Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
What’s the state of DeAI x DePIN?
Recorded: May 27, 2025 Duration: 0:42:40Space Recording
Short Summary
In a recent discussion, crypto enthusiasts explored the intersection of decentralized technologies and AI, emphasizing trends towards privacy, community-driven projects, and the potential for token launches. Key insights included the importance of ethical AI development and the need for innovative solutions that prioritize user engagement and decentralization.
Full Transcription
Music Thank you. Yo, my man.
I don't know if I have your mic on.
Give me one sec.
Hey.
Be on stage.
Thanks for the invite.
Yeah.
Look, just hosting a random space.
Just trying it out.
I haven't hosted one in a while.
So getting my feet wet again, just doing it random.
Dude, random is the best method.
That's where the, you know, get the best content.
So, no, I'm all in on it, man.
Love the topic.
I mean, more deep end than AI.
I think it aligns with what I'm doing, but I also use AI daily,
so the convergence is interesting.
Yeah, Tom just posted it out.
Awesome.
So, and Russ, remind me, so you're building hubs,
and so the focus, there's a D a deep end focus for you guys as well?
Right.
So in order to be decentralized, we basically run the nodes to offset the computational load.
So instead of spending all this money on AWS servers, we let our community mine for our native token down the line.
But yeah, so there's the line, but yeah.
So there's the D pin aspect.
Love it.
Mm-hmm.
Love it.
Yeah.
Yeah.
That was something, uh, when I was working for stacks or block stack back in the day,
um, I got really, uh, excited about, uh, privacy and then decentralization.
So that's sort of like been with me ever since
loving it how's hubs going right now i mean we're cruising along um i i would love to hear back from you guys but i know it's a waiting game that's what i'm saying i get i got a lot
of sticks in the fire and you guys were my favorite so i was like gonna kind of buy some time um but no we're we're moving um just trying to figure
out which uh which direction to take like i'm very i don't know like i'm looking into zion
looking into wire okay and if and and finding a home is really important to me because wherever i go i'm going
to be the the club penguin of whatever uh virtual or whatever blockchain i end up so i'm going to be
bringing them a lot of users i don't know it'd be great to like the founding team i try not to be
picky but it'd be nice to actually jive with people running the project that i'm under yeah
jive with people running the project that I'm under yeah yeah no I see fans
oh in here I was also with the art give me one second while I shoot all this
stuff out fans oh can you you able to speak
let's see and on tight invited you to speak so let me know if you can't see anything it can't hear anything
or say anything um yeah but russ no i love that club penguin uh explanation i think so i think we
can sort of kick off um i think it'd be dope to hear from you
on this. Just had a few questions in mind on DEI, D-Pen, et cetera, and just want to get a sense for
everyone's, you know, vibes, but also, you know, facts on how the space is pushing forward. So just, you know, I think you have an interesting
place where you can talk about both the AI side, but also there's just, especially with it being
communities, there's just a whole bunch of, you know, tech that you all are bringing into the
platform. So kind of want to start on. Yeah. Where see you know what would you say the state of things is
i mean there's the state of things and then there's like my stance on things um i'm not much
of a person to ride trends um but when it comes to ai i mean even with my platform um something
i would like to sort of broadcast is that we're almost like an AI safe
space because I don't want little things following me around to feed their algorithms to further
get me addicted to things I don't want to click on or look at. So anyway, I'm not trying to rabbit
trail on that one. So I have a particular stance on AI and just how I see it kind of going and unraveling in the future.
So I don't know.
But with the convergence, I mean, are you saying that with the way things are automating and then with D-PIN and having it where an AI can actually have access to a centralized bank account and things like that?
Yeah, I mean, I think there's two ways to take it.
I think one on the, like,
what's the state of the actual interaction of DEI, D-PIN?
I think that's a really interesting take.
I was also mentioning, you know, those two separately.
Just what's the state of DEI and D-PIN?
