TBPN - Diet TBPN: October 31, 2025

Episode Date: November 1, 2025

Our favorite moments from today's show, in under 30 minutes. TBPN.com is made possible by: Ramp - https://ramp.comFigma - https://figma.comVanta - https://vanta.comLinear - https://linear.appEight Sleep - https://eightsleep.com/tbpnWander - https://wander.com/tbpnPublic - https://public.comAdQuick - https://adquick.comBezel - https://getbezel.comNumeral - https://www.numeralhq.comPolymarket - https://polymarket.comAttio - https://attio.com/tbpnFin - https://fin.ai/tbpnGraphite - https://graphite.devRestream - https://restream.ioProfound - https://tryprofound.comJulius AI - https://julius.aiturbopuffer - https://turbopuffer.comfal - https://fal.aiPrivy - https://www.privy.ioCognition - https://cognition.aiGemini - https://gemini.google.comFollow TBPN:https://TBPN.comhttps://x.com/tbpnhttps://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231https://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://www.youtube.com/tbpnLive

Mag 7 earnings recapping what's happening in the public markets, the trillion dollar tech companies that have been powering and holding up the entire global economy. Will they continue to hold up the global economy? So far, the answer is yes. I'm glad the chat is enjoying the costumes. Jordi, who are you today? You got to be living under a data center to not know who I'm dressed up as today, none other than Mark and Drieson.

Spitting image. Spitting image. How am I doing? I think you look great. Of course, I am Elia. Siskiver. If you are watching the recording of this at some point in the future, I recommend whenever I'm talking, put it on 3X speed to get the full mark experience. If you're, if you're not taking Halloween this seriously, what's going on? You've got to be taking it seriously. Tyler, are you

taking Halloween seriously? I'm taking Halloween extremely seriously. Well, who are you? Sam Alman at 2008 W.W.D.C. Trump cut China tariffs after he met up with Xi Jinping. The U.S. agreed to lower a fentanyl-related tariff on Chinese goods to 10% from 20% after China promised to crack down on chemicals used to make the often-deaddle-bred precursors. Precursors. Beijing also promised to ease some of the controls it imposed on exports of processed rare earth minerals for one year. China dominates the production of the minerals, which are needed to make everything from smartphones to submarines. Wow. The bigger news in tech world, of course, is earnings. Amazon posted a big jump in profit and revenue. Net profits searched 39%

and revenue rose for the third quarter. The last earnings, they were sort of left out of the AI narrative, all the other hyperscalers in their cloud businesses, telling great stories about acceleration and growth, and Andy Jassy was a little bit on the defensive. Well, now he's back on the offensive. They beat on the top line. They beat on the bottom line.

They also beat on AWS revenue. They were, you know, proud member of the lag seven. Yeah. You look at Alphabet. I mean, not the lag seven, but they were, a lagging part of the Mag 7. It's a pleasure to introduce you to a new meme. Alphabet up 247% over the last five years.

Amazon is still only up 62% even after the pop this week. So the only thing better than Halloween is Halloween AI edition. In case you missed it, tagged Ilya and Mark. And of course, no AI here. No AI. Did this one the old fashion way. Microsoft. Microsoft.

What's your Microsoft tape? Topline beat. Earnings beat. Stock traded down three to four percent. After hours, it's recovered quite a bit since then. Satya, in my view, has already gotten a lot more credit for the company's opportunities in AI than Amazon and even Google. People have known about Google strengths, but the big questions around search have held back performance to date. There are so many companies that are trying to build Excel plus an agent on top of it, which is exactly what Microsoft is also doing with Excel with co-pilot.

And they're model agnostic. They're going to be able to bring in any kind of intelligence that they want. And so it's also fascinating because they have access to your opening eyes IP. Slack was very much the same thing where Slack was this product that was so much better than the alternatives. Like if you, I don't know if you ever used like workplace chat pre Slack, but it was. I was basically born on Slack. I never knew John was so ripped. There's been some other comments.

Yeah. Yeah. Baby, let's go. Ilya. HGI has been achieved. Let's go. Let's talk about meta. Yes.

Top line beat, earnings miss. Massive, like a $12 billion. People are not talking enough about what happened. It is crazy that you could be operating such a big company and just be like, oops, $10 billion. I mean, think of how they've been spending. This is like, you know, I'm sure meta and the executive team, it's like after a weekend vendor checking like your credit card statement.

Okay, so apparently it was a one-time tax charge related to the Big Beautiful Bill Act. What? Is that why the stock traded down 10%? Is it specifically because of this one-time tax charge? I think it could be a little bit of vibes in the mix. The timeline loves the core meta-business, but doesn't have a lot of faith in Zuck. People thought Bezos, who had a similar level of control over Amazon,

was crazy to spend like he did building AWS, and he was ultimately completely vindicated. People thought Mark was crazy going all in on the Metaverse and VR. So far that part of the business has not shown any signs of being ROI positive. Reels is now a $50 billion revenue run rate business. Let's go. So Tesla beat earnings missed, a beat on revenue missed on earnings. The stock reaction was down about 4% after hours.

