Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
The Conversation with @MartinShkreli
Recorded: Dec. 23, 2025 Duration: 1:34:10Space Recording
Short Summary
In a recent discussion, participants explored the evolving landscape of technology and economics, focusing on trends in AI, quantum computing, and the implications of reshoring manufacturing in the U.S. The conversation highlighted the challenges and opportunities presented by these shifts, emphasizing the need for strategic adaptation in a rapidly changing environment.
Full Transcription
Thank you. Thank you. Oh, testing, testing.
Does this work?
Hey, can you hear me okay?
Yes, I can.
Let me bring up one of my friends.
This is going to have an interesting discussion.
I feel like based on the things that I listed,
we're going to probably start with Quantum first, and he's
kind of enthusiastic about this as well, but he
kind of largely shares your sentiments, so it'd be interesting
to kind of have a discussion about that.
So, yeah, we'll start
with that. But yeah, it's good to have you, man.
It's awesome. Likewise. Also, thanks to
Isabel for setting this up for me.
Very nice. We're going to have this kind of be a
very small panel at the moment.
I will just have you on and Aude and myself.
So it's going to be kind of like a triangle, basically.
We're going to do the triangle thing.
So, yeah, we'll have it be like that for now.
Yeah. Aude, you can hear us?
Yo, Aude.
These haven't spaces issues, eh?
Aren't we all?
Yeah!
Yeah, it's really bad.
Okay, I can...
Yeah, okay.
There we go.
Alright, thank God.
Yeah.
Alright, excellence.
Yeah, the co-host thing didn't stick it wouldn't work yeah spaces is mega broken like if this whole thing fucking collapses we're just gonna go to
fucking discord that's what's gonna happen yeah uh i think this is like a minor product for
everybody like the whole idea of the live streaming product is not the most consequential
to anybody. So like, none of the platforms really have it. And WebRTC is pretty finicky.
Like it just doesn't like, always work.
Well, this one here is especially retarded because it's Periscope, right? So they're
like using a thing that's already been obsolete for many, many years.
Are they really?
Yes, yeah, they're using it for everything. You just look at it. Even live streams. You
ever wonder why the live streams have very similar bugs
to how this works as well?
It's because it's Periscope.
And it actually shows it to you.
It shows the source as .pscp.tv, which is Periscope,
and then it even tells you the whole thing.
That's right.
All the bugs are similar across.
And I think for quite a period of time,
a lot of the information was funneled through a similar protocol.
And that's what made all the interactions and things possible as well, like notifications, which also chunked up in the same way that spaces do.
And I don't know why, but it's actually gotten worse.
You know, like when you have a space that's running for over eight hours, there's some really interesting glitches that tend to occur.
I have no idea what those are.
I feel like the spaces team at the time tried to solve it,
but most of them are largely gone.
Like, the three people that I knew that worked on the Spaces team,
they're all fucking gone, man, so I don't know what's going on anymore.
Yeah.
I mean...
What are you going to do?
I feel like Twitter was doing a pretty good job
of holding everything together with duct tape.
And when Elon came in and he just fired everyone,
everything just started to go into shit.
Because there wasn't anyone keeping it up.
Yeah, I think his priority might be a little bit different from ours.
In that he's, you know, trying to get, you know,
his AI sort of effort more.
That's sort of the valuable thing to him.
Yeah, I think that's part of the reason
why this was actually a good investment too,
is because X is a pretty good resource
if you want to train your AI.
It's probably arguably one of the better ones,
alongside Reddit,
except the fact that people on Reddit are kind of insane.
But in terms of just format, I think X is one of the better platforms.
But there is no point in using this platform if you're just going to be running it like shit and making it so no one wants to use it. yeah and also i think the ai centric focus is definitely obvious i mean look at the
recent crossover with the military that they've had so yeah he's just mega ai maxing this is also
why the space computer thing is a massive thing for him. Because
firstly, you want to be like
the default delivery company for all
things space. And that's why you have Starship.
It's beyond the thing with just Mars.
I think Mars is like the thing where you kind of
dangle it and say, yeah, this is pretty cool.
But Starship is just going to be delivering
shit on our stuff into space. And if you're
going to be the default delivery company, that's going to be pretty
good.
And so it's AI adjacent. So it's like, hey, what are we going to deliver shit on our computers? And what will we run on those computers? Well, AI, of course.
Makes sense. So quantum, speaking of. Yeah, there's a lot of quantum brain rot going on. So I think maybe we outline what quantum actually is and what we currently actually have.
Let's start with that.
It's like, what is it actually?
What is quantum computing?
Let's start with that.
Well, I think a lot of, you know, I actually think of this a little bit differently.
So I think maybe reversing it and asking why do people care?
And I think people care because the largest company in the world is a computer company.
And that has never been the case previously.
So computer companies and semiconductors, they're not as terrestrial as like,
okay, I know what Meta is, I know what Apple is, I know what Google is.
You don't exactly use, you know, some of these, like, AS, you know,
very few people know what ASML or Applied Materials is or things like that
because they're fairly, you know, mundane kind of, like, you know,
equipment providers that...
Okay?
Adrian.
Adrian, you're putting out the Spaces music.
Why?
I have no idea.
Is it gone now?
Yes.
Oh, thank God.
I like doing it.
I'm just literally listening.
You know that sound?
Yeah, yeah, yeah.
Unfortunately, I do.
So, yeah, I my my point here is that we normally wouldn't give a crap about quantum other than but for the fact
that compute is now interesting again whereas previously i think you know there was this
idea that well who gives a shit about you know some some random semiconductor company at the end
of the day you know the biggest companies in the world were never hardware companies. And now we have that. So what do we do about it? And I think,
you know, people want the next thing, you know. So I think that sort of bears, you know, keeping
in mind just because I think there's like, quite a bit of, you know, questions as to why bother.
And I feel like, you know, quantum would be,
if NVIDIA weren't 5 trillion,
I don't think anybody would give a shit about quantum.
And I think, you know, as I thought about this myself,
you know, and had a position myself
in terms of like, you know, what do I do next?
You know, I think there was like a,
this idea that maybe, what do I do about it?
You know, maybe I'd be better off, you know, sort of,
you know, thinking about that instead of thinking about, well, why, you know, why is quantum
overvalued? Maybe, maybe I should be thinking about why is compute so interesting to people?
And I think that's kind of a better, we're going to talk about quantum a lot as you wish, but,
you know, I think that's almost a better, more productive way to think of things. Cause I think that's kind of a better, we're going to talk about quantum a lot, as you wish, but, you know, I think that's almost a better, more productive way to think of things because I think that, you know, why, you know, why is there this desire for quantum, you know, and I think that's, you know, generally the main reason.
You know, the bullshit articles that come out that's like quantum compute, this quantum computer is 50 trillion billion times faster than like a general computer than a supercomputer.
You know, you hear all these things from fucking IBM and Google.
And secondly, a lot of these things claim that, oh, we're going to be able to cure cancer because they can fold proteins more efficiently.
And we're going to be able to break
these types of encryption. So it's being sold to governments. It's being sold to medical industry.
It's being sold to the general public as if it's a good thing. And also we're going to,
it's the breakthrough, right? We're reaching a stagnation where we're kind of struggling to go
below a certain size for our transistor. So we need to go quantum because they are infinitely faster.
Meanwhile, all the statistics about, oh,
this computer is like 500 quadrillion times faster,
a complete bullshit because they're not even actually doing anything.
They're just using sampling.
Yeah. I mean, I think,
I still think it bears keeping in mind that people want to go faster.
And the question is to what end and why?
You know, I think companies like OpenAI and X and others, you know, they do want to go faster.
There is a reason to go faster.
But, you know, what is it?
You know, what is it for is another question, right?
Like, are we trying to do math models faster?
Well, quantum can't help you there.
And that is what we're trying to do faster and you know the
very notion of speed is you know we're trying to think about speed in a lot of different ways
algorithmic speed is more important than clock speed typically especially if you're saturated
in some algorithm like metmoles you really want to be able to do metmoles really quickly
you're going to want an
algorithm that can do that quickly. And quantum has no such algorithm. There's no way to do
almost anything faster on a quantum computer with algorithmic speed, right? So you need some
algorithmic speed up. And very few people understand the cores of algorithms. I mean,
I'm not exactly Mr. Algorithm myself, but I know enough to know that, you know, there are processes that are slower, there are processes that are faster, there's o-notation
and other ways to delineate which processes are faster or slower. And quantum has one major,
you know, method for huge speedups to go from exponential to polynomial, which is a huge
difference for those of you who don't know algorithms. That is the difference between something not being possible and something being possible.
And quantum has one such algorithm where you can do that, and that's Shor's algorithm
for factoring and for period finding and quantum Fourier transform, which are
very algorithms, breaking encryption.
Yeah, very, very limited scope there. Almost nobody,
almost none of the, if you've surveyed Amazon EC2 or something like that, very, very limited scope there. You know, almost nobody, almost none of the, if you surveyed Amazon EC2 or something like that, Azure, almost none of the compute needed today is related to this, right?
That's a very cursory kind of curiosity.
People like me are curious about it, but not too many people really give a crap about that.
We care about matmoles.
And there's no way to do, especially non-sparse,
dense matmoles quickly, at least without a different method. And we'll talk about that in a second, because I have something that could give us some hope here that's less,
you know, frightening maybe than what we're used to. So in any event, I think that, you know,
there tends to be this misunderstanding of what is speed anyway. And speed is, whether it's
clock speed, which is one thing, or algorithmic speed, which is an entirely different matter
altogether, I think that clock speed is something we can all understand. How fast does this thing
were? And every one of us has had a computer that's X gigahertz. Well, the quantum computers,
very few people know this, but they're extremely slow. So the Transmon superconductor computers are around a megahertz. And you and I have all
had megahertz computers a long time ago. As kids, they're not very useful or interesting.
The IonQ and laser trapped ion computers are much slower.
They're actually in the kilohertz range.
So clock speed, you're dead in the water.
There's just no way to get the clock speed, you know, any better.
The, at least at the moment, you know, maybe down the road they'll be better.
But at the moment, they're very, very slow.
So the only way you can win is algorithmic speed.
Now, how do you make an algorithm faster?
Well, this is really hard.
This is what the entire field of quantum information science is.
And I talked to the White House about this.
And they agreed that you're better off developing more quantum algorithms than you are developing quantum computers.
Because you can't do anything with a computer until you have an algorithm that gives you
that provable speed up.
If you make a machine that can exploit the laws of physics
to make some algorithm run faster,
and when I say run faster, I don't mean clock speed,
I mean number of steps.
And you could take the number of steps from say 10 to the 38th,
which is what you would need to hack a Bitcoin key,
to say 10 to the 10th, which 10, or less, 10 to the 5,
which is a really, really small number of steps that no matter what your clock speed is,
you could do that very quickly. Whereas also, no matter what your clock speed is,
you can't do the 10 to the 38 quickly. So that's what an algorithmic speed is. And
we only have one of those magic tricks. And it really is a beautiful magic trick. Peter Shore, MIT, you know, a bit of a quixotic.
Some may call him a top autist or top tist, as we say in my community.
Peter Shore, really, really neurotypical guy, you know, came up with this brilliant, beautiful
exploit of nature.
And that's why people are excited about quantum.
It's not for practical reasons.
Scientific community understands that there's nothing out there that's going to win in quantum. My girl actually says it's like the
Large Hadron Collider or CERN. You're happy it exists, but you wouldn't invest in it unless you
want shares of CERN. It would be very awkward. So I think that's a really good way to look at it.
And again, very happy to take on quantum bulls, but I've almost
never met a quantum bull that really believes in their own BS. I think they understand that,
you know, in a lot of ways, they're taking advantage of that zeitgeist that we're in an
exciting period for compute writ large. And that exciting period is kind of, you know, should be,
you know, celebrated, but also isn't really something that's going to
yield riches. I do think there are some approaches that might yield riches, but I think that, you
know, we still have a long ways to go on, on all of these fronts. So I don't know that if you want
to pause there. And I think, I think out before last year, we didn't even have real quantum
computers because they couldn't really do anything.
What they were doing was basically they were doing something called sampling, which was where they just track the state of a quantum bit.
And then they compare it to a computer, a traditional computer simulating that and comparing them and being like, look, the computer knows its own state.
The quantum computer knows its own state. The quantum computer knows its
own state better than the simulation on a traditional computer does. That's kind of
like me holding five fingers behind my back and being like, hey, how many fingers am I holding?
And the guy gets guesses. And you're like, well, I knew it 500 million times faster than you,
because I knew it instantly. And it took you like five seconds.
Right, right. No, the circuit sampling, the boson sampling is definitely a
scientific fraud basically. But even if you were trying to sort of, you know, do that, you know,
excuse that, you know, I still think you're in a, you sort of place where my favorite thought experiment is,
what would you do if you had a quantum computer right now?
And when I talk to people and they say, well, I do some calculations really fast, I say, which ones?
And almost never is there a real answer here.
And it belies this understanding of what computing really is.
And computer science people sort of know what it really is.
They have an understanding of this because that's their profession. But I think the average
Joe is no clue. And that makes it really, really tough for, you know, people to size
up whether this is a good investment or not. But I've started to come around to this idea
that maybe this is rational, that maybe if you're the average Joe and you saw NVIDIA
go up to, you know, say, you know, what is it, five trillion or
something, you know, that you don't know, and I don't know, and you don't know on the call here,
one doesn't know what the heck is the future will hold. And I think that's actually quite reasonable
perspective, because how could you know? So why not buy a little bit of this trash company,
a little bit of that trash company, and, you know, pray that,
you know, somehow that that will add up to something. And I think that's, you know, very,
very reasonable from the perspective of, you know, a novice investor. I mean, why not, right?
Why not think of it that way? And, you know, maybe in 10 years, you'll have the next NVIDIA,
and that's not long to wait. So I think that's very reasonable.
And in fact, it kind of makes me start to think, okay, well, what else should we be doing?
If that's the case, you know, why not, you know, try to, you know, make the most of this excitement
and try to build real next generation computers and ask ourselves the question, what is coming after NVIDIA?
Because it's probably not NVIDIA. The same way NVIDIA sort of, you know, surprised everyone in the, you know, sort of
Intel, Wintel sort of monopoly thought that'd go on forever, and it didn't. So where do we stand,
you know, in terms of next generation compute? What does that look like is almost a better
question. I don't think quantum is ever going to get there for obvious reasons. But, you know,
it's still, I think this whole odyssey where I've lost, you know, personally
lost millions of dollars, you know, trying to figure this all out. Um, this was actually kind
of an interesting insight for me, despite the painful loss of money. Uh, it's an insight to me
that, you know, maybe we are ready to start having this discussion about a post and video world.
Um, and I think that's kind of, you know,'s kind of exciting, because I think if that's the case,
and the world is ready to think about post-NVIDIA,
there's something for entrepreneurs to do, and there's something for investors to do,
because ultimately, I think that's what we're talking about here, is money.
Interestingly, none of the AI companies have shown any interest in quantum whatsoever. So I think that, you know, is very telling about
quantum's potential for compute, because there's so much desperation for compute at companies like
OpenAI, Anthropic, and so forth. They're willing to do ASICs, they're willing to do all kinds of
GPUs, LPUs, they're willing to invest in things like next generation, you know, energy supply just to
power them. But if they're that convinced that we're going to need that much energy, then why
the heck aren't they trying quantum? And the obvious answer is, you know, they're not interested
in quantum because it can't do anything in terms of bandwidth. It can't do anything in terms of
algorithmic efficiency. These cupids aren't stable. I mean, there's nobody serious out there
that thinks quantum can actually be a great compute substrate. Now, you know, what's after that,
I think is really interesting and important. And I think there's a lot of sort of non-Von
Neumann or other types of strategies that we can exploit to maybe get around some of these
bottlenecks that I think we didn't foresee. On the ASIC bit specifically, I think we didn't foresee.
On the ASIC bit specifically,
I think like a post-NVIDIA world will just kind of be ruled by China.
Not going to lie.
I feel like that's kind of where things are going
at the moment
because I can't really think of any European
or in general Western companies
that can kind of like do all these things.
So I think like,
it's probably we're it's
it's it's probably just going to come out of china and i think asics are going to be the future
anyways because like we're just using these things it's like bitcoin basically it's just like bitcoin
right people were using just like you know general compute to mine the stuff for ages until
eventually hey why don't we just talk and use asics instead and it makes a lot more sense
so eventually you're just going to have basically more purpose-built compute
because there is demand for it anyways.
So I wonder who's going to be doing that.
And I don't know of any Chinese company
that will be doing that,
but they're actively working on it.
And they're the ones that have a really good understanding
of hardware itself to the point that
they can create optimizations for hardware
that are better than the very optimizations made by the point that they can create optimizations for hardware that are
better than the very optimizations made by the companies that actually make the hardware
itself.
That's crazy to me.
Yeah, I mean, I think that there's still hope for America.
I think that we can probably figure out a way to get these, get new machines and, and get them in a,
you know, in a place where, you know, we may not need this Taiwan bottleneck. I'm not a geopolitical
guy, but like, obviously having the entire world, the world's economy, you know, hinge on this tiny
island next to China is probably not such a great place to be. You know, it doesn't take a political theorist to see that, I guess.
But, you know, so what can we do outside of this?
And I think that there had been, there's been, you know, a lot of rich, you know, non-Von Neumann approaches.
And it may be time to... Do you think there is enough political capital, especially with what's going on right now, to bring the Taiwanese people that are actually in the semiconductor business into the U.S.?
Because I'm seeing a lot of anger from Americans that are even against these fucking semiconductor people from coming in.
So I think just based off that i'm not super optimistic
yeah i mean i i think we don't need taiwan you know i think that america's got an ability to
innovate and that there could be a way to actually make a new kind of computer um i don't know what
that looks like,
but it could be any number of things that really could change the game
and how we pick up the computer world.
And I hope that there is an effort here that will create this idea of a new computer.
And what it looks like exactly, nobody knows,
but many ideas have been you know thrown around there and you know we'll have to see if if there is a you know if there is an avenue here
that that could make sense so you know i'm i'm excited you know that you know there are efforts
ongoing and you know potentially could be you know sort of um implemented in the in the not-too-distant future.
So I think we should probably double down on non-trad compute
because ultimately, I mean, think about where the limits are
in transistor-based compute.
I mean, they're in physical limits that are really just not possibly
traversable unless you try a new kind of, you know, new kind of compute, which,
you know, ultimately, you know, why are we going at three gigahertz a second or four gigahertz a
second? Why can't we go faster than that? And I think, you know, we have to ask.
We can go faster than that. But the problem is when you go that fast, the heat that is generated,
it makes it completely unreasonable to go that fast we have chips uh that are very special
single core that go much faster but it just doesn't make sense for the average person because
you would be spending more electricity on actually cooling your chip than you would be
yeah so let's move to you know a non you know some other metric you know some other
framework that doesn't necessarily use transistors. There are a lot
of substrates. You're right that there's a heat wall there and it's impossible to traverse. So
what are we going to do about it? Yeah, but the problem is when you're talking about moving
something to something completely different, you're talking about investing trillions of dollars and replacing infrastructure that's already existed.
And by the time your new technology manages to catch up with the old, it's already behind.
But we've done this before, right?
I mean, they had to go.
Not to this scale.
Not to this scale.
Maybe.
I mean, I think it's doable.
It's ambitious.
But I think, you know, to go to, for example, you know, I'm a fan of optics, but you can be a fan of memoristers.
There's so many different, you know, ideas in next generation computing.
What exactly to do? I don't know, of course.
I mean, for you to replace what you have right now, you don't, it doesn't need to just be better.
It needs to be better now right no doubt
about that at the same time if you're just like it's going to be better in a theoretical future
time if we invest 10 trillion dollars into it right no one's going to put that money in unless
you can well hold on hold on because look at quantum i and q's raised three three billion
dollars you know on that billion on that very premise.
That's tiny amounts.
You're talking about when you're talking about something like semiconductors.
Just on the fab, just on the fab, they're spending $40.
You're indexed to a very, and I'm not trying to fight, by the way.
You're indexing yourself to semiconductors where, you know, I assume, you know, when innovations happened in the past and there have been, you're indexing yourself to semiconductors where, you know, I, I assume, you know, when,
when innovations happened in the past and there had been, you know, major wins, you know, that's
what productivity is all about. Can we index ourselves to a new process that won't be what
it looks like right now? You know, that will look like something maybe a lot more efficient.
And you're kind of missing, I But you're kind of missing,
I think you're kind of missing the point
because even if it's more efficient,
even if it's better,
for you to come in and be like,
hey, not only are we going to make better chips now
using a completely new process,
but we're also going to need to move
the entire world to this new process.
All the software, all the hardware,
all the vendors, all the hardware, all the vendors,
all the governments.
This is going to cost trillions of dollars.
What's TPU doing?
What's that if it's a better technology?
I mean, TPU is doing just that.
Yeah, but the TPU is a specialized device.
Understood.
It's a specialized device that is connected to your CPU and GPU.
Sure.
And it's still based off our transistor technology.
It's not a completely brand new technology. And that's my point is that, you know, why, you know, with some really big new idea, there's
no reason to believe that, you know, we couldn't try for something bigger, you know, and just,
you know, as you know, as an innovator, I mean, the point that nobody's done it before
isn't, you know, that doesn't work.
You know, there's nobody ever made a laser before either.
And, you know, there are always new ways to do things and exploit the laws of physics that result in success.
You know, I think people told Elon there's no way they can launch a rocket.
You know, I think just to dismiss any possible innovation is not, you know.
But my point is the innovations of the past, the innovations of the past have usually been built on top of the things that have come before it.
Not necessarily. Look at wireless.
You know, wireless is a good example where, you know, do you remember the whole trope of like Korea and some other countries that they skipped the idea of like having fiber in the ground and, and other, they had no telecom infrastructure and they just started with wireless better off kind of never having the, the vestigial
kind of like attachments.
So I'm, I'm just saying that, you know, it's important to be open-minded, you know, there's,
there's, you know, obviously, but even a wireless that you're talking about right now, it still
relies on cables underground.
For us, yes.
Go to the towers.
Yes.
But that, that still is using the underlying technology
that I'm talking about.
Fiber optics?
No, not fiber optics specifically,
but going to the tower and the wires
that go to the tower and the towers
that then give you the 5G.
Fiber optics.
The 4G.
But my point is that we're dealing with technologies of the past that then drive us into the future when we need something new.
When China came in and they just adopted mobile phone payments.
Yes, they skipped credit cards, but that was because people dealt with credit cards first.
They were like, this is
kind of a shit system. Let's make something better. But you can see how well credit cards
have stuck around in the West. Yeah. I mean, I think that this line of argument is basically
saying, you know, you can't do technology. And I think there's always a way to do technology. And,
you know, one of the things you can do, for example, is atomize, you know, what it is you're trying to optimize. So a good example is the matmul. This is the
workhorse of today's economy, multiplying two n by n matrices, or n by m matrices, I should say.
And how do you do this faster? Well, this is an n cubed problem in computing. n cubed is,
you know, if it's a 10 by 10 matrix,
it's 10 cubed, just a thousand steps to calculate
this 10 by 10 multiplied by another 10 by 10 matrix.
Obviously, as we get huge numbers of these dimensions,
the scales pretty terribly.
If you could make a device that just did mat moles
at a linear complexity, so just 2n,
instead of n cubed, you would change the entire world of computing.
Now, you don't have to change any operating system.
You don't have to change any networking.
You don't have to change anything outside of just the matmul going faster.
We have the abstractions that wrap around that.
And I think that's the idea of quantum too, right?
Is that we're not gonna replace Windows with quantum OS.
We're gonna just do, I guess, key cracking
or whatever else these things will be good for.
And you'll have a QPU adjacent to a CPU or a GPU
whenever you need to do quantum period finding,
which will be never.
But assume you have this magic MatMol tool,
and it's this little MatMol semiconductor. Maybe it's a semiconductor, maybe it's optical,
maybe it's a memristor, maybe it's magic, whatever it is, you know, this little tool is something invented by you. And you are going to become a trillionaire, just like Jensen became
a billionaire, because he invented parallel computing, you've invented a new way to do
MatMols really fast. You know, this is the idea.
You know, it's not necessarily about…
You're talking about specifically something that's application specific, right?
Well, Matmol is one algorithm, right?
It is literally…
For AI.
And this Matmol would be for AI, right?
Well, you'd use Matmol all over the place, including in graphics, but yes, I mean,
AI is the main place. Yes.
But I was talking more about general computing.
Okay. Well, there are no bottlenecks in the rest of general computing.
We were talking about going from transistors to something else.
Right. Right. So this is that.
So if we're talking specifically about matmoles, that kind of…
Well, that's where 99% of the effort is
expended right now. That's why we are trying to build data centers to cover the errant waters of
the earth, is that we are trying to do matmoles faster. And I, for one, don't like this idea
that we're going to create nuclear reactors, that half of our cities are going to be in data centers,
that we want data centers on the moon and in space, when we can find a way to
do matmoles faster. That is the better approach here. And I think resigning ourselves to, well,
is it going to be NVIDIA that does the matmoles at n cubed? Or is it going to be the TPUs that do
it at 0.9 n cubed? When maybe we should be thinking about erasing that cube and replacing
with a squared or even no exponent.
That's what I think would solve this huge computing bottleneck.
And that's why quantum,
this is what the investors are thirsting for, right?
They're thirsting for who is going to, you know,
win this war of the long run
and who's going to look like NVIDIA did 20 years ago
and who can be that champion that solves this?
And I'm not saying I know who it'll be or what approach it'll be.
Nobody knows that.
But I think that maybe one of the ways you could do that, which, again, the data center is just sitting there doing map bowls all day, maybe get off transistors.
And this is what we did for comms.
We used to send data over electrical lines. We don't send, we used to send
data over electrical lines. We don't do it anymore because it's not that efficient. It's better to do
it over fiber optics. It's more bandwidth and it changed the whole world, right? Internet,
when it was over copper lines was pretty shitty. I'm sure you remember. And now with, you know,
this massive, you know, throughput, the internet's pretty awesome.
I can see you, I can wave hi to you, we can send each other all the music in the world
or whatever.
So I'm just merely saying, is there a way to do matmoles faster without necessarily
staying in the transistor regime?
I think everyone's exhausted, like algorithmic
advantages there. There's like a N to the 2.65. It's a galactic algorithm. It's impossible to
actually implement, but it is, you know, one way to do it. But, you know, how do we really
transform the game on mammals? That will change our economy. It'll change scale of GDP and so
and so forth. And that's kind of, I don't want to get bogged down on this, but I think that's
forth. And that's kind of, I don't want to get bogged down on this, but I think that's where,
where, you know, I'm excited. And I think many, many VCs are excited. Having said that, you know,
like, for example, we've seen X-Tropic and some other kind of like very weird, different
thermodynamic computing, optical computing is very, very hot right now as our, you know, again, just non-Von Neumann,
neuromorphic, you know,
what are the ways that we can,
you know, do matmuls faster without,
you know, getting...
I've been thinking about neuromorphic too.
Yeah, there has to be a way, right?
Like, man will win this.
I don't think...
There's all these fucking retards
that are like,
no, we can't
have neuromorphic because it's going to be too similar to human brains and you know it's going
to be an ethical issue i'm like shut the fuck up bro i i i think that's really funny because i was
just trying inventorying what people are doing out there in this space and i joked with my girl like can you can you make mammals uh uh can you do mammals in
animal brains it's kind of gnarly probably good for halloween and frankenstein you know there's
an adjacency to that actually because people have been trying to use uh human brain organoids as
compute there actually already exists a kind of solution for that yeah yeah they've been like
brain cell compute yes yeah yeah yeah. So the Swiss companies,
there's like the Swiss company that's doing this or a bunch of other startups that are trying to do
this kind of thing. But like these same people that are going, hey, we should like use the
use neurons for compute are also the same people are saying, hey, we shouldn't just like do
neuromorphic computing. I'm like, huh? Well, we're doing it electronically. It's, you know,
that's what a neural network is inspired by, right?
So it's a little bit silly to not use the architecture that God, so to speak, gave us.
But I think that the bigger question is, can you use biological matter, which is a totally insane concept?
I think yes, but I think it's exceedingly unethical and I don't think anyone should be
doing that, personally. I think that will
open up a door to creating
something that will be the
undoing of Azul for many
more reasons than just creating a weapon.
You're bastardizing nature
to an extent that is hitherto
undreamt of. It's crazy. You're going to be creating
something like that. What do you think about DNA
storage then?
Well, I don't want to jump off this just yet because I think this is really interesting
because, one, I don't think that biological matter is really that great of a substrate,
but I do think that, you know, it's worth exploring. The company FinalSpark is the one
maybe you're thinking of that actually is SAS access. Right now, you can SAS, you know,
Actually, it's SaaS access.
Right now, you can SaaS, you know, you can actually, you know, use this right now as a, you know, and rent a biological computer, which is, you know, completely insane.
And I'm going to look into some of this stuff, but, you know, so far, you know, they've gotten it able to play Pong and stuff like that.
I mean, there's some stuff it can do and most things it can't do, and it'll never be really precise, you know, so there's a lot there's a lot of like you know issues with that but I am really curious about where this will go um you
know not because I think like I said it's it's some fantastic you know way to do things but I
do think that there is a kind of you know you know our brains are obviously doing doing lots
of calculations and the question is you know can we can we learn from that? And to the extent that Hebb and-
Yeah, like furthering neuromorphic compute
by just kind of understanding how it works
through mediums like this, right?
Is that kind of the point?
Yeah, yeah, exactly.
And again-
I get that.
I mean, we drown rats all the time
and we kill like a shit ton of mice.
We also like skin a bunch of hamsters.
So it's like, okay, maybe there is a little bit
of leeway for the greater good.
Who knows? If you can grow it in asters. So it's like, okay, maybe there is a little bit of leeway for the greater good. Who knows?
If you can grow it in a vat and it's just cells, you know, I think that's probably fine.
I mean, I don't... Yeah, like how alive is it really?
That's usually the point.
How alive is it really?
Like, are you genuinely causing like actual spiritual pain to this thing?
Because like if you're doing that, then that's going to be problematic.
And I think like as soon as you reach a certain level of complexity with these kind of like neurons,
what ends up happening is that you have like like, this emergent kind of, this whole, like, structure.
And that's where the problems start beginning.
Oh, totally.
Yeah.
I think that...
This is where the demons come in, so...
Yeah, I'm with you.
I think that the...
It'd be unacceptable in every way.
But I will say, you know, there is still this question of, is it possible? And, you know, if you want
to get really Frankenstein, the idea is like, can you take like the closest biological animal
or even something that's like, you know, people use pigs and pharmaceuticals and other,
other like higher order species to, to actually get, you know, an animal to compute for you,
whatever, you know, mathematical thing you're trying to compute. And again,
not something I'm interested in, but it's definitely an interesting, you know, mental experiment of, you know, a thought
experiment, if, you know, is it even possible? What is the limit of our biological matter and
computing? Because ultimately, maybe that's where we're going, right? The first time people did
neural networks, they said, there's no way this is a good idea. Yeah, well, because they just didn't
have the computer at the time as well. It made sense.
People thought it made sense even in the 1920s.
But it's like, okay, well, we don't have the computer for this.
It's totally unrealistic.
And then we did have the computer.
It's just like, oh, wow, holy shit.
This actually works really, really well.
Yeah, yeah, exactly.
That's the story.
And I think the closer we get to maybe biology,
which is obviously we've gotten pretty close with the way we do AI right now, but, you know, getting it closer and closer and closer to a point where
it's, like, really, you know, there will be fun to
sort of see. In terms of DNA, I like DNA. I think DNA has got...
Yeah, DNA data storage makes a lot of sense. It's not really, like, alive in a sense that it has a
solar or anything like that. Maybe I'm wrong about this, but it really is an interesting thing that, like,
stores information by default.
I mean, that's literally what it's meant to do.
It's supposed to store information and carry a blueprint, effectively,
for some sort of construction.
So DNA storage makes a lot of sense.
If you can make that better than a fucking NVMe, then yeah, go for it, man.
I think that's a really good idea.
Yeah, I've been trying to read some papers on what enzymes would you use
to process information with, you know, still not clear, you know, how do you unspool that DNA?
Obviously, our body finds a way to do it and it does it in this like somewhat fault tolerant way.
If it's too fault tolerant, then you get cancer.
But, you know, it's somewhat fault tolerant.
So it'll be interesting to see how that sort of plays, because I do think there's a lot of potential there.
Although, again, this is probably longer term.
Microsoft's invested a lot in this space.
I'm curious what you two think about DNA storage.
That's what we're talking about.
Yeah.
Did we cut out?
Did we cut out?
Yeah, so did we cut out? Did we cut out? Yeah, a bit. Yes. So I guess the question with DNA storage is, is can you reliably get it to sort of give you the answers you want?
So you know, in databases, there's a lot of like, you know, what's your read time? What's
your rate? How much do you have to go through the data such that you
can actually know that what you've, you know, you can't search, for example.
There's no great search tool for this stuff yet.
So all that would have to be built out.
It's kind of to your earlier points, right?
But, you know, this doesn't exist yet.
And it would take a really, really long time for us to get rid of our cheap storage.
And tape drives are, you know, quite expensive. But the point is that you can store a lot more on like a gram yes of dna then
it's a football field of of tape versus a uh maybe a size of a grape of dna so really really big
delta there and you know for archival purposes, why not?
You know, and it lasts longer.
I mean, human, you know, DNA in general.
It lasts thousands of years.
Yeah, yeah, maybe even more.
So I think that's an exciting new next-gen tech.
And again, you know, I think the quantum bubble tells us that people are really excited about, you know, doing something different. And, you know, one, one thought experiment I tell, I asked my friends is, would you have invested in, in NVIDIA, um, 20 or so
years ago, knowing if I told you that AI would be really important and they'd be the category killer
and our whole economy would be based on it, et cetera, et cetera. But there'd be no profits for
25 years or 20 years. You just have to sit and wait. And all you would see is like flops per watt going up or something like that,
which, you know, you would understand is like you're behind Intel
and everyone will always say you'll never beat Intel.
And then eventually you beat Intel.
And then eventually this new magic new sort of thing comes about.
That's almost exactly what quantum is right now, right?
Like that, that is what, you know,
is largely being, you know,
sort of thrown around is like,
you're behind, don't know when,
but eventually you'll be ahead.
And we don't know when,
but eventually there'll be an application.
But the difference is that when NVIDIA came out,
there was a ton of companies
that were working on the same issue.
Sure.
Right.
So you have to pick the correct one.
Yes, yes. that's what makes the
quantum stuff even more flabbergasting because it's like well okay maybe one of these guys will
be worth something one day but how do you know it's and when nvidia first came out as well they
were behind everyone else yeah i have to ask my father-in-law who believe it or not my father-in-law who, believe it or not, my father-in-law worked there for 20 years and has a lot of fun old stories about NVIDIA.
I watched it from Jensen Huang himself.
He's like, when we first started, everyone picked the right technology except us.
So he was talking about how they were actually uh behind all the other graphics companies
would your uh what your father-in-law is a father-in-law right did it did i guess
would your father-in-law be comfortable with coming on an x-base or something like that to
talk about uh you know kind of these origins because that would be really interesting i'm
sure he would but ultimately i think that you know his position on it would be you know i can
tell you all the old war stories but at the end of the day you know a lot of this stuff, you know, his position on it would be, you know, I can tell you all the old war stories, but at the end of the day, you know, a lot of this stuff is, you know, kind of, you know,
now gone in history books, you know, and maybe what's more interesting is what's coming up.
And I think, you know, the early days of NVIDIA, how it all happened, it's been somewhat chronicled.
You know, I think the, you know,
it's a lot more interesting to see where we're going from here
because I think one of the cool things about matmoles,
and this is what I'm explicitly working on, you know,
literally at the moment, is if you could do a matmol faster,
let's say, like I said, you could do it in linear time
instead of exponential time,
you really would change just everything about
the world.
And so if you just zoom in on this one equation, that's all you need to solve.
And it sounds easy, but it's not, obviously.
The reason I'm saying this is because if you look at the reasoning paradigm, you know,
this sort of recurrent reasoning, you know, kind of almost a loop that many discovered
before even opening AI operationalized in 03 but
the sort of chain of thought and reasoning models that are literally maybe like uh elo wise elo a
thousand points higher and maybe iq wise i don't know i would say 20 or 30 iq points higher than
the straight old school gpt2 read through to the final layer kind of model,
where you just have, you know, whatever the output layer spits out, that's what you get.
That huge improvement is based on just feedbacking into the same series of matmoles. So what stops
intelligence possibly from growing to superintelligence is just how fast can you do this damn
multiplication. And if nature has converged on the neural network as our supreme processing,
you know, sort of architecture, which at least it seems from our brains that that's the case.
Now, could there be something better? Sure. Airplanes are better than wings.
You know, so maybe there is something even better out there. But ultimately, if we could get this
part really, really fast, you know, when you ask ChatGPT or Anthropic or pick your favorite one,
what, you know, to answer a really tough question, it'll sit there and think for five minutes.
In those five minutes, it's doing trillions of matmoles or billions, maybe trillions.
And the question is, you know, what if I asked it to do, you know, quintillions or,
or add, you know, add many, many more orders of magnitude?
What could come out?
Could Fermat's last theorem equivalent type work come out?
We saw recently an ex-DeepMind guy kind of, I don't know what you want to call it, but he tried to present a Navier-Stokes solution largely seemingly through AI.
Maybe things like that will actually happen in the near future.
New materials, new pharmaceuticals,
things like that.
So it really does all rest on this.
And transistors are not done by any means.
I think it's just a question.
But that's also why we're talking about
neuromorphic computing, right?
Because also one of the issues with,
okay, you're not running trillions of iterations
in your head right now.
You're physically incapable of doing exactly that. yet you can come up with more complex things than the the chat gpt with all of the internet's like archive days like you know chat gpt didn't
come up with like tsmc right for instance none of these things existed beforehand like all of
these things were built created and thought of by humans right or like probably ever sent this who
knows right but still it came from the brain somehow.
And so maybe the whole thing
is not just about making things faster,
but figuring out a way how networking actually works
by saying, hey, we have like, even in the brain,
you have like areas of the brain
that are designed for specific purposes.
And they all like work together very, very well.
So they're very well networked,
which is also interesting
because if you have a person
who's like in a definitionally dumber, right?
They have a bad network, right, their brain structure is bad.
So, like, if you look at, like, for instance, autistic people are really good with, like,
you know, computation because they kind of are like computers in a sense. They're like,
I have a few bits missing, and so they basically networked a whole bunch of other things together
and so they could just, like, you know, go band for band with, like, the most insane bullshit and
have, like, insane context windows and just, like, you know, do band for band with like the most insane bullshit and have like insane context windows and just like, you know, do what we're doing right now.
So, well, like, you know, higher functioning autism, not the one that's like retarded.
It's the whole spectrum needs to be known as Asperger's.
Thought I'd need to clarify that.
But yeah, I think it's like, it's not just about speed or like scale of the thing, but
more so about like how efficiently you can make that happen.
Like what's the networking solution?
What's the organization?
What asks the systems to do their jobs and how does it
promulgate or like come up with those questions? And there's a great essay in the like less wrong
rationalist, you know, sort of community, which I abhor, but you know, sometimes they have some
good ideas. And one of them is, you know, is the internet conscious? And are corporations the form of superintelligence?
Because if you think about the scale of effort and organization it takes,
the capitalist sort of gradient results in some of these inventions happening.
You know, to do this, the promise of money was one of the key gradients
that sort of allowed some of these things to, you know, succeed.
Well, it's kind of like a company is kind of like networking,
if you think about it.
The most effective company just does like,
if you imagine it like a brain,
you could say that like different departments
and people within it,
like, okay, so departments are areas of the brain
and then people within that department are neurons.
And then basically all of these neurons need nutrition.
So if they do a good job, they get more nutrition.
If they don't, they get replaced by something else, right?
Like, you know, it's kind of,
it's similar to how the brain works although that you have like damaged neurons
that get like replaced that's why when you drink alcohol parts of your brain actually die a little
bit so it's like things like that basically this is the jeffrey it's quite interesting it's quite
interesting that when you look at a lot of these uh a lot of these public companies they're actually
very very badly run yes but they still somehow still somehow work. Right, right. Because their brain structure is incredible, right?
You got like so many good neurons down there.
A lot of times they're abusing moats.
And, you know, that's sort of the Buffett-esque, you know, approach of like, it doesn't matter
how dumb the CEOs are.
If Coca-Cola has, you know, taken the world by storm, it's going to take the world by
storm and it'll never change.
Even somebody trying to literally damage Coca-Cola couldn't do it.
It's that good. And I
was going to mention that some of what you're talking about is the Jeffrey Epstein best buddy,
Marvin Minsky's old theories on brain as well. I feel bad for all the MIT guys that got caught up
in taking Epstein's money, literally all of them, Chomsky and all their hardware guys too,
Summers. It's just an incredible thing how he infiltrated
not just the wealthy world, but also the science world. But in any event...
Guys, we're just there for snorkeling. We promise.
It reminds me of Minsky at least. But I think that your organization of it is really important.
And the question is, can you have many AIs trained on many different corpuses?
And the capitalist AI is kind of still the ultimate gradient winner because it'll allocate compute to where it's most efficient.
And I think there's some reason to believe that.
And it sort of dovetails with Elon's point here, which is like, could you make billions
of more humans through robotics?
And they also now have that volition and desire to maximize certain
things like capital or or whatever they're they're sort of programmed to maximize and you know i i do
see like for the first time in my life i really believe in this like singularity kind of concept
but i do think this the matmul barrier needs to get broken and that'll you know sort of be where
civilization will take
off. And right now, I think the general plan is just power through it with more of the same,
which, you know, we need an exponential step function, not, you know, just a...
Yeah, right. But also at the moment, you have like this current problem where people see and
identify a singularity and it's objectively true, but it's not the singularity that was initially
sold to us. Like the initial singularity was like the definition that you'd have like data and you
have like these computation systems that would like basically come
up with new types of things all the time. And it would just become this like fully recursive loop
of infinite improvement. You don't really see that at the moment. But what you do see is the
singularity of energy and resources. That's all you're seeing right now. Like, you know, there's
like a ton more data centers. There's a ton more energy being like allocated to these data centers
is getting so bad that even the hippies are being cast to the wayside and people going hey guys i think we
need nuclear power again like maybe we don't need to concern ourselves with the thing that
physically cannot turn into a fucking nuke and maybe just like use it oh by the way we also have
ways to use nuclear waste now so how about we just fucking do that immediately before somebody
else takes our lunch you know and this is like we're debating about who's gonna get the cookie
here it's like oh fuck it we'll just figure that one out later. Let's just capture it.
You know, let's capture it for ourselves.
We can like figure all the bits and pieces out later.
That's kind of like what's going on right now.
This is what explains quantum, you know, the quantum sort of bull run is, you know, is illogical as it is.
And you can wax on, you know, how dumb it is till the cows come home.
The share prices, I mean, I think I've told every single financial institution on Wall
Street about what I think about quantum, and it has not changed a bit because at the end
of the day, the major question is, you know, what's out there that can solve this problem?
This problem is huge.
And even if there's a chance that we come up with a new quantum algorithm or we, you
know, a lot of people say, even Peter Shore said, there are cases of algorithms that were not able to be discovered until the computer was available to test them on. And, you know,
while I disagree with that a little bit, I think it's probably true that if you had a quantum
computer you could hack around on, you could probably come up with a quantum algorithm a
little more easily and certainly test them. On that same notion, I have to ask,
what do you think about the whole Beth Jezos, like, extropic thing? What do you think about that? Yeah, so he's a friend, and I think
that, you know, you can't short a private company. So to me, when I look at private companies, I just
wish them the best. And as a friend, I 100%, it's 100x, you know, where I want him to win. So
however way, any way I could help him or whatever,
I extend myself as much as I can.
But I also acknowledge that startups are extremely risky.
So I don't know 100% how things will pan out.
I think he was a little bit surprised when,
and we should get him up here
because he's sort of an expert on all things compute.
I think he was a little surprised when Unconventional raised a $500 million round.
Is he here?
I don't think he is.
No, no, I mean in general.
Just get him up.
Yeah, yeah, yeah.
We could do like a follow-up at some point
if you'd like to do that.
We could probably do that next week
or some other day.
Depends.
There's a lot to talk about.
Compute is just one narrow thing.
But, you know he you know
unconventional raised 500 from a16z and this was the nirvana founder so nirvana was the i was in
jail at the time avidly watching this um nirvana was uh intel's neuromorphic acquisition and you
know this was a you know this is still never seen production, interestingly.
So Intel made a bunch of acquisitions, like really forward-looking acquisitions that were kind of exciting, and they were never able to productize it or tape it, even tape out
in some cases.
So, you know, he then went to, so he sold Nirvana, great success.
He started something called Mosaic ML, sold that to Databricks,
which is largely an A16Z company,
and now has unconventional AI,
which sounds like it's just analog compute.
And if you watch their pitch,
it's the same exact thing I've been saying for the last hour,
which is we need to try something else.
Let's not ship H100s or B100s or V100s or R100s into space. And it's sort of a
nutty idea, but it may be all we have. Unless someone like Unconventional, which has an eye
popping 5 billion valuation, and you might ask yourself, okay, well, INQ's 10 billion or 20
billion. Unconventional is a company that literally just started,
but it has a good CEO, and it's worth $5 billion.
It gets back to my point of like,
somebody wants this big win,
and the big win could be worth more than $5 trillion.
And if you get the big win,
it's arguable that you've ushered in a new phase of computing
that back when NVIDIA was trying to gain ground,
it was not very clear that you needed a graphics processor.
And it certainly was far from clear that AI would happen, you know, and you would need
parallel processing to do it.
And so now we kind of know what the sort of trophy is.
The trophy is really important.
And we know whoever gets it will immediately have orders
like open ai and all these other companies will immediately buy buy your chips and throw
away their nvidia chips instead of a football field maybe it's just a it's a server rack
um wouldn't that be great everyone will be happy environmentalists the the companies the users the
only people that'll be unhappy or you you know, maybe data center people and
NVIDIA people.
But that's creative destruction for you.
So the question is, you know, will it be unconventional?
Will it be BethJazos?
Will it be me?
Will it be you?
Will it be whoever?
And I think Beth is, his thing is like, it's a very, like, I don't quite understand it, but it's a really neat idea about using the randomness of nature to sort of exploit certain computational algorithms.
And again, until you could do MatMol, I'm not sure what we really need some of this for.
Because, like, we kind of know what, you know, what the ask is.
Now, one of the big asks, as you may know, is also the bandwidth between chips.
And that's increasingly gone optical. And to the extent, you know, that's sort of,
you know, getting, you know, that's a bottleneck that's being sort of relaxed.
There is some excitement there too, but I don't know. It's, it's, it's a crazy new world and,
and adapting to it and learning it from a, you know, from, from being a biopharma drug expert
and skipping into this space over the last decade or so.
It's been a lot of fun.
Damn, I couldn't unmute fast enough. Jesus Christ.
Yeah, for some reason I couldn't hear what he had,
like some of what he was saying because they kept cutting out.
Yeah, I think maybe we're ready for next topic.
I hope the recording is going to be awesome because like that's,
that's probably a good to be the only thing that's like actually definition, like high definition is going to be awesome because that's probably good to be the only thing
that's actually definition
like high definition is going to work out
I don't know
it's really gay
can you hear me?
yeah I can hear you
the host connection is the one that actually hears the most
what else did you want to
chat about?
well I think it was like after we addressed some we addressed, uh, some of the compute stuff,
and I think we can probably even do a followup on that at some point in the future. Cause you know, that's an ever growing thing. It's really interesting to talk about that. Um,
let's talk about economy. Let's talk about economy. I think like
kind of where things going, we're going to see a large correction because the countries that
normally had done all of the manufacturing no longer do it for instance and there's kind of
like this race to bring it all back and i think this is currently being done especially by america
in a way that is way too quick and therefore it will burn more bridges than you will have the
ability to build bridges in the future to kind of like, you know, ensure that dominance is retained.
What do you think about that?
Like general gist, because, you know, you have like this speak of golden age
and all these things.
But when I look at it quantitatively,
I can't really figure out how to put that together
in a way that it makes sense.
So what do you think?
I guess from the AI perspective,
I think the economic benefits will take decades
because old business processes stay around for much longer than we give them credit for.
But I think that, you know, and there's a J curve, like you lose money implementing some
of the early efforts. And then when you figure out how to really use these systems, you get them.
So I think it's really tough for presidents and people like that, like lay claim to too many
successes because, you know, they're because they're decade-long planning things.
That's why Singapore kind of had this big success.
I think we were talking about that the other day that if you can plan a country's outlook for 20 years and not have somebody tear it up four years later, it sure would be nice.
And it's especially unproductive to literally, you know, have it switch every four
years. You know, that feels like, you know, really, really frustrating. That's one of the
arguments of the Singaporean government. It's that like a lot of people are like, a lot of people
say it's not really a democracy because the same party keeps winning every fucking year.
Every fucking time there's an election, not every year, but like the Singaporeans argue that that just means that the party is actually pretty good and effective at doing their job and they can actually plan for longer.
It's not a sign that there is no democracy.
It's a sign that the government, the party is actually doing a good job.
Yeah, I mean, it's kind of nuts to just tear up the playbook every four years. I mean, imagine in four years we see, you know, kind of the, I don't know, anthropic people in the White House.
Yeah, I mean, it's kind of nuts to just tear up the playbook every four years.
And people like that, they say, well, it's all about AI safety now and AI is going to be regulated. And then four years later, we're more, I don't know, techno-optimist people go into the White House and, you know, they say,
tit to hell with these AI regulations. Everyone should be able to do whatever they want with AI.
And you keep changing. It'll be, you know, four years. You're just going to end up with stasis
that, you know, lets other people, rightly or wrongly, exceed you, whether the right answer
was A or B. The worst answer is neither. Pick a direction and go. And I think that's, you know,
kind of dangerous. And the same thing
applies to tariffs and things like that. So America's kind of cooked in some of these ways,
unless we really pick on a direction and go. At the same time, we have really unique frameworks
for capital formation and entrepreneurship that are obviously kind of unparalleled.
So hopefully, we don't mess with that engine. But I kind of think we are in a golden age in some ways.
And Elon's $750 billion is a good sign of that, that we're starting to get into this
like post-money world, which is he starts to bring this up now every time he's asked
that, you know, none of this is going to make a difference.
At some point, we're not going to have universal basic income.
We have universal high income.
And that all the things that we need to live are going to be provided for. And I really
do believe that. And to the extent you call that a golden age, I'd say, sure, that is
a golden age.
Okay, but here's the problem though. Who defines… Wait, go ahead, Al.
Yeah. Wait, so you actually believe in universal basic income?
I believe in the singularity.
In terms of UBI, I think that sounds like inflation to me, unless you have this economic engine that's so profound that it can sort of create something out of nothing.
And we're not there yet.
But I think, again, I'm not a geopolitics person.
I'm not even a, you know, sort of economic social theory person.
But I do think that to the extent that you have somebody worth $750 billion and growing, it does.
And then you have, you know, a welfare state that has largely succeeded.
You know, we can exist in this weird sort of UBI-like. We're almost already in
a UBI state. What welfare state are you talking about? America. How is America's welfare state
something that succeeded? We're all functioning. There's very little poverty. In many ways,
it's a hybrid corporate government state. And again, this is not like some high conviction thing for me.
Like I said, I'm not a social scientist.
But we barely have to work in America anymore and we can make ends meet.
It's really an incredible shift.
I think YouTube, TikTok, Facebook, and other companies that provide, quote unquote, influencer, what we're doing right now, you can get by on a very minimal amount of work via those
companies and then also via productivity improvements and then finally via government handouts that
are provided for by people like Elon or anyone that's called to pay taxes.
And it seems to have large work.
Right now, that system doesn't really work.
Most people are barely functioning. They can barely afford food. They don't have enough money for housing. They don't own their own home.
Where do you live?
Right now, I'm in the U.S.
Where?
Well, I'm not going to disclose that.
And this doesn't apply to me specifically.
I'm just talking about the average person, which is part of the reason why Trump got elected in the first place,
because there was all these people that were pissed off about the fact that they couldn't afford anything under Biden.
Yeah, I don't I don't see the same America you do.
We have record low unemployment, record high wealth.
Our GDP has grown dramatically.
I mean, those those numbers are all very misleading.
Firstly, the GDP is only increasing because of AI.
Everything other than AI is decreasing.
Secondly, the unemployment rate is extremely misleading
because the unemployment rate only counts people
that are actively looking for work.
Well, we have many unemployment rates.
And actively looking for work.
There's a lot of unemployment.
They're making a minimum amount of money that's way below the minimum wage.
So if you're making like $100 a week, that excludes you from unemployed.
Well, there's all definitions of unemployment.
You have you one, you two, you three, four, you five, you six.
So you can look at any of those.
Yeah, there, six. You know, so you can look at any of those. And they're all... Yeah, there's six, but it's close to, I think, 4% now, right?
And that's not record law.
I mean, that's amazing.
It's close to record law.
I think record law is like three and a half or something.
But you're not counting the fact that a lot of people have just given up on finding more.
Well, then you have to look at the other unemployment metrics, and those are also close to a record low too. I mean, just because we can't
have 100% full employment, which I think even Keynes would say that it's not possible,
doesn't mean America is not doing well. Again, our GDP is the highest. We still have the most
economic liberty. I don't know. I think America's doing fine. It's hard to
argue. If we're not doing fine, who is? You know, it's hard to imagine, you know, given our
huge stock market wealth and all this stuff. You could argue it's a plutocracy, but at the end of
the day, I think, you know, our needs are met. You know, Trump did get elected, you know, for a lot
of reasons. It's not 100% clear that it was only, you know, because there's a dissipating middle class or something like that. And again, I sort of pivot back to Elon, which is, you know, his comments, which are when people say, God, you're so disgustingly rich. Isn't that terrible? His response is, we're going to have money's basically meaningless.
Money is basically meaningless. We're going to have so much plenty that there's going to be basically nothing to want for and that this is our future.
And that whether I have a trillion or quadrillion000 per fictional token, it sort of shows you that
maybe we are in a post-money world, or maybe just we have like a huge barbell of like massive
wealth on one end and nothing on the other. But there's clearly something strange going on with
the nature of money, given, you know, kind of some of these interesting signs we're seeing.
Well, there's a bunch of issues, right? First of all,
you'll have the issue that if money is infinite, it's worthless. And first of all, the idea of
having an age of abundance would mean that you get resources from somewhere, but eventually
these resources are going to run out. That's another little bit. I don't think we'll ever
live in a world where such a thing can actually exist. I think the US is a really good example
for many things.
As a welfare state, I feel like it kind of has succeeded,
but it's also kind of tearing itself apart because of that exact reason.
It's a thing that's going to last for a bit until the current generation goes, hey, why don't I have enough money?
And it just decides to not contribute to the pot of money anymore
that's keeping all the people who are useless alive.
The welfare state is something that's spent $40 trillion in the last decades, and it has
had no effectiveness in reducing the poverty of people.
Well, our poverty is pretty low, but I would go to, going back to what you said, Adrian,
the interesting thing there is that the current generation has moved on to a new currency,
very interestingly. They moved on to attention as the currency.
I think attention was always the currency, to be honest.
It was just that we didn't have the technology that would allow us to maximize it at the time.
That's what I would think.
Let's say the 1980s was pretty close to the information age, but, you know, people had magazines,
and they did, to a certain extent, have, like, phone connection,
and you still had social status things, right?
Like, for instance, you'd meet up physically in a place that you'd call a spot, right?
That's how they used to call it.
You meet up there, you do a whole thing,
show off each other's trinkets or whatever the fuck you got going on,
and then probably, like, wreck someone's car.
Like, that's kind of how the gameplay loop goes.
Today, we do it differently. Today today we just meet up in simulations we have virtual spaces like
these for instance as well discord is like very popular amongst the youth now because of exactly
this it's like a it's like a place mental physical place that you know you go to to meet other people
with similar interests right so it's like attention is more valuable now not because
like it's a new thing.
It's just because technology has enabled it to the extent that it is one of the optimal ways.
Individuals now have access to it.
Before, attention was largely controlled by governments, right?
And major organizations made by people that controlled those governments.
Hollywood news stations, so forth.
Exactly.
If you ask kids who are in their teens today who they admire the most,
it's not Elon or Palmer Luckey or even a movie star.
It's, what's his name, Adrian Ross?
Aiden Ross?
Aiden Ross.
Yeah, Aiden Ross and what is it, Kaisenaz and the fucking guy who barks around like a dog.
What is his name?
Aisho Speed.
All these guys, and that's the hero now.
It's because they have the most attention.
They have the most riz.
They have the most motion, as some of these kids say.
It's fantastic, you know, that that's their priority because I think it shifts the point that, you know, they're being provided for.
And, again, I think we're asking a really important question here.
Our colleague here is asking, are we sure that that's like an infinite plentiful good? Because this is all bull market talk, right? Like there is more attention out there than there is money. The actual point of attention is turning that into something that is useful. Like if you can
get attention, that's pretty good. But if you can't turn that into something useful,
it's effectively worthless, right? Like attention on its own isn't inherently valuable.
What is valuable is what can be done with it once you have it.
But what if this is all BS and that, you know, debt markets come home to roost and we, you know,
all this welfare has to be snipped away and, you know, we get back to, you know, debt markets come home to roost and we, you know, all this welfare has to be snipped away
and, you know, we get back to, you know, a recession where we're reminded of this, you know,
this fantasy that we see in the bull markets at the top where we're thinking about quantum
computing and covering the world of data centers in space and all that stuff just pops, you know,
some needle comes out and pops. Well, if you want to go down that path, let's discuss the eerie parallels
that we have with the Roaring Twenties, right?
Like, I think that by 2030,
we're going to see one of the greatest recessions
that's ever recessioned.
You'd like, you'd take the most lame
and dipshitting way to explain it.
That would be one of the things that I think is happening.
You won't have time to follow your favorite influencer
because you'll be in a coal mine somewhere.
Yeah, and I think that's actually kind of a good thing because at the moment, like, we kind of live
in an age of abundance as it is, right? Like, the thing that I always say when people say,
hey, the economy is pretty bad, I always say that's like...
It's an age of artificial abundance.
Right.
Yeah, artificial abundance.
Yeah, artificial abundance.
To his point, have we just borrowed it from the future and that, you know, we're going to have to pay the bill someday?
Maybe.
Maybe a little bit. But like, you know, at the moment,
the economy isn't bad enough.
Yeah, yeah.
The economy isn't bad enough yet.
Like, when you still see,
like, people always say,
okay, the recession is coming, like, tomorrow.
I'm like, no, no, no.
Because the fucking OnlyFans girl
who's providing zero fucking value
and is probably AI generated
is generating more money
than a guy who's, like, building
the next digital brain
that will be controlling all assets
of information in the future. So it's like, uh, when I say welfare, I include stuff like that.
That's like a proto welfare, a new welfare, if you will, and you welfare, that's like this new
form of corporatized, you know, sharing of attention. Also, I think welfare has been
wildly inflated as well at the cost of the working person.
Like, if you can get disability payment for fucking anxiety, I'm sorry.
That's just too much.
I have anxiety.
I'm sorry.
I'm offended by this.
Like, I mean, it's like when I first started going on a space, the first time I was on a thousand listener space, I was super fucking anxious. But you
power through it, it's like, okay, well, do I get
a fucking disability payment or do I power through it?
No, that's not anxiety.
That's not exactly anxiety.
Well, yeah,
in a sense. But that feeling, you can beat it.
It's been the same with other things. Human interactions.
It can be tricky.
It comes out of different forms.
Some people have a medical version of that that's really ugly and there's not much.
Yeah, that's also true.
And so, you know, whatever it is that's debilitating, I agree with you that, you know, our society, look, that wouldn't exist, you know, 100 years ago.
Right. You know, the nobody's that coddled, you know, nobody was that coddled.
So I'm with you on that. But yeah, that just... Yeah, like there's bad cases, of course.
But I think that the majority of the cases that people think is like,
okay, this is something that needs a payment is like not even...
Right, right, right.
Should even be on the scale.
A great example that bothers me a little bit is the idea of a trauma dog or what have you,
an animal that you...
An emotional support animal?
Yes, yes, yes.
That sort of...
Hey guys, remind me.
I'm just going to go on this plane, and this is my emotional support python.
It may or may not eat your dog, but that's fine.
Yeah, I think that's a little bit silly.
So we have those kinds of things, and you're right.
Then, you know, welfare of, you know, even things like, you know, this is maybe a little too...
I don't want to be politically incorrect, but, like, there are other things, like, you know, that we've just, as a society society have said, yeah, maybe that's a good idea that now we're starting to say, okay, this
is a little crazy. This stuff is nonsense. You know, let's get back to a more trad way of viewing.
And I think we're going to see something like that with the UBI thing. I think we're going to
see something like that with this as well. Because like, okay, so let's say the basics of life are
provided for you. So what are the basics?
Like, what kind of, let's start at nutrition, right?
So for a long period of time,
you're defined, like,
you're defined as standard of nutrition by the FDA
is such that it basically says,
here's a thing that you can eat at this amount
and it may not immediately kill you type of shit, right?
So if you then have this expanded onto everything else
and you say, okay, now you get your, like, daily rations, but your rations are like subpar as fuck.
And you have like the inverse food pyramid that's being shilled today where you say, hey, we need to eat 80% proteins, sorry, 80% carbohydrates and 20% proteins as opposed to the other way around.
It's like, that's what's going to fall.
That's what's going to become of us at some point.
become a bus at some point, right?
Right.
If you have like people provide this risk.
On the food, the FDA comes in and they're like,
hey, this amount per day is okay.
So every single food item that you eat
starts fucking including that shit at that level.
And then you get like 10 times the amount
that you were supposed to have according to the FDA
because every single food item fucking includes it.
Yeah, it's like daily intake value, right can see the percentage of percentages of daily intake value like for instance you can have like a what is a 200 milliliter bottle of juice right and in
it is like here's the daily intake value of like sugar it is 120 percent of that right like how the
fuck are you going to be drinking the whole thing without exceeding the daily intake value and then
of course you know becoming inherently unhealthy, right? Like these, there's
like then a statistics game that's going to be played. Plus the quality of the food is another
thing. Cause like, okay, we can make super cheap food, but you know, as the old saying goes, you
get what you pay for, right? So the bread that you're going to be getting, for instance, comes
from the most glyphosated field on the fucking planet. You're going to be eating that and you're
going to start breaking out in hives,
and you're going to have genetic deformities
and all these kinds of things.
That's what's going to end up happening
when you have these systems of government
provide nutrition for you on your behalf.
It's like the essentials of life.
Okay, well, we have government housing.
Well, okay, that's a little bit...
It's still a house made out of paper,
but I didn't have to pay for it, right?
But when it comes to food,
these are like you live with real- life consequences where eventually it just becomes unacceptable. And I
think that the whole UBI thing is like, it doesn't take into account the fact that we have thermodynamics
and that effectively energy will run out. And who's going to pay for that? Who's going to do
the work? And I have seen what happens when people do nothing and get paid to do nothing.
Like what comes out the other end is like a set of individuals who have an animosity for productivity will actually drag down the collective consciousness of society through their effectively uselessness.
Right. And that's kind of like where I see this going.
It's a very social thing. Maybe I'm like too pessimistic about this, but that's kind of like how I see this going. It's a very social thing. Maybe I'm too pessimistic about this, but that's kind of how I see these things.
I've seen what employment does to a person,
but I've also seen what unemployment paid for does to a person.
It's way worse.
Hang on, did all that go through,
or did we have some sort of issue?
I heard you.
Martin, you still there?
Yes, sir.
Just don't have a lot to add.
Okay, yeah.
Yeah, I get that because I feel like
you're coming from the side where you kind of like
this has to be a thing.
I get this and this is a very unpopular
opinion at the moment that I'm putting out there. I kind of like, this has to be a thing. I get this, and this is a very unpopular opinion at the moment
that I'm putting out there. I kind of risk
myself in doing so as well, because it's like
saying, hey, NVIDIA is going to have some competition in the
future, like, in
2024, 2023, where
everybody's like, oh, if you say anything against NVIDIA,
we're going to absolutely flatline
you guys.
Like, it's, you know, it's kind of like that.
I didn't say anything back then, so I'm going to say things now, right?
Like, that's how this works.
Although I did say a few things, and then, yeah, yeah. all right anything else i think about we kind of talked about the economy
about the economy, the economy
the economy bits i think for a bit um
bits, I think, for a bit.
Manufacturing. Let's kind of get
into that so we can kind of round that out a little bit
and then we can probably get into more policy
adjacent things, I think. And I don't mean
like politics because that's just memetics
and we could discuss that objectively as well
but I don't think it's inherently productive at the moment
because like it's a very charged topic
beyond even the thing that I've just discussed, which is like
very quantitative in nature. It's still kind of a touchy subject because it's like pseudo-political
manufacturing also has kind of become that in a way but so how exactly is america going to retain
or like actually return to manufacturing dominance and what is it going to dominate
on specifically because i think maybe
it's just going to be weapons manufacturing right so is that the thing that they're going to be
maximizing as much as possible or will it be something else or both like that's that's kind
of what i'm wondering what do you think about that it's a good question i mean it's like you
know do we do something are we going to do things that we're not good at?
Or do we let others sort of do them?
And I feel like there's definitely a lot of questions here as to what...
When I look at the military-industrial complex,
a lot of people have this weird brain rot
where they look at the military-industrial complex
and they say this whole thing needs to disappear.
I'm like, actually, no, hang on.
If we take this money, if we take that money
and we put it into something else,
like, I mean, that's a lot
of money and there's a lot of things that can be done with that money.
The military is absolutely
necessary. You need to defend
your borders and also the military
is necessary for ensuring that
you actually have trade between countries.
But,
yeah, there is excessive waste and fraud in the US military.
That's why I think like companies like Andrel are actually an improvement over like a lot of the contracts that you have right now.
Where it's like cost plus where any, any.
The government just pays an unlimited amount of money and if it goes over budget that's fine
yeah
there's also an issue with
people being expensive in the US
yeah that's a huge issue right like we have a in the US?
Yeah, that's a huge issue, right? Like, we have a great, you know, sort of system.
With China, obviously it's sort of fraught
with long-term issues.
You know, will we really be able to, you know,
keep this going over huge periods of time?
I think that's kind of the ultimate fear, right?
Yeah, the Chinese
are also quite clever in that they just take
everything.
Right.
They're like an actual empire that
doesn't really give a shit about whether
or not people think it's an empire. They just go like, hey,
we can make everything, but like cheaper and better and faster.
Here, just take it.
You know, that's like their asymmetric form of warfare,
because they knew they couldn't win militarily.
So they were like, what can we do other than that?
You can't really win culturally all that much either.
You have to do that within your own borders.
So how do you actually conquer the world?
Well, you just make everything, right?
Yeah, you make the entire world reliant on you.
You make the entire world reliant on you.
I mean, look, how many operations are there actually, like that the United world reliant on you. within it that don't want to be imperialist, but ideally, if it wants to retain its dominance, it has to become an empire, has to be an empire, it has to acknowledge the fact that it is an act as such, because China is doing exactly this. They're going to all of these places, they're
doing the strange business deals, and they're coming away with all the resources, and they're
just selling it back to everybody else. So they're like, hey, I have all the stuff. You're sitting on
a shiny, like, mountain of gold. If you have all the gold and you just give it out, then it's like,
okay, well, you just control everything, including the value of gold.
Right. So America kind of has to do that. And I don't really see it doing that at the moment.
That's the main issue, I years, which you can see with deals like XAI, End that might actually save a large portion of their dominance, if they maximize that, I think.
I mean, hey, you can make all of the resources, but can you make all of the weapons? Your resources are useless without weapons, right?
Yeah, it'll be interesting to see how this plays out, because in a lot of ways, some of the earlier criticism, this would be very useful to have a kind of a,
you know, this kind of benefit of, you know, manufacturing that is, you know, in America.
It sort of solves a lot of problems having this middle class.
But what's middle class when a dude has $700 billion?
It's sort of, you know, there's people that even aren't impressed by being a billionaire anymore.
We've gotten to a point where, you know, there's so much excess in one dimension that, you know,
dimension that, you know, it's hard to, like I said earlier, sort of determine what does
it's hard to, like I said earlier, sort of determine what does this mean?
this mean?
You know, it's starting to lose a little bit of meaning when the amount, orders of magnitude
somebody might have in wealth is just so much more than some others that, you know, where
are we going with it?
But I do think a manufacturing base is kind of this like somewhat strategic, you know,
certainly strategic.
I mean, it used to be that if you had a
billion dollars everyone knew who you were right and you could do whatever you want and you could
pay for whatever you want now if you have a billion dollars you're still operating on a lot
of the restrictions that average people do even even though you have less restrictions on you
than the average person but you're still more restricted than someone that has 500 billion, 100 billion.
Yeah, this is sort of the word buffet point, which is, you know,
I use the same phone as you do, I eat the same food you do,
you know, there isn't some, like, magic thing that I have that you, you know,
you can't have other than maybe a private jet and control over some assets,
which, you know, that's kind of our, you know, the main thing that, you know, you know, he's got that you can't have.
If that makes sense, I, you know, the disparity between, you know, the billionaire and the average person has shrunk and shrunk and shrunk.
And the wealth perhaps between them has grown and grown and grown, which makes you wonder, like, well, what what is the you know, what is the point of being so wealthy?
And I think some, as we mentioned earlier, some of the youth have said the same question and said, well, maybe maybe it's just attention and I can skip the wealth part because who needs it?
And it'll just come with the attention, perhaps.
In some cases, this is true, but like the value of...
I think it's less so about owning things as much as it is about being able to control things.
I think that that's kind of always been
The most important bits that people didn't understand is that it's not about stuff that is like publicly associated to you or like
Legally associated to you but stuff that you have control over
I think like I mean, why would you need money in the first place?
Well to eat and sleep in a nice place
But also to do things that allow for you to have some sort of security right the only reason why we have money is because we want to control bits and pieces of
the world so that we are secure and any part of us that we like to protect is equally as secure
and also in the future is secure if we happen to pass on which we will so it's like it allows you
to use the resources to impact the world in the way that you want.
Exactly. Which is one of the things you want to do within your life.
Your order, right? In a positive way that is.
Not everybody shares the same sentiment on that, but it's what it is.
So it's largely just a control game, and I think people are starting to realize that.
That's why sometimes attention, if it's captured in the right way, is more powerful than money, because you can derive money from it.
Yeah, sure.
But you can also move things around without ever having even given anyone money to do so,
which is, I think, the most interesting that people don't initially consider.
We were talking about it a few days ago, remember,
when we said that you couldn't pay any amount of money to get the same attention
that OpenAI got from the Ghibli thing? Exactly, yeah. that you couldn't pay any amount of money to get the same attention that, uh,
open AI got from the Ghibli thing.
Exactly.
Yeah.
We're also talking about like,
why,
why like there was this,
even,
even yesterday,
I think it was yesterday.
And it was the day before.
it was a day.
It was,
uh, the,
the day before yesterday,
it was like the,
the conversation.
It was a discussion of like,
why did open AI like allow the video generation
so our video generation at the time to be so costly to itself well basically because it was
free advertising you had all these like influencers and such that wanted to bring out a shiny new
thing because everybody wanted to interact with sora videos as you basically just like had to
give people the ability to create their own fucking content that they can then basically
just like shill to the masses and then get more customers in there for right that's like that was the whole thing it was like advertising it cost
them a whole bunch of money but technically speaking i think that they got more attention
and more users off of that than they ever cut off of an advertising campaign yeah because that was
the advertising absolutely organic advertising campaigns are the best i mean there are demonstrations
of demonstrations of this everywhere me you everybody you, everybody. The speculative value of a thing
is a lot higher if it has grown organically
than if it had been
kind of accelerated. There's a lot of people
that have been artificially
platformed
that just kind of disappear
and get destroyed within less than a year.
Hayley Welch, for instance, is a great example of this.
The weird hop to a girl, like all that bullshit.
Rug your community once, everything completely gone.
Like, the attention was so artificial that as soon as there was no market in doing that anymore,
the whole thing just basically collapsed because, you know, the bitch didn't retain anything.
So there's, like, that's the value of attention.
And also it's what attention do you have?
Like, if you have, have for instance attention of people
that can change a lot of things about the world that it's infinitely more valuable even though
it's like smaller by mass than if you had like a million retards right like an army of retards
is pretty good but it's not as useful as like a thousand people who control maybe 10 10 of the
world's you know resources for instance right even a billion retards wouldn't be as good.
Yeah.
That's kind of where we get into
policy as well, because that's just management of
people, right? We have
two sets of religions. We have real religion,
and there's fake religion, which is
law, right? Which
exists for a good purpose and largely to
manage well a very large majority of people that cannot manage themselves because they don't have
any religion anymore and that's like policy and then politics and all that kind of like goes
around in a circle and then people go well who's who i'm like well you're all technically the same
just one of you has a soul and the other one does not. You have to figure that one out on your own.
Well, guys, I have to run.
I don't know what other topics you want to discuss, if any.
I think we kind of, like,
I think we covered as much as we
could. We could probably do a follow-up in
like a week or two, whatever you're kind of down for.
Whatever's organically relevant, I
think. Maybe we could do a thing with DefJs if he's down like it depends whatever like this is generally
like the the conversation is more so like a forum where we have like this this this this trio where
we go round and round until everything's mined and then we just kind of finish off wrap it up and just
have a little little decompression beyond that point so i think we could probably wrap it up
here to be honest. Perfect.
Right then. I mean, thanks for coming. Thanks for your time. This is awesome.
Out as well. Yeah, this was a great space. We'll see when it comes up again.
Yeah, I'm going to head into Discord now because it's like the time of the day where I usually do the daily thing for my subscribers. Oh, I got to check out your Discord. Can you tell us how to get there?
Yeah, yeah, it's right up at the top. It's great. I can give you the content creator role as well. That'd be great. Yeah, I'd got to check out your Discord. Can you tell us how to get there? Yeah, it's right up at the top.
I can give you the content creator role as well.
That'd be great.
Excellent.
The link is up at the top, discord.gg forward slash noetic.
Go join that. There's a bit of a chaos
in general chat right now because a bunch of people decided
hey, let's have a cat fight, which I think is kind of funny
because I'm like, hey, this is hilarious. We're like discussing
technological things and
there's just a fight going on there in the background. We go like, oh, that's a nice little source of dopamine. I was like, hey, this is hilarious. We like discussing technological things, and there's just a fight going on there in the background.
We go like, oh, that's a nice little source of dopamine.
I was like, what the fuck?
But yeah, so join the server.
I'll end the space here right now.
Thanks a lot, Igor.
Thank you all for coming. I'm going to go to the next episode.
Insights
P