But at the interaction of them, I think one, the, you know,
obviously like using compute for agents is important.
But I'm seeing, too, there's been sort of industry trend moving away towards building infra just in general across all of crypto.
But then especially in DEI and Deepin, there's been more of a focus on the consumer side.
And so at the intersection, you've seen, you know, folks bootstrapping that.
So take like your beamables of the world and others,
folks essentially bootstrapping their network with compute.
And then like you guys,
and then overlaying a bunch of useful and interesting consumer stuff without
having to like manage your back end
or at least outsource your back end
to a high cost environment,
like in a high cost way.
So I think one,
making sure that you all vertically integrate
and focus a ton more on consumer.
That's why I'm seeing a ton of interaction.
But then I'll say two,
on the AI side, there's been a focus on,
you know, taking the next business model with agents. So you have like the kind of,
I say, Gen 1 of agents was just like agent launch pads. Now you're having these consumer apps for
agents. So Hubs, again, is a good example. I've seen others where it's like, okay,
Hubs, again, is a good example. I've seen others where it's like, OK, how are we enabling agents to stream or, you know, build environments in one prompt?
When you say street, I just let me take one point, like agents stream. Like, what do you mean by that?
What do you mean by that?
Yes, I've heard, I actually just saw a pitch today of someone building, I forget the name,
but they're building a platform for agents to, like I said, like they live stream.
So they'll have folks who will go in and watch, say, an agent perform in a game.
And you can, in the chat, you know, suggest things for the agent to do.
And you can, in the chat, you know, suggest things for the agent to do.
But go on its own, go off and do things according to, you know, the sentiment of the chat and what it's also kind of programmed to do.
I mean, I'm. Do you like that? I mean, what was your.
Yeah. Like, that's my question. Yeah, well, I think
more
I'm sort of calling out
the stuff's happening.
Some of it's memetic,
to be clear.
I think,
especially for the business model,
they were focused on...
Because it's not like an agent launch pad.
So you can't like launch an agent with a token attached,
but there's a platform,
you know,
the platform allows for purchases of the agents,
you know,
asset,
you know,
et cetera.
So if it like has t-shirts,
et cetera,
then you can earn revenue from the different business models that
the agent's able to do. Yeah. So let me, I guess an example for me that gives a little light on
my stance or sheds a little light on my stance. Like if I have an issue with a platform
or a software I'm using, let me just, I don't know, GoDaddy, for instance, I get a hold
of customer service for GoDaddy. Let's say it's the chat. If I find out that I'm just talking to
a very convincing AI, like as a consumer, that battles me. Like I really cherish the human interaction and I also believe that it it should be a human getting the value
I don't know it's just to me I mean maybe it's just a moral stance but I really I mean I'm almost
trying to shy away from the parasocial relationship that comes out of you know twitch streamers who
have 50,000 concurrent viewers and they can't even memorize one name
in that flood of messages.
Like it's just one entity called chat.
That's what it is.
And these people have an emotional attachment to them,
but it's not a value exchange.
So I see, I don't know.
I think with the agents,
you're even furthering the value exchange.
It's definitely not gonna be one-to-one.
It's one very wealthy executive.
It just makes things feel like they're less authentic.
But it seems to be what people want, which is cool.
I don't know.
It's something, yeah, sure, you can throw money at and it'll do well. But how do we really feel about it? i don't know i it just it's it's something yeah sure you can throw money at and
it'll do well but how do how do we really feel about it i don't know i'm i'm i guess the only
time will tell uh yeah it's more harmful than it is on a macro sociological level and again that's
you know that's just where i come from i'm'm trying to create a platform where we take Dunbar's
number seriously, where you can only have 150 working relationships at any given time.
And we build a platform around that and try to create community in a way that's sustainable
online, that goes beyond, you know, QD, text chat, things like that, but also doesn't replace
real world community
anyway that was a shameless plug sorry about that no dude no no feel free and like i said this will
be recorded so we'll have folks be able to uh listen and hear that great explanation of hub
so no dude um so dude i'm curious so on the ai side what makes you so what would you say the state of consumer DEI looks like?