If Elon goes straight shot to the full self-driving equivalent for humanoid robots, basically no teleoperation, the project should be years behind what's going on with 1X, I would imagine. Just like the Waymo dynamic, Tesla's going full moon shot, full send. No one debates the fact that like Tesla's Robotaxy fleet is a year or two behind Waymo's right now. Now, the Tesla bull hope is that you come from behind and start compounding because you have just structurally better costs, way more manufacturing capacity, more of a flywheel on the actual data and the training. Elon Musk on data centers in orbit, SpaceX will be doing this. He's going to do data centers in space. It's easy to be bullish on space right now. But if you're investing in different categories, you have to imagine that if there are adjacent opportunities besides Star-Lar,

link for SpaceX to do. They will enter them if they are large enough categories. And it's also just fascinating to look at the space industry versus the automotive industry. The automotive industry is the most brutally competitive, challenging business where Elon is having to compete with every car manufacturer in the world. In space, it's like, you know, you can imagine that being not a perfect monopoly. It's certainly having every possible advantage in order to dominate in that category when the time is right. Apple beat top line. beat on bottom line. China sales decline year over year.

If you've been paying any attention to Apple, you know that their China business is slowly going away. iPhone revenue they missed just by $300 million. $50 billion. John, your take, you said, sort of the inverse meta-core business, not at risk due to AI, but not investing aggressively. Consumers seem happy to just open an LM app.

could take. I said Apple's trading it 40 times PE with very low growth. This feels steep even if I don't think they're threatened by AI in the near term. We should close it out with Nvidia just to finish this little earnings wrap up. John, your take was on top of the world, but is that head that is the head that wears the crown heavy core AI factory build out going flawlessly but unclear why NVIDIA is pushing data centers in space, humanoid's quantum computing, mumbo jumbo. Maybe it's real. Maybe just forward of the same old AI story? I said I don't think Jensen would be just chugging beers on camera in Korea if he was having a rough one.

He is the most value creative founder CEO in human history. No one's created $5 trillion in any capacity. I'm sorry, but no company generating the cash meta is with the margins it has should ever be raising money. Of course, they announced another bond sale yesterday morning. Don't pay every AI scientists like their show hey, Otani. Yeah. When I hear that they're setting up a big data center and they're raising debt for that particular deal, I don't have like red flags flying off in my mind. AI is not a bubble. Enterprise AI is very strong. Nearly 150 enterprises each process one trillion tokens with Gemini models for a wide range of apps. And then they say, wait, that's less than a million bucks per year per enterprise or 150 million of annual revenue, i.e. 0.3% of GCP annual revenue.

What percentage of GCP annual revenue is AI, is Gen AI? Sometimes the hyperscalators don't know. Tyler, what you got? A trillion is just like not a lot at all. It's just not a lot of things. Open AI is doing like just on the API like something around like $250 trillion a month. If it's a trillion over the past like year, even if it's 150, I mean, that's like, that's pretty bad. This was crazy.

I wasn't sure if this was fake news or not. But on the Coinbase earnings call yesterday, Brian Armstrong. wrong. When there was a prediction market that people were betting on which words he would say during the earnings call and he waited until the end and he just read through and read off every single word that he had missed. Because I was tracking the prediction market about what Coinbase will say on their next earnings call and I just want to, you know, add here the words Bitcoin, Ethereum, blockchain, staking and Web 3.

We've had prediction markets created around different refuse to be done. And it is wildly distracting when I'm looking at the chat and people are like, say this word. Say that word again. It's crazy. Say it three more times. Here, Microsoft seemingly just revealed that Open AI lost $11.5 billion last quarter.

They have the least profitable business in the world? Who's losing more money than Open AI? These are not losses. These are investments. So like if I raise $10 million for my startup and then I'm like, okay, I'm going to buy of office building for $5 million. That's not negative 5 million earnings. That is negative 5 million cash flow.

I've lost 5 million cash, and I don't have liquid cash anymore. And so I would assume that a lot of Open AIs losses are actually investments in CAPEX one way or another. The first half of 2025, they're going to do 3.5 in losses. So this is basically double that. Palantir shoes former employees for stealing company secrets, The company alleges pair violated non-competes to create copycat AI firm.

Interesting. The two employees, they're accusing them as stealing confidential data to build a rival company called Percepta, backed by General Catalyst. I would not want to steal from Palantir. Bad idea. That seems like one of the worst companies to steal from. They will find you and they will kill you. Black Rock stung by loans to businesses accused of breathtaking fraud and entrepreneur tap private credit funds to finances, telecom businesses, AV in the chat. Large-scale fraud is actually a sign of good times.

I mean, that's true. Please, Mr. President, respectfully let us buy this truck. Let's go. Is this a paid truck specifically? I don't know if these trucks are blocked. They're blocked in the U.S. or just you would have to import one yourself. I think this one from Toyota internationally sells for like 10 to 20 grand. Like it's very cheap.

If I ran a company, my first act would be to ban hard-boiled eggs in the office. Hard-boiled eggs are really an insane snack. It's not more protein than a steak in a bag. You've got to go Chesky mode. Throw a steak in the bag and take it with you everywhere. Like I wouldn't have a problem with somebody going if there's like a cigarette smoking area, like a back alley, and you can go into the back alley and scarf, scarves, hard-boiled eggs.

That seems reasonable. Am I going to go home and trick or treat with the kids like this? No, I do not want to traumatize my young children. Yeah. I will be dressing up as something else, but we will see you back here on Monday. We'll talk to you soon.