 1: Put some of them together. The judge is doing bits. The judge is doing bits. Apparently, Elon's lawyer's microphone turned off four times in the course of his opening statement.

Speaker 2: Maybe the screen turned off at one point. I saw some different on that.

Speaker 1: Apparently on the There were technical issues. Fifth or so time the judge says, what can I tell you? We are funded by the federal government. So

Speaker 2: That's wild.

Speaker 1: She's running she's running bits. But I highly recommend going over and Yeah.

Speaker 2: Mike Isaac and some fantastic coverage.

Speaker 1: Posts are still rolling in. Yeah. It's a lot to go through because it's it's it's hours and hours and hours of content but we'll try to pull out some of the highlights and Yeah. And go through it on tomorrow's show.

Speaker 2: The other thing you should listen to is Patrick O'Shaughnessy had Paul Tudor Jones, one of the greatest macro traders of all time on invest like the best. It's an hour and eleven minutes and you should go listen to it. So go take a listen. It already has 5,000 likes, 1,700,000 views, not nearly enough downloads. So go add it to your podcast player right now. And enjoy that for the rest of the day. Enjoy the rest of your day and we'll see you tomorrow, 11AM. Lead us five stars now. Podcast, Spotify, sign up for our newsletter tbpn.com.

Speaker 1: Can't wait to see you tomorrow. Have the best afternoon of your life. We love you.

Speaker 2: Goodbye. Flashback.

Speaker 16: You're going flashback.

Speaker 2: Out. Goodbye.