So for you and then even say, you know, what you're seeing in the market, is there a focus on privacy for consumer interactions or just building better or more consumer interactions?
What would you say on the consumer side?
Now, I wish fans would chime in
and have something more intellectual to say i mean this is i mean i i definitely hopped in here
because this is not that one particularly is not my biggest area of expertise okay um in the the
i mean i can interpret it to my best ability yeah i mean, being decentralized AI, I mean, obviously that is a better scenario
than, you know, concentrated control. Can you give me an example of what that even looks like
in terms of real world consumer value? Yeah. So, I mean, for example, if you have like an agent chatbot ecosystem or if you're building, you know, agents and games or, you know, social media platforms that have agents interacting in them or, you know, just consumer applications where you're either using open source models or, you know, crypto or crypto technology.
So ZK proofs, et cetera, to verify or protect people's identity in consumer applications, stuff like that.
So I think for you guys, it may be more of how are you thinking about, you know, say.
Online interactions using DEI.
So for you, I think this is actually a super interesting question.
To what degree do you find that people in your ecosystem or potential, you know, prospective members of your ecosystem, et cetera, care about if you're going to use AI, we'll make sure that you're using like, you know, using,
you know, like, um, Prime Interlegs open source model versus like chat GPT or something like,
are you getting those questions or do people care that you're using?
Well, I mean, I, I've gotten, you know, VCs trying to get me to inject these, uh, AI into,
He's trying to get me to inject the AI into my platform.
And although we use it as sort of a onboarding mechanism
in a way, it's still like, I don't know,
like it's such a tough question because I keep going back
to the fact that I truly want to be an AI safe space.
So I don't want to like do a conflicting statement there. And I definitely see to be an AI safe space. So I don't want to do a conflicting statement there.
And I definitely see the value in the way that we actually have AI in world would be labeled as such.
I hate it when you feel like you've been deceived talking to an AI and it was actually or talking to a person who was actually an AI.
That's definitely something we don't want to happen.
I don't know.
You brought up privacy.
You brought up a whole lot of things there. But I'm still on the moral aspect. And let me kind of pivot a bit. The reason
why AI is even so trending, I think, especially with consumer applications and you got these
cursor AI, things that are allowing your day to day
nine to fiver to have an opportunity to break out of the matrix as that one dude, that one
bald dude would say.
I believe it was the same reason why meme tokens popped off and everything else.
And I guess it gives an opportunity for people to have a chance in their free time to actually be a contender for some innovation to
happen without having to have the right connections. So, I mean, I can, I see why
it is becoming such a thing because people are really tired of their dead end jobs that,
you know, they're forced into via this modern day slavery that we're all subjected to.
So anyway, I mean, it means it's freedom
at the core of it all, right?
And that's why it's doing so well.
Whether I agree with it from a moral stance
and then just the fact that it's replacing people
and I believe people need people and that's just where I stand.
And I would rather have a team of people that are smart than, you know, less of a team.
And then just a bunch of AI that have a great pipeline.
So I don't know, man.
I'm still kind of up in the air with it.
Can I say that?
That is my answer.
It's just like, I'm still not sure where it all settles with me just yet.
Because I think it's still so new.
I approach it cautiously.
I'm just texting fans on now.
You're good.
You're good.
Opening wasn't a too roundabout answer.
It's just that was kind of a loaded question, right?
It's like I don't think the right answer has even been discovered yet.
Yeah, you're totally fine.
Chris, welcome.
Sanzo as well.
DMed a few of you guys.
Chris, let me know.
Raise your hand if you want to join.
And happy to let you guys, let you up on stage.
We're chatting about the state of DEI and D-PEN.
I'm not sure if you, if you guys have followed Hubs yet, but have Russ in here.
