Reed Albergotti on Anthropic's numbers, open source AI bans, and teasing his Satya Nadella interview

Speaker 2: Very cool.

Speaker 1: Stunning.

Speaker 2: Anyway, we have our first guest of the show. Reed from Semaphore is here. He's the San Francisco bureau chief. You know him. You love him. He's got scoops. The new Porsche's got scoops. It's... I swear they

Speaker 1: made this car for the lone werewolf journalist that's The gumption. Scooping around scooping around SF. That...

Speaker 3: I I don't know how many journalists you guys know, but you know if you buy a super car, you can no longer be in journalism. You have to leave. That's... Return. And be a journalist.

Speaker 1: Yeah. Fuck the trend. Fuck the trend.

Speaker 2: Yeah. You you need to be driving like a... I don't know, like an Oldsmobile. Something broken down.

Speaker 3: Oldsmobile or

Speaker 2: something or...

Speaker 3: A very old one.

Speaker 1: Yeah. Something that you can pull pull out, know, pull up outside of a of a technology company office and smoke a cigarette. Yeah. Just act sort of mysterious.

Speaker 2: Even like a '19 A

Speaker 1: detective.

Speaker 2: Even like a 1998 s class, that has some weight to it, but it's going to be depreciated. You can pull up and have some authority there.

Speaker 1: Can we get...

Speaker 3: I know if you've car. You... Yeah. Was... Even owning a car is probably...

Speaker 1: Yeah. Go ahead. I don't know if you've been following, but any predictions for the Tesla Roadster later? We've been debating, are there gonna be ramps involved? Are there gonna be... Is it just rocket thrusters?

Speaker 3: What are guys are really... It gonna be

Speaker 1: all the kickflip? Things like that.

Speaker 3: You guys are you guys are in a very different place with cars than I am. And my... Like, I I have a Model y, and I'm just like, I just need a bigger car. That's all I care about. Just need more space for the kids and the gear, you know, and all that stuff. Like, a Roadster, I'm just not... That's not where my head is at right now. Yeah. Go go go go

Speaker 2: The Model y l, the long wheelbase version gives you an extra 20 inches or something.

Speaker 3: I was super excited. I wanna go... When my lease is up, I'm gonna for sure go check that out.

Speaker 2: Yeah. Yeah. Seems like...

Speaker 3: Just make it bigger. That's my that's my ass.

Speaker 2: I have one on order and they pushed me back because they can't make them fast enough. I was supposed to get it in November and now it's January.

Speaker 1: Yeah. I think my theory is that Tesla is, like, starting this crazy inflection where enough people have had the magical experience of self driving.

Speaker 2: I've heard

Speaker 1: so And many what I'm actually fascinated by is, like, I... It doesn't feel like Elon is actually giving it enough attention. And I think in part that's because he wants to merge Tesla and SpaceX. And he owns more of SpaceX. Right? So he wants Tesla to be at a lower valuation

Speaker 2: Yep.

Speaker 1: When that happens. We've both Good considering. We've both, like, put, you know, we both, like... I bought a Tesla a few weeks ago. John has been trying to buy a Tesla. Both of them were delayed by months and months Yeah. For, like, basic models. And so Yeah. I think that inflection is happening and I... But I don't think you're gonna get the like Yeah. Every single every single person I know now is having that like magical moment.

Speaker 3: I'll tell you, I have a broken shoulder right now from playing ice hockey and activated my FSD. It's kind of amazing. You just sit there with your broken shoulder and have the car drive you around. It's not it's not so much how hard I'm going. It's like how hard my old man body went to the ground when I was brutally tripped in Oakland. But my point is... Wait.... Ice skating rinks.

Speaker 1: Was it too broken to throw hands after after you got taken out?

Speaker 3: Way too way too broken. Drop the Way too broken. It's just me sadly skating off the ice thinking this is probably the last time I lace them up, but Yeah. You know. Dang.

Speaker 1: We should have been We should have been there. We would not have let that let that let that go.

Speaker 3: Yeah. This is the second time FSD has helped me with a... With an old man hockey injury. So I got to say, it's, you know, there's some marketing opportunities there.

Speaker 2: No. No. The the FSD is is like overpoweringly good to the point where people who don't like electric cars because they wanna hear an engine are, but I still had to get one as the second car, like, for whatever reason. And then the political thing, people are, like, look, the midterms are coming. Like, Elon's gonna be out. I like... I I I don't like him, but I'm doing it anyway. And I've heard that from a lot of people too. So people are begrudgingly buying them just because they

Speaker 3: Everyone said they would never buy a Tesla. And I... Every day at school drop off, there's a new Tesla, you know, that's that's post... You can tell it's post Elon because it's got the new... Post Doge because it's got the new headlights,

Speaker 1: you know. Yeah. Yeah. I'm getting the bumper sticker that's... I bought this after FSD became so undeniable that I felt it was the safest option for my family and others on the road. Bumper sticker. Right?

Speaker 2: Yeah. Yeah. Exactly.

Speaker 3: Well, my bumper sticker just says morally superior.

Speaker 1: That's all.

Speaker 2: That's all.

Speaker 1: Oh, okay. That's good. Yes. Yes. Yes. That's that's ideal.

Speaker 2: And then they can apply it to whatever the topic du jour of the day is. It is. Whatever it is. You know, you're prepared.

Speaker 1: Yeah. I... My other car is the moral high ground.

Speaker 3: Yeah. Right. There you go.

Speaker 2: That's great. Well, what... Yeah. What what what else are you what else are you tracking? Are you... I mean, we were at the start of the show absorbed by the anthropic numbers, obviously, historically massive, but still not beating expectations. And so, yeah, what are you tracking today this week in general? What's captured your fascination?

Speaker 3: I think... Okay. So so shameless plug, I've got Satya and Adela on my new podcast Amazing.

Speaker 1: That which... That'll air next week. So I was in... That's the best.

Speaker 3: I love the I love the music. So that's gonna air next week. But a little teaser for that is, you know, I asked him this question. You'll have to watch it to see his answer, but, you know, Microsoft's doing great. Wall Street loves Microsoft, but, like, look at the growth of all these companies. NVIDIA is now out earning them per quarter.

Speaker 2: Yeah.

Speaker 3: You got the hyperscalers. You've got SpaceX AI coming into the fold. Like, is there a world in which Microsoft is still a great business, like, a few years from now, but they're not in the mag seven? Oh. And I heard a little gasp from the the Microsoft comms team when I asked a question, so you'll you'll have to... But I I think this is...

Speaker 1: I'll just say Satya Satya's my...

Speaker 2: There's comms people can get aggressive. Are sure somebody didn't chime in and say, you are sure you don't want to talk about something politically controversial instead? Anything but the

Speaker 1: price I'll I'll just say, like, Satya is my favorite communicator. Like, he's like... In in the Mag seven. Like, he he's the one that I think is generally like the most candid Yep. And actually pick up on on how he feels about different things in a way that the other CEOs... Yeah. Like, does it in a way that's just... Again, I I think he's very candid. I think he's very fair.

Speaker 2: He's not stiff, he's not informative.

Speaker 1: Yeah, exactly.

Speaker 2: And it

Speaker 1: feels like he's just on some set of talking points and just always looping back to that. He's actually like, oh, let me answer this question. So I'm excited.

Speaker 3: And if he does have a talking point, it's like he it's like he's written a blog post on exactly the question that you're asking him, and he's got, like, the perfect... He's thought it all out.

Speaker 2: Yeah.

Speaker 1: Yeah.

Speaker 3: Yeah. And, yeah. I mean, this this was up at the... They've got the CISO Summit up there, and

Speaker 2: Sure.

Speaker 3: He's been thinking about security. I think the way he... It's actually really interesting to hear him talking to these CISOs, because he was saying... He's like... He's been sort of... It's almost like a let down. He's like let down by these models. He's like, I don't think the alignment is working. He doesn't think we're gonna mechanistic interpretability, so you have to put all these guardrails around this stuff. Mhmm. And that was that was fascinating to me. Like a little bit... There's like a sliver of defeat there, but also like, we're just gonna figure out how to make this work. Like, we're gonna put secure, you know, deterministic software around it and Yeah. You know, we'll use it that way. Yeah.

Speaker 2: What is your read on Xbox Labs? We talked about it yesterday, and I I love the the decompiled mashup games. I mean, they're super fun as, like, memes. I don't necessarily actually think that they're blockbuster games, but I I thought it was interesting that Xbox was leaning in, but what's your interpretation of that story right now?

Speaker 3: Oh, yeah. I totally agree. I was talking with someone, I I think someone at Xbox about that, you know, earlier in the week. I think Asha is probably the right CEO for Xbox right now, because I think she's like, you know, she's a little more forward thinking, she's flexible, and I think they're like, yeah. Mean, if you can't beat them, join them. Like, there's nothing you can do. If some kid wants to, like, mash up, like, decompile a game in their, you know, in their home using Claude, like, there's not much you can do about that from an IP perspective. So, like, I think they have to... The video game industry is gonna have to rethink everything. I think it's probably gonna accelerate cloud gaming, is my guess, like, in new new ways, and I'm excited by that. And Gen z feels loves

Speaker 2: like they're in such a good place as effectively a private company, because they're owned by Microsoft, so they can they they can kind of like hang out in the sub balance sheet and completely upend the business model. Whereas if you're an independent video game company, you you have a business model that relies on you earning a certain amount from micro transactions and selling a certain number of disks and having this particular business model. And you're gonna miss your quarterly earnings if you upend everything, but they're kind of in a situation where it's never been a big driver for Microsoft. It's been awesome for culture. It's been really cool to work at Microsoft and be like, oh, we got the Gears of War over there. But financially, there's a lot more flexibility. Do you agree with that?

Speaker 3: I totally agree. Yeah. I mean, they they still... I mean, the console business is no joke. Like, it's like Yeah. It's like a great business. Yeah. And I think they have that, but they will... I think they realize like they probably have to start disrupting themselves a little and they can do that. I agree. They have the luxury.

Speaker 2: Yeah. I mean, it feels like it's just good to be in hardware too because, like, we saw the Mac Mini was not ever built as like an AI device. Like, it's very much the opposite thesis from NVIDIA, the graphics cards, right, which got used as for for all sorts of AI stuff and eventually developed server chips. But the Mac minis blew off the shelf and you could see people sort of decompiling or hacking Xboxes and being like, yeah, it's just a good bargain. It's a good product. It's reliable and I can run anything on it or I can install an agent on it or I can, you know, actually hack and vibe code and do all these different things. And so they start flying off the shelf and actually becoming like differentiated and they're just going to sell a lot because even if they're not even if they don't have that huge margin of IP on top, it's still just a great product to be selling.

Speaker 3: Yeah. I also love the... I love this concept of just like hacking old hardware. Like, you can actually do that now. It's like

Speaker 2: Yeah.

Speaker 3: Why not? I mean, this stuff is so expensive. I'm like going through my drawers now looking for old hard drives. Because Oh, like I don't wanna buy a new hard drive.

Speaker 6: But storage is expensive.

Speaker 3: Yeah. Mean, it's like, why not? Like, just try to... Just try stuff. It's it's an exciting exciting time.

Speaker 2: Yeah. How are you how are you thinking about the open source alliance? Like, all the things that NVIDIA is doing, the different groups there. Open AI has been kind of like one foot in the other, like we don't want to ban it, but we also aren't really open sourcing stuff. Meta is like open sourcing stuff, but you know, has something very close source that's working now. Anthropic is still very strong against, but what is the shape of the open source alliance these days?

Speaker 3: I think, yeah. First of all, I think the hyperscalers would love to ban it. I mean, I was talking to our DC people, like the open fill funded groups are having dinners talking about open source is gonna be banned. I have no idea how that's even possible. I mean, that just seems so draconian and and really ultimately just like a fool's errand that... Yeah. I mean, I think this stuff is gonna... Is gonna total... It is taking off. Right? Okay. Open source

Speaker 2: I want to fight you on that because I think it actually may be very easy to ban open source. In in the sense that like we banned we banned file file sharing. Like

Speaker 3: Right.

Speaker 2: And and it's like, yeah. Obviously like torrents still exist. And you can go and find an offshore website that has a torrent, but you will get a letter from HBO if you download the latest season of True Detective illegally. They will find your IP so you got to be using a VPN and then it just becomes cumbersome. And you can't have a big data center hosting files like mega upload, that whole saga went, came and went. And so, it needs to be distributed like weights that are, you know, on this like dark web torrent network that's kind of cumbersome. And then the really dangerous stuff is always like a million dollars of compute. And I don't think you're going to be able to run anything as an attacker on a Mac Mini that can do anything but bounce off, you know, CrowdStrike plus mythos plus GPT cyber plus Palo Alto. Like these organizations are going be running like every...

Speaker 1: To be clear, you're not you're not advocating for it.

Speaker 2: No. I'm not advocating. I'm just saying that like...

Speaker 1: Yeah. It's like banning serving it is quite different. The other thing...

Speaker 3: I think that you're... Oh, go ahead.

Speaker 1: Yeah. Yeah. Was just going to say that the... Like the zero day that Expo found Yeah. Chrome was... That affects... Would have affected a billion people Yeah. Was like $40,000. Right?

Speaker 2: Yeah. Yeah.

Speaker 1: Okay. So...

Speaker 3: Okay. Okay. But you're assuming that it... That you you actually get it banned. Right? Like, I think I think you're right. If this gets fully banned, like, there are technical ways they can just make it a total headache to Yes. Download open open... They could just, like, kill that industry. Yes. But, like, is it... The legal framework for just saying, like, any kind of open source AI model is is banned is gonna be so difficult. Right? Yes. What are... What is the theory there?

Speaker 2: Yeah. Yeah. Yeah. So so like, as a practical like, as a practical matter of how things get done in Washington, that is more intractable than the technical question of like how do you find out that a data center is serving an open source model and send them a sternly worded letter, hey, stop serving this because we banned it. But, yes. So so so I agree with you. It's harder to actually ban than enforce a ban that that happens potentially.

Speaker 3: Yeah.

Speaker 2: Interesting. I've always... What do you think, I mean, from talking to the DC folks, what do you think of this idea of like you can sort of chart the path of technology regulation by just summing up the market caps of various open letters? Like when the open source letter happened, I was like, oh, this is like very rough for... To be the anthropic position of like, let's ban it. Because I was just looking at... It's like Nvidia 5,000,000,000,000 and then you have another company that's like, you know, a couple trillion and I'm just I'm just adding up the market caps and I think it was like $25,000,000,000,000 of market cap was saying like we support this thing. And whether or not DC is coin operated, it's just a fact of the matter that when you have that much economic weight behind it and that many people's pensions and all the different things that like that might dictate the actual outcome.

Speaker 3: Oh, yeah. I mean, let's let's face it, like DC is somewhat coin operated, like it's not, you know, it's oversimplifying it and I'm out here in the Bay Area. Easy for me to say. I don't have to go to dinner with these people every night.

Speaker 2: Sure.

Speaker 3: But I mean, yeah, I think you totally can. But I also think it's like who... It's like there's not some law that's being violated by this already, but you have you have to create a whole new idea, a whole new legal framework. It makes it it makes it really difficult. And then, like, the other the other thing I think with... Like, a name a tech regulation that was actually good, or or didn't have some major blowback, like even section two thirty, which is almost like Yeah. Unregulating rather than regulating is like hugely debated today. The DMCA that that governed all the downloads and torrenting and everything, that's a huge mess. Like, that would just Yeah. Created headaches for years. Still, it's creating all these issues with doing, security research. Right?

Speaker 1: Like, you don't... You can't... You have to

Speaker 3: be very careful about how you Yeah. Research bugs and software because of that. So, like, I don't know. It just... To me, it doesn't make it doesn't make much sense. It makes a lot more sense. Like, it's like you have the money behind it, and also just like the logic behind it. Like Yeah. There's... It makes logical sense. Like, let's have open weights. Let's have open source harnesses and sandboxes. Like, I wrote... I had a little scoop today on Perplexity open sourcing their sandboxing technology. I mean, like, let's let's, like, have... Let's crowdsource, like, figuring out how to keep this stuff in, you know, in silos and make sure it's safe. Like, that makes a lot more sense than like, let's have two two labs or labs companies Yeah. In Silicon Valley, like figuring this out all out for us. Like, that that never works. Yeah. Yeah. I'm preaching to the choir. I realize that.

Speaker 2: No. No. No. No. It's interesting. I'm I'm I'm thinking about, like, the, like, AI is maybe, like, the least frustrating it will ever be. I'm just imagining, like like do we get the AI utopia or do we get the doom scenario where like the median case is like probably like we get cookie pop ups. So we get some like annoying thing that happens that like is like constant and like and it's just like the the cookie pop ups development and like the way that like that's how the internet played out is just so depressing every time I have to click that button. I'm just like, how did the... How was this the outcome? But I don't know.

Speaker 3: That's That beside the was a consumer thing. Right? And like Yeah. I I think there's actually... Like, my theory right now is like the reason there isn't more consumer stuff happening in in AI Yeah. Is that there's just not enough compute. Like, we don't have you just don't have enough data centers to run, like, the frontier models for consumer applications. Yeah. And and I think open weights is probably the way that happens, like, Before you get to, I don't know, like, the web two point o of AI, where it's, like, all in the cloud, you're probably gonna have some sort of like the Netscape of AI first.

Speaker 2: Yeah. Yeah.

Speaker 3: Yeah. Runs on your computer and it's, you know, it's not why... It's not like everyone uses it, but it's sort of like early adopters get into it. Yeah. And we're like not even there yet.

Speaker 1: I mean...

Speaker 2: Yeah. I mean, you see it with like, you know, Muse is having like a crazy moment. Open AI responds with Dot, but...

Speaker 1: Dave's Hot Chicken responds. Dave's Hot Chicken responds. Apparently working with Nikita.

Speaker 2: Yeah. Apparently. But like Dot, I think is like still rolling out through the paid tiers because it's like Yeah. It's actually difficult that they were like opening the flood gates and being like, yeah, billion people can just run this constantly. That's a lot of compute and so it it is tricky. Yeah. Where are you thinking the bottleneck goes next? Like, is it money? Are we gonna run out of money? Is that the real problem?

Speaker 3: Yeah. We... That's actually... I was talking about that this morning actually Yeah. With our our finance team because we're... They're obsessed with the fact that these companies are borrowing money.

Speaker 2: Yeah.

Speaker 3: I'm like, I'm on the other side of this. I'm like, this is not like, you know, this isn't like the financial crisis or like 2000 again. Like, this is a very different thing.

Speaker 2: Yeah.

Speaker 3: But, like, the banks are getting to the point where they're like, yeah, we just can't lend out this much money, and so I think the borrowing has been opportunistic by these tech companies. It's like, why not borrow that money? Like, investors like it. But if the banks are like, yeah, we're just... We're we're hitting our limit here

Speaker 2: Yeah.

Speaker 3: Then, yeah, I think money does become another bottleneck. The hyperscalers, you know, the the mag seven, they have enough cash. They can they can find other ways to fund this, but I mean, yeah. I I'm actually probably gonna write about this for tomorrow in the newsletter, how this is not a... This is a bottleneck economy.

Speaker 2: Yeah. It it flips so fast. I mean, I remember Peter Thiel... Bottleneck,

Speaker 1: I'm going to invest I'm to invest in cash. Oh, okay. Yeah. You got to get

Speaker 2: some But of I I I remember there was a Wired Tech Conference maybe twenty twelve ish, twenty fifteen ish. Peter Thiel, Eric Schmidt, who was then the CEO of Google debating uses of cash at Google. And and Thiel's point was Google has $50,000,000,000 of cash on the balance sheet. You don't have anywhere to spend it. You're out of ideas. We're stagnating. Go do something with it. And he's saying, oh, well, like we're spending a billion dollars on this self driving car thing and we're going to work on cancer research. It was all the same promises but it was just a decade too early where there wasn't actually real ways to deploy the capital. And and this was the true this is true across, you know, all the hyperscalers where Zuck would go off and spend $10,000,000,000 on something that wasn't really working. And now the opportunity is here and we've switched from like, oh, you don't... You're you're not you're not spending your cash, that's an...

Speaker 1: Also Google did invent the transformer.

Speaker 3: They did.

Speaker 1: They did. Did. Self driving cars. Yeah.

Speaker 2: But now Yeah. Now we're in the, okay, go spend the cash because there is an opportunity that can actually consume that cash and you still get criticized, but I don't know.

Speaker 3: No. I know. And it's like everyone's like, you know, everyone's talking out of both sides of their mouths. Right? Yeah. It's like, oh, they don't want big tech to become so powerful. And it's like, well, they're actually being forced to like spend all their cash right now. Like, that is a good thing. Right? That means there's there's like competition and disruption Totally. And you know, I guess if it doesn't work, then they'll go back to just having, you know, the most profitable cash cow businesses in history. But, you know Yep.

Speaker 2: Yep. Yeah. Mean, we saw that today with like, there's a rotating out of the Nasdaq, it goes straight into the Dow. And and you know, it's like, oh, if if the AI trade's blowing up, like get back into SaaS. Get back into the social network. Ad tech. Like the the stuff Yeah. This money just like flows around people hunting for the for the latest thing. Anyway...

Speaker 3: They're pretty well hedged.

Speaker 2: Thank you so much. Awesome. Are you

Speaker 1: well excited for the Satya episode?

Speaker 2: Yes. Go. Great to hang.... To go listen.

Speaker 3: Thank you. Yeah. Download Future Proof.

Speaker 2: Future Proof. Gotta plug Yeah. Podcasts are available. Thank you so much for coming on. Cheers, Rachael.