Wish there was, like, a public chart, like, where you could just see the number of normal people and the number of people who have their lost to the ether.
Are you normal? Yeah.
How are you?
I'm good. It's Friday, and we're trying to do a thing where we we usually go out to dinner every Friday, but we're trying to switch it up and do lunch every Friday.
Oh, okay.
Just you
and Louis? A little Friday afternoon lunch? Little A siesta?
A siesta comes after.
Oh, really? That's after lunch?
A siesta is a nap that people take after lunch.
Oh, that's the nap.
Think it was invented in Spain. Yes. And it's literally, like, a real thing. Like, everything just shuts down
for hours. Town to shut. That's so cool. I wish we did that. I mean, I can't really I wouldn't wanna take a nap, but it sounds fun.
I'm not someone that really gets sleepy after eating in, like, late afternoon, but so many people do that I'm just convinced there's some biological thing where we're supposed to do that for the most part.
So I I thought it I didn't know it was Spain. I thought it was, like, Italians, and they all just ate, like, a ton of pasta, and then I get sleepy after sleepy after pasta. So I can imagine wanting to take a nap. Like, a big baguette. I don't know. Or is that France?
Baguette baguette is French. Well, you're really confused about Europe. I'm really confused. Going on over there.
What do people eat in Spain? I have no idea what they eat in Spain.
Like, I'm imagining in your head, you're, like, imagining Europe, and it's just like a table with, like, pasta and a baguette. Pizza. And, like, and pizza.
Yeah. American pizza.
Yeah. What do Spanish people eat? Spanish food is really good. We eat a lot of it here, like tapas.
It's like why like, oh, tapas. What's a tapa?
Tapas isn't really a thing, really. It just means a lot of small plates, which is a great way
of eating.
Like, they just give you Like, Ethiopian. Different things. Yeah. It's kinda like you do a few food. I mean, same concept where you're just getting a lot of things.
Mhmm.
Yeah. But there it's not like stuff you dip in other stuff. It's all stand alone.
Oh, gotcha. Do they like like, like, fried rice? No. That's Asian. Wild rice? What am I what am I thinking about with rice and Spanish people?
Yeah. They, what's there's that famous dish with the seafood. I'm forgetting I'm forgetting the term now.
It's probably something different because I didn't know there was seafood involved. It does make sense. There's a
lot of seafood in Spanish
food. Yeah. That makes sense. Spain on the ocean.
Like in it. Paella.
Paella? No. Never heard of it. Sounds awesome. Well, I don't like seafood, but but there's rice. I like rice a lot.
There's rice in it. I forgot you don't eat meat for a second. How can I forget?
I do eat a lot of rice. I love rice.
So my favorite thing and may oh, no. This isn't the thing I'm thinking of isn't isn't, hang on. I gotta look this up because it's driving me crazy. Ceviche. I'm thinking of ceviche.
Ceviche. I don't know what ceviche is. I do not.
Okay. Well, ceviche is raw. It's kinda like sushi. It's like a mix of different raw fishes or, like, just stuff like that. But they use a ton of lime and citrus, and the citrus acid cooks the fish. I think it is, like, nice, like, tangy, sushi like flavor, but it's, like, you know, technically cooked. And that's interest I think it's
a Like, it literally cooks it like they don't use heat, just the citric acid.
Yeah. Wow. It's like a form it, like, it's like a version of it being cooked. Yeah. And it's Interesting.
The history behind it is interesting because I haven't even recently understood this. So that comes from Peru. When you think Peru, South America, you have, like, an image in your head. But there's a there was some point in history where a ton of Japanese people moved to Peru, and their cuisine is crazy interesting because of that. Because the ceviche is like sushi, but, like, South American style.
Oh, but South American interesting because Japanese people moved. That's so cool.
Yeah. And now they have, like, whole weird dedicated cuisine, but it's from that. So there you go. A little fun fact for you.
Did is paella actually a thing, though? That that word?
Is a thing. It's a Spanish dish, and I don't like it because they just put, like my my perception of it is they're just like, let me just take every possible sea creature and just throw it in the pot and put rice with it. Like, they just put everything. I don't even know what I don't even know if there's, like, an official set of ingredients, but it'll be, like, clams and shrimp and, like, oysters. Like, just, like that might not even be correct, but I just know that I just don't like it.
It's like every time we take a spoon, there's, like, a different sea creature on your on your side.
So I
don't like paella. It is really popular. I don't like it already.
I feel like that word, Paella, I think that's a town in Maui. I think that's why that word is familiar to me. I'm pretty sure. Yeah. It's like a little town in Hawaii, and they have this awesome health food store called Mana Health Foods.
Oh.
And it's super cool. Anyway
How are you imagining when I say Paia, how are you imagining it spelled?
Like, P A I A, maybe?
There's two l's in it. What?
Oh, okay. I had the flu. That's I l l a. We didn't do last week because I was sick, and I've still got the cough. So it's kinda gross, and I'll try not to do too much in the mic. Maybe Chris can edit out all the coughs.
Sounds fun.
I'm just gonna cough over everything you say and just make it hard for him.
Is this a new thing where when you get sick, you just stay sick for a month? I feel like this just was not a thing before.
This was not a thing before. I thought so I thought I had a cold because it started out just like every other cold I get, and I get, like, one every quarter. I don't know. Like, a little two day, like, kinda have, like, sore throat, whatever, a cold.
Yeah.
I thought it was one of those. It was not one of those. It was the flu, I think. Influenza a, bird flu. I don't know. There's a lot of words floating around. But I was out for a whole week, and the cough is still going. And it's been two weeks. So I think and I've heard that from other people that a cough, like, lasted three weeks or whatever, which sucks.
When I had COVID, that that's when I had that. So I'm wondering if, like, it's, like, COVID related now. Because, actually, everyone just has when they get sick, they're just off for, like, a month. You know, they're better after a week.
I thought I had COVID because, like, before Casey saw on TikTok that it's influence a, I thought it was COVID because I could've sworn I couldn't taste anything for, like, two days. Stuff that I ate all the time just tasted like nothing. Maybe COVID mixed with the bird flu. I don't know. I'm not a scientist.
When COVID was, really peaking, you know, when all the news reports were like, New York is the epicenter of COVID.
It is
a disaster.
Pop up tents and morgues or whatever.
Yeah. Yeah. I was like, I'm going for really nice walks every day because there's no traffic. Liz lost her sense of taste for two days and had no other symptoms at all.
Really?
It was really weird.
But she got the taste back. Because I've heard some people lose taste forever. Right? Yeah. Yeah. Some people die forever.
Because they can't taste the food?
No. No. Like, some people die, and then they're dead forever from COVID.
Yeah. Okay. That also happens. Yes.
You can lose your taste forever or your life forever.
You're right. You're right. COVID's not good. I agree. It's not good. Controversial.
Is it still is it actually still active, though? Is there still COVID?
I think so. I think it's kinda like the flu now. You know? I also like, this is also very stupid. I didn't put it together that the flu and the Spanish flu are the same thing. And the reason we have the flu every year is because the Spanish flu happened.
Wait. What was the Spanish flu? I don't know history that well.
The Spanish flu was, like, the last big pandemic prior to COVID. Oh.
And that was the fir that was the start of the flu? That was, like, the first flu?
Yeah. There's this word I think it was called endemic. Like, a thing like that shows up and then becomes endemic, which means it's just, like, part of the world. Yeah.
So that's what COVID okay. Well, then that's actually comforting that, like, there's precedent for this new disease humans have never had, like COVID, and then it's just kind of around. And it's not gonna just suddenly morph into something that kills us all, it sounds like. The flu didn't do that.
Well, I think there's, like, funny evolutionary pressure with diseases because the most optimal disease doesn't kill the host. It, like, keeps the host alive so the host can infect as many as possible. On everybody. Yeah. So there's, like, a pressure to, like, not be as lethal.
But then I think in theory, it could still be lethal. It could be effective if it was lethal after, like, a month. Right? So that's a big answer to make this so scary. But I think that's why COVID rapidly, like, became more about infecting people than being lethal.
I heard this back in the COVID, like, when, say, the twenty twenty, twenty twenty one days. I remember somebody talking about, like, a disease that just kills you immediately, like, doesn't get to spread, so it's not that actually dangerous Yeah. Well, to, like, a population. Yeah. That makes sense.
But, yeah, what what about the one that's, like, a time release? Kills you after two weeks exactly after you've infected everybody. Hoo. That's scary. Let's not let's not create any of that in, like, a lab somewhere. That sounds bad. Yeah. What what's the deal with that? Could you actually tell me? Because I I didn't keep up all that well, but I know, like, everyone says people lied, and it actually was made in a lab. Is that true now? That's a fact?
I mean, we're never gonna know. I think somebody out there knows for sure. It's like someone in the Chinese government knows. But, like, I doubt we're ever gonna know anytime soon.
Okay. So it's not established fact as Twitter made me blush.
It was just so funny because when COVID was happening and Jon Stewart has a really funny bit about this where he goes in the Colbert show, and he just keeps repeating, the lab in Wuhan was called the novel coronavirus lab. Do you think it's somehow related? Like, he just goes off, like, a five minute run
about that. Wasn't it, like, just really close to where the first case was or something?
It was literally in the city, and then people were like, no. The lab's there because there's coronavirus is there, and they're there to study them. And it's just, like, there was a time where saying that, hey. Maybe it was leaked out of the lab by accident was, like, this weird political landmine, and then you're, like, you weren't allowed to suggest that. Not saying that, like, they engineered it on purpose and released or anything.
Just
Yeah. You can study diseases, and they can get released by accident. And, like, an Occam's razor thing is, like
The simplest.
It says the wound coronavirus left. Where does virus come from? Who knows? Like, maybe we should check there first.
Yeah. The the hard thing for me about these debates or about, like, hot topics like that is I can't follow the discourse. Like, I just can't get in the headspace of either side, and I'd I'd lose the arguments. Or I'm the same way with compute conspiracy theories. Like, I can't ever, like, keep track.
Okay. So why do you think they did that? What was their motivation? Like, I can't keep it all straight in my head, and then I just lose the train of thought, and I'm I'm out. It's like I'm not smart enough to be a conspiracy theorist. I feel like you have to be pretty smart.
Well, it's funny you bring this up because I had this I was doing a bunch of self reflection yesterday, and I had this thing I'm gonna try where I'm interested in stuff. Like, I'm interested in stuff that's going on
Me too.
In the world. Oh, okay. Yeah. I'm sure you're interested in that too. I'm sure most people are. Mhmm. Things have just gotten to a really bad place, I think, when you're trying to learn about the world because
I feel this.
You have these two options. Right? So yesterday, I heard that there are some, like, explosions in in in Israel. And I was curious. Okay. What what what's going on? Like, is there another attack? So your options are Google it, and you land on, like, the worst written article ever where there's, like, this site is garbage. It doesn't load. The information you want is, like, like, one sentence in, like
And then an ad and then Yeah. Yeah. Yeah.
It's just like it's just so hard to parse it there. Mhmm. Twitter will have the information pretty clear, but it's really hard to avoid the crazy extreme stuff in the comments where everyone is saying, like, the most batshit insane thing. Mhmm.
Mhmm.
And that just, like, is a drag mental drag on you. And I realized, hey. I think I just wanna experience a world through AI. Because AI solves this perfectly where when I ask Grok, hey. What's going on here?
It tells me exactly what I need to know. There's no garbage in there. It ignores all the extreme political stuff that shows up everywhere. And I'm like, I think maybe AI came out at the perfect time because I think just taking the raw feed of the Internet, it just gotten to a point where I don't think I can deal with it anymore.
It's just
it's too much.
Yeah. It's too much. I'm not able to sift through it. Like, I feel very vulnerable to, like, being swayed or just, like, believing things that I thought were just supposed to believe and that wasn't actually true. Like, I just don't know what's real anymore. It's funny the Grok three three thing. Like, Grok three just came out. So you use Grok, like, for your daily questions about the world. It makes sense. It's connected to the Twitter. Right?
Yeah. It makes it, like, 10 times better than any other l m because I can ask it about what's going on today.
Yeah. So Grok three is a good example where it came out, and it's like some people are like, it's the greatest thing ever, and some people are like, this is the most dog shit model that's ever been put in the world. And I
don't know what's true.
Yeah. It's everything that happens is like that. It's like there are people on extreme ends, and they're the loudest, and they get the most reach, and that's the thing
that ends up in your feed. Yeah. And and, Jay was bringing this up yesterday. He was like, you you know, for us as SST, we're we're in you know, we have a company and being able to parse what's going on in AI and, like, navigate that is very critical for, like, our future as a company. Like, we really need to, like, have a good understanding, a good thesis, and make good bets to, like, make it through to the other side.
And he was like, I think we need to not consume anything in, like, the public discourse about it because it's just such ridiculous, insane noise. And you, like, can it's very hard to have a clear head and, like, a clear vision of what's gonna happen when you just have all this crazy stuff being talked about all day. So, yeah, I'm trying to, like, cut back on on that. I'm trying to, like, only again, this whole thing of right only thing that I've I've talked about before, like, I wanna, like, try to really focus on making sure I'm not taking in
Yeah.
That energy. Yeah.
Yeah. No. I I I feel that, and I'm not doing a good job. I think I'm still bringing in that energy. I don't need it.
It's hard. It's really, really hard this day.
It's it becomes like you're yeah. You get stuck on a problem or whatever. You just you wanna numb for five minutes, then, like, you have your go to place. Some people, that's Hacker News. I don't understand it. But for some people, it is. For some people, it's Reddit. Some people, it's Twitter. And it you're inevitably getting fed a lot of stuff, I guess, at this stage.
And I think like I said, I think there's a lot of potential for AI to help fix this. And we we Jamie made this funny point the other day where I was talking about how sometimes I'll suddenly feel like, Hey, for the past couple of weeks, like going through Twitter has been making me feel really negative. I won't realize it right away. It'll take a couple of weeks
and
then I'll realize, oh, I followed this person a month ago and they're not explicitly doing anything wrong, but they're, like, reposting stuff or, like, somehow injecting stuff into my feed that I that is technically making me feel that way.
Mhmm. And
it always takes me a little while to notice. And when I notice, I I, like, track down the source. Okay. This is, this is what's going on. This is why I'm feeling this way.
And I'll like cut the source out. And then Jade made the point was like, Yeah, I bet that happens all the time due to some algorithm change that in this case is something I did. I like controlled it so I can undo it. Yeah. But theoretically, there's some algorithm change that could happen somewhere that now suddenly is, like, making me feel a certain way.
Mhmm.
And that's that that must happen all the time, and it might be hard to to deal with. So I think there's an opportunity to make some kind of product that I still wanna take in information. I still wanna scroll. I still wanna, like, you know, consume stuff that's going on. Like, I'm not trying to cut all that out. I know some people are that's not where I'm trying to go. Yeah. But give it to me through, like, this cold neutral AI that is pretty boring in a lot of ways.
Mhmm.
Run it through that, and I can kind of see how that could be reliable.
Can you just can you just ask an LLM, like or ask Grok, like, give me the news? Like, what's going on?
You can. Yeah.
And it does, like, give you the top. Okay.
Yeah. So that that's, like, a great starting point, and I think that's what I'm gonna try to, like, build some habits around. But I can see how there's some kind of opportunity for, like, a new kind of product that doesn't involve other people at all. It feels like social media. It feels like something, but
Yeah.
It's not generated by other people.
But okay. Now I have questions. Like, do you imagine this thing, like, having, comments that are not from human? Yeah. Yeah. Good. Like, Reddit, but, like, AI and Yeah. The all the actors are AI?
Yeah. And you you have to be, like you know, if you wanna go that far, you'd have to be pretty clever about how to, like, make that interesting and engaging, like, generic. Like, I think AI does a great job of summarizing, hey. Microsoft put out this new quantum CPU. Here's the deal. Here's the criticisms. Like, here's why it might be great. Here's why it might not be. I think it does a fantastic job of just coldly Yeah. Failing that.
And taking a step further to have, like, something that seems a bit more human, I think, is harder, but I can see how you could do that.
It is interesting just to think, just as something as simple as the Reddit model where, like, AI agent or sorry. AI models or whatever, little people that are actually AI, they have their own little account, could actually upvote stuff and actually comment on stuff. Like, what would the what would that result in? Like
Yeah. Yeah. Like because
because the LOM output is very, like, vanilla down the middle. But if you took, like, a whole bunch of it and you let them all interact like that, would interesting novel stuff actually come out of it?
Yeah. Like, there's definitely, like, tropes of people that like, there's obviously negative ones, but, you can create a version of the AI that tends to lean very libertarian, or you can make one that tends to lean, like, you know, very socialist and whatever. And you could, like, have them
Interact.
Engage in something within parameters of, like, not being too extreme and, like, actually understanding what the other person's saying before responding.
Okay. I'm sold. Maybe AI is the answer. It solves all this information overload and just biases and all these weird things that are going on in the world.
Yeah. Yeah. So I think that there's something there. And to yeah. It would be a consumer product, which is great. Especially now, there's so many appetite to try anything AI related. So if someone's interested in building something, I think there's something to explore in this category. I I'd like I feel like I want I want this.
We'll build it, and we'll build it in the terminal. No. I'm just kidding. Great.
Built in the terminal. Perfect.
So I wanna talk now that you've just said that, like, you're trying not to consume all the public discourse, I wanna go into some public discourse. Mhmm. But before we do that, so that'll be dessert. We can eat a little meat. Do have you, not meat, but vegetables. Whatever. Just you know what I mean?
By the way, I'm going to Korean barbecue for lunch today.
I'm sure you are. Yeah. I'd like to I'd like to actually eat out tonight. I've been trying for, like, a week to get us to order. We just don't eat out ever.
And there's, like, a a restaurant in Nixa that is like they're an Italian restaurant, and they have a whole vegan menu, which is super rare in The Ozarks for there to be a restaurant with a vegan menu. But, like, the it's a husband wife, and the wife is vegan, and they have, like, the whole vegan menu. It's amazing. Anyway Nice. Shout out to Piccolo's. Did you see the Satya Satya?
Is that the the the quantum thing or the thing where you're trying to wind back the hype?
Wind back the hype. But I'd also like to hear what about the quantum thing. That sounds exciting. Go first. The well, I guess, did you hear, like, his comments? Yeah.
Yeah. I you
know where I heard about it?
I I was trying this grok thing, and that's where I heard about it.
Nice. Okay. Well, you probably know more than me. I've just heard, like, a reaction to I haven't heard him in the actual statement. Just kinda like, I've heard what he generally said. But, basically, like, kinda calming down and just he he said things that just, like, resonated as a person who's also tired of people like Sama. I don't know if he's actually tired of Sama, but was there something he said about, like, pulling back on the OpenAI partnership, or was that just a reaction stuff?
I okay. So I think there's, like, a lot of weird dynamics here where OpenAI has a weird deal with Microsoft where if they achieve AGI
Mhmm. I
don't did they, like, not did they kinda release Microsoft's ownership over it? It's something weird like that.
And a lot of his comments were about, like, how much he hates this term AGI and, like
Yeah. So I I kinda wonder if that's related where he's, like, trying to change the definition. Mhmm. Not change. I think his point of view is correct. Like, why do we we had AI, and we're like but then we, like, assign that label to something that was pretty primitive. So we're like, okay. Now we need AGI because, like, we already used AI. Oh, shit. But now we assign AGI to something too primitive. Now we need ASI, and that's, like, the next level. So it's just like, stop stop doing this.
Yeah. He's saying, like, let's call it 10% growth. Like, the world is growing economically at 10% rates. Like, let's call that. That could be AGI.
Yeah.
Basically, is that what he's saying?
Yeah. Yeah. And that and so I I actually just posted something, before I got on and kinda relate. I'm just gonna read it out loud. Yeah.
So I wrote, I'm not sure if the highly visible AI companies we see online are representative. We see a lot of activity in unheard niches, integrating AI to handle very specific business flows. I thought there's gonna be 10,000 of these companies each doing 10,000,000 ARR. So we're obviously biased because we only see, like, whatever segment of the market we focus on. But there are a lot of AI companies that are using SST, and every single one of them is identical.
They're in some weird space. I I I hung out with a guy, yesterday, and he's, like, in the ERP space focusing on manufacturers in South America. Okay? Like, you never hear about this. And he's building a and, like, his, his father has been in the space for forty years, and now they're building technically AI based tooling that is getting into these businesses that and now can kind of solve these problems that, you know, you couldn't really do before with software, like, stuff that AI kind of unlocks.
Yeah. But they are just if you look at them, they're just a standard software company. There's nothing that crazy new about them. 90% of their work is AWS, setting up Glue jobs, getting all the data in one place. Yeah.
And then, like, AI unlocks a bunch of things for the reason that they exist. But there's, like, ten years of work for them. And then there's, like, ten years of work ahead of them from the do to get in all these little places and then continue this and I think I keep bringing up Jay, but he had another good point here too where he was like, we had this whole software e in the world concept, but it didn't really go that far. And what we're seeing with AI might just be a continuation of software continuing to get into more places where it still hasn't gotten into. And I think it's very aligned with what what Satya is saying, which is, that might just be it's a very boring outcome, but 10% year over year growth
say.
Is insane. And it might be involved with, again, 10,000 of these companies getting into all these little places and covering the last set of possible to automate things that we haven't automated yet. Mhmm. And it seems more and more likely that that the, quote, unquote, boring outcome is maybe what's more realistic.
Yeah. I loved you've tweeted something about this, I think, in the last couple weeks. Just the idea of, like, building apps. So, like, enough of the shovels, and we're just so plugged in to the to the whole, like, dev tools scene, I guess, on Twitter that it does feel like that's the big focus for the people that we're around. I love the idea of, like, this technology enables all kinds of new products.
Go build those products, like, end user products. Yeah. That I mean, that was like StatMuse. When we started StatMuse, it was like, hey. What like, all this NLP tech exists, and this is, like, old school NLP tech.
Nothing like what you could do today. But all this, like, all these tools exist, and there's just no consumer products that are, like, using NLP. There was, like, Wolfram Alpha, and that was it. And it was like, why can't why doesn't this exist for, like, looking up my favorite sports player stats? So that's very much always been my thesis of, like, use the stuff that's available.
And so often, we we think we have to build new tools when, like, there's plenty of tools to build cool applications. It's way more fun, I think.
Yeah. It's way more fun When you're in a moment where everyone's trying to build shovels, it's, like, 10 times more fun because every single week
There's a new shovel.
Shovels like, oh, suddenly your models are now 10 times cheaper. Suddenly, there's, like, a better you know, it's just it's so much better to be on the receiving end versus trying to compete in this crazy ass noisy place. Right?
Mhmm.
And I think the final thing is is I think everyone people love people love this stupid quote, which is when there's a gold rush, build shovels. Like Yeah. Just look in history. The winners are the people that use the shovels. They're not the people that built the shovel.
Intel is about to die. Right? Apple built their company on top of Intel, and then they eventually just escape from Intel because they own the end value of the thing that they build with the end end user. Right? It's always better to be there. Yeah. And people just keep trying to build these shovels. And what bothers me is we're so early. We're, like, nobody really knows a question of what are good ways to use AI? Mhmm.
What are good architectures to use AI? What kind of product features does this make sense with? Like, we haven't discovered any of that yet. And then there's already people writing stuff about here's how you should build AI stuff. And they don't that's not their product. Their product is something unrelated. Like, how can you possibly know? Like, go build the end product.
Then you'll know what shovels are needed.
Five years from now, yeah, we'll look back in hindsight and be like, oh, like, this is a common thing. Let's, like, you know, make a standalone service for that. But, yeah, the ordering of this stuff just makes no sense to me.
Yeah. I love that tweet you had talking about, like, AWS came after Amazon, like, all the different examples of
Yeah.
The the big shovels coming after the products.
Yeah. You develop expertise first, and you can't get expertise without shipping something.
Other things Satya said, the infrastructure, like, basically, obviously, we're gonna build way too much. We're gonna invest way too much Mhmm. CapEx, and he's excited to lease from all the people that build too much infrastructure. It's interesting.
Yeah. I mean, that that's a basic rule of it's impossible to invest the correct amount. You can only over invest. If you wanna invest, you lose. It's unlikely you're gonna precisely invest the current correct amount.
So the only result is over investing. And then when the froth ends, then there's, some interesting stuff. I was kinda thinking this other day. I was like, you know, everyone likes to say they're contrarian, they make contrarian bets. But I think the craziest contrarian bet you can make is to build a business that needs GPU capabilities, but is too expensive currently.
Betting that in five years, all this GPU stuff is gonna be on the market for crazy cheap. Like, is there some kind of business that can really take advantage of all that stuff, you know, having to having to be sold off when some of these, you know there's only gonna be a couple winners. So the losers are gonna, like, have to liquidate.
Can can I make a really dumb comment or observation? Like, we're sure the GPU thing is the right way to do this. Right? Like I
It just it's so weird to me, like, that a thing that was made for, like, rendering video games then turned out to be really good at, like, this AI training and just just, like, the the serendipity of that and then, like, thinking how many bazillions of dollars are gonna be put into building these GPUs and these huge data centers full of these GPUs. And then what if somebody's like, oh, actually, if we just do this whole thing, it's the way better. And, like, it looks nothing like GPUs.
Possible. And it's kind of our like, if you look at okay. So in favor or kinda against what you're saying, a GPU, yes, was used for graphics. But if you think about what it is, it's just picking an opposite set of trade offs from a CPU. A CPU is about having really, really fast performance, but not a lot of parallel work.
Right? You were trying to get the fastest single thread thing. If GPU is like, okay, what if we flip the requirements and said, it's okay if each core isn't that fast, we just need to do a lot in parallel.
Parallel. Okay. It
happened to be that graphics was a great entry point into that, but then parallel computing in general is useful everywhere.
I see. This is why I said it was a dumb comment because I knew I don't know anything about any of this. It just it just feels like a huge amount of money and time and effort are gonna be put into building these huge data centers. Yeah. And it just feels like it's hinging on some very, I don't
know. Assumptions.
Assumptions. Yeah. Some some solid assumptions around GPUs.
I think I think I think the part where you're right is, why NVIDIA and why NVIDIA GPUs? And that's not really related to the hardware. It's more related to the, like, the software stack on top. Vibes. Yeah.
People people like using CUDA. And then
Well, there's no there's not much of an alternative. So everyone kinda under invested in the CUDA concept, and so NVIDIA is winning. But, yeah, like, conceptually, like, I don't see why that gap isn't closed given how just how much money is on the table.
Yeah. So the okay. So I know the answer that, like, this is we're still decades or whatever away from this, but what about quantum? Because I feel like
That's the thing that Microsoft announced yesterday.
Yeah. Yeah. There's there's constantly been headlines that, like it feels like we're actually making progress in the last year on things that people thought were gonna be decades away. Yeah. So, like, what happens when quantum comes along and it's way better at AI training and inference and, like Yeah.
It's possible.
Oops. But maybe that's twenty years from now, and it's like you gotta do this in the meantime anyway. I don't know.
Yeah. It probably is. Like, I think it might end up being I'm trying to think of comparisons where, like, where something like this happened, where there was, like, an investing in something and then something kind of came out that made it useless. I'm sure it's happened before,
but
I feel like transitions are always slower than than we'd expect.
Yeah. Yeah. What was the announcement, though? What did Microsoft announce?
So they they built a chip which has what they're saying is a new form of matter or built with a new form of matter Woah.
Or a
new state of matter.
I'm in. You got me. And
they're saying they've achieved things with it, that are a little bit further ahead than than they expected. And I read a bunch of the, criticisms. It's like, it ranges to this is nothing to, like, oh, this is good progress, but I don't think it's gonna, like, really lead to the next step in the way that they think it will. But from Microsoft's point of view, they're like, this is the next big step, and it makes our vision of enabling this, like, possible in the next five years. That's kinda how they're they're they're pitching it.
But, Microsoft, man.
That does seem like, yeah. How are they it's amazing how relevant Microsoft has remained. When it's like it seemed like they were irrelevant twenty years ago, and then they've just been more and more
ten years ago. Yeah. Yeah. Yeah. Turned it all around. Amazing.
Pretty pretty crazy. Yeah. I do think, like, all the headlines I've seen with quantum stuff, it seems like in, like, ten years, we're gonna have something. There's gonna be, like, actual products Yeah.
That are quantum powered. Yeah. This is an area I need to, like, look I need to understand better. I understand some of it, because one of my friends took a quantum computing course in college and, like, he would, like, tell me all about what he's learning.
I know the word cubit. That's all
I know.
I don't understand any of it. But
Yeah. It just confuses me because I'm like, how is this in the chip? Like, they were showing pictures of it, and they're like
Yeah.
I don't yeah. Quantum computing's in here. I'm like, but, like, how, though?
Oh, like How's
it in there?
Yeah. It's very confusing. Lots of fun, exciting, like, things happening in technology. When you think about, like, the fission no. Fusion? Fusion.
Fusion. Yeah.
Fusions is that's the hard one. Like, there's been huge announcements on the fusion side.
And I
feel like all these big progress, like, news headlines with quantum and with fusion and AI, it's all converging, and then AI is just gonna help us, like, get over the hump.
Yeah. That
well, AI did something in the research setting. Like, it helped somebody. This was a Google paper. Right? When I saw the headline, I was like, whatever. But then it was like, oh, it's from Google. There was some, like, thing that helped them develop a something medicine.
I I I think there's been a bunch of stuff in research, that are just less flashy, but more just there's a bunch of tedious work. Again, kind of what I was saying earlier, there's been a bunch of tedious work that was hard to automate before that now has been automated. It's funny because they were, like, someone from Google, I think, trying to, like, do this whole hype thing that, you know, Sam Altman's doing. He was like, oh, yeah. With DeepMind, we've done, like, 2,000,000,000 PhD hours of research already.
Or some I'm making them or something, like, ridiculous. I'm writing that.
Mhmm. And I
was like, oh, that's really crazy. And I was talking to Liz, and because she used to work in research, and she used to run, like, a, I guess, a lab. And then she was like, this wasn't related to this topic, but she was describing something about some of the work that they were doing back then. And you realize what two billion hours of PhD research means is they were, like, doing some research on, like, analyzing social media and, like, crawling with other stuff. They were manually going through every single host and, like, cat like, checking off a bunch of things.
Yeah. And then being like, oh, this
is for AI.
Yeah. Exactly. So it's not like, oh, like a genius. It's not like a genius research. It's more just they're just tedious work that that exists. And it's definitely super useful. But when they say, like, PhD level research, they're trying to imply that it's really smart. But Yeah.
Yeah. Yeah.
It's probably just stuff like that.
Interesting. Okay. If I have to pee, do we need to just go?
Yeah. I probably need to leave in a couple minutes. But what was your dessert? You said we had dessert.
Well, the dessert. Yeah. I I don't wanna leave everybody hanging. I just feel like the world is crazy right now with all this stuff. What what is the deal? The US and Russia are, like, buddies and, like, Elon and all this stuff is so bizarre. I don't know what to think about anything anymore.
That's actually what triggered my self reflection yesterday because I was seeing something from Elon being like, oh, all the community notes about on me about the stuff I said about Ukraine are wrong.
Those are false.
And so we're gonna fix the community notes thing. I'm just like,
what the
fuck is going on? Like, why we are trying to help Russia effectively. We're, like, taking the Russian angle on this situation weird on its own.
Weird. Yeah.
Why is that affecting my social media experience? Like, why is the features of social media changing because of this?
It's so bizarre.
I'm like, I I think I'm I'm very I can deal with a lot, I think, but I really think the world has gotten too weird for even even me. I just can't tap out.
It feels like like, it it really feels like there's this dynamic now where if your enemy politically in The US believes something, you have to believe the opposite Yeah. Then you have to, like, go very hard that direction. Like, there is no letting some things just be like I don't know, like, that side cares about it, but I don't care. You have to care, and you have to care hard.
Yeah. And,
like, that feels like this whole Russia thing is like, oh, they like Zelensky? Well, we we love Russia. We love them. They're the best. Like, what
in the world? Because somebody is asking tweeted that thing yesterday?
Yes. I'm going crazy.
Okay. I'm realizing that this all triggered it. I saw your post for that. I'm like, what's Adam talking about? And then I looked at the explore tab, and then I was like, oh, this is probably what he's talking about. And I had this whole, like Yeah.
No. I've been on the for you tab because my following tab's terrible. Yeah. It's because, like, I follow people randomly.
Yeah.
But the for you tab is something right now. It's something.
It's a
It it's just the bizarro timeline I didn't know we were headed towards, I guess.
It's so weird too. And then, like, the president of Argentina did a crypto rug pull, like, last week.
What is going on?
Yeah. And then
it's like then Elon is like, we're gonna
give every citizen $5,000. I'm like, what the what?
It's just the whole thing is a little too much for me. The whole Elon and Trump partnership is just it's just not heading in good places. And I don't really have strong pin opinions about those people. It just feels like too much for the world. It's just it's too much.
I think the way the one little bright spot I'm seeing is that I had a lot of people that I, like, generally looked up to and was like, oh, these are people that, you know, I like like the way they view things and, you know, I can learn from them. Mhmm. Then I witnessed how whatever this thing is that's happening where, like, there's, like, this crazy political effectively, like, virus that exists out
there. Mhmm.
I see it. I see these people I admire coming in contact with it and then just not being able to get past it. Like, I remember, like, I remember during this happened in, like, the other way. So in, during COVID, I remember there were people that like, COVID just broke them. Like, they just they were just, like, so scared of doing anything that was, like, gonna, like, get them sick.
And, like, they just couldn't escape it, and they, like, never really recovered even after COVID. Mhmm. And I'm passing the same
thing. Cases.
Oh, sorry. Yeah. Yeah. Yeah.
Many such cases. Yeah.
I see the same thing with politics where I'm like, this is, like, the most potent drug ever created, and it is, like, messing people up where they just cannot look at anything outside of this lens, and they obsess over every everything. And I see so many people that I'm like, they're clearly smart, they're clearly capable, they've achieved things that I haven't achieved in my life. Like, they're like, they've done more than I have.
Mhmm.
But they couldn't beat this thing. And there are people that have. Like, there are people that kinda, like, you know, I admire them and they, like, escaped it, but it just, like, narrowed the field of people that I'm like, I can look up to them. And I think that's kinda like a good thing because I kinda cleared out a bunch of noise.
Good. But also, like, my the the thing that I would always fall back on when it felt like social media just wears me out and I wanna get off of it for a while, the thing I'd fall back on is, like, most people don't feel this way. Most people are normal. But I don't know. Is it increasingly, like is it is it an adrenaline number of people that are normal?
Is this potent drug infecting more of the general population? Is it becoming like is this where we're headed? Because I feel like I've not taken the pill or whatever. I'm not I don't feel politically charged. But is it just inevitable? Are we all gonna end up am I gonna be on here ranting about how great some terrible person is? Some leader? I
mean, I I look at Elon and I'm just like, I think about what his potential is and he just got hit with this thing. And now he's just wasting his time in a bureaucracy doing all this, like, really stupid stuff and, like, completely detached from reality at this this point where, like, yeah, I'm sure he's still working on everything, but whatever mission there was to get to Mars, it's gonna happen a little bit slower now. Right? Because he's, like, derailed by this. I'm like, this is, you know, for a society that has an Elon, we don't need him here.
We need him working on Mars. He's actually extremely bad here, and he's extremely good at the other thing. And, yeah, it it's just crazy. Like, it's just it's just no matter how smart the person is, like, you just never know who's gonna be, like, especially infected by it. Mhmm.
Yeah. I wish I wish there was, like, a public chart, like, where you could just see the number of normal people and
the number of people
who have lost their lost to the ether. Yeah. So I can know, like, how dire it is. Anyway, I'll let you go lunch. Enjoy your siesta or whatever. Your Korean barbecue.
Yes. Alright. Alright. See you. Yep.