Fantastic founder with fantastic thoughts about the future of both D-Pen and online communities.
I know Hubs is a bit different than focused purely on DEAI, but definitely has some great takes on it.
So glad to have everyone here.
I want to sort of keep the conversation going.
Curious for everyone.
One, yeah, this is sort of a controversial question, but on purpose.
So curious to hear people's thoughts.
Welcome, Conway.
Curious, which do we think needs the other other I want to say it this way
Which do we think has
The most important things to contribute to the world
DAI or D-PIN
And why in people's heads
I mean I'm going to
Definitely jump to D-PIN
And I mean
I think it's all about Cutting out the middlemen in the right kind
of ways. The middlemen that don't actually provide value that are, you know, that are people. And I
don't know, like to me, you know, because the way I understand D-PIN and what it pretty much allows
is similar to how Bitcoin is replacing banks, right? And so it's just an evolution.
And then we're kind of connecting into legacy systems
and we need that.
I mean, it's just part of the process.
So yeah, and then not everything needs to be on chain
and with RWAs and things like that,
it's kind of impossible depending on the amount of data
you're trying to put up there.
So yeah, I mean, in terms of liberating humanity, depending on on the amount of data you're trying to put up there so um yeah i mean that's that's
in terms of liberating uh humanity i don't know if that was exactly your term um but i think deep
in at least is the quickest or maybe not i won't even say quickest because i don't know again like
i said i think ai is still being discovered uh in terms of of how much it alleviates us. Anyway, Conway's up here now, so I'll let him chime in.
Yeah.
I don't know if I'd say that it's more important,
but I'd just say DAI is very interesting
because there's certain models that I think won't be created
if you don't start adding decentralized incentives
to getting them created. One that's really top of mind for us is POV expert knowledge models.
So this is like, imagine you've got AR glasses on and you want to learn how to do something very
esoteric thing. Currently, there's a monopoly on this type of model,
and it's like Ego4D.
Meta creates that.
It's very out of date.
All of the models that exist and all of the data sets that exist
are behind closed doors currently.
So, I mean, it'll probably happen anyways
that these start to be decentralized over time, but
crypto, I think could really accelerate that timeline. And also, I don't know if everyone
knows this, but like stable coins are already the way that most contributors to data sets are
getting rewarded. Like if you want like scale AI, like they pay all of their contributors and stable
coins. So anyways, long story short, I think that's an
underappreciated area. And it's one thing I'd like to see more in the crypto AI overlap is
funding the creation of models that don't exist. So as opposed to doing like, oh, here's a decentralized version of this existing LLM, you know, or similar capabilities to an existing LLM, like that's not as ambitious as like, let's go create something that doesn't already exist.
And so Vision AI, I think in particular, is really, really primed for that.
primed for that. I think it's awesome. And have you heard of, I forget their name.
I think it's awesome.
And have you heard of, I forget their name.
There was a team that was just founded. Oh, I forget their name. Chris saw your DM. Appreciate
it. Anyway, they're funding, they're creating a new research model or model for commercializing LLMs.
And it was called Platform Research.
I forget the name.
It was one of my favorite companies coming out, Conway.
But I appreciate that context.
I think, yes, I agree.
Also, there's not enough, I think, focus on the cost of training runs. And this is what this
company was focused on and lowering the cost of training runs. One does a few things. One,
it allows for non-transformer based, et cetera, models to have, you know, a viable path to
commercialization without just having to build, you know, kind of having the API as a tool for recouping the cost of an initial like 80 to 100 million dollar training run.
But second, you know, I think a lot of that point gets to fundamentally who do we want to be able to innovate at the cutting edge of AI?
at the cutting edge of AI? Should it only be, you know, entities with, and I don't want to ask
this as a leading question, just sort of keep the combo going, but should it only be, you know,
large R&D houses or large kind of tech companies that can afford to do AI research or to the
degree that you can, you know, use Web3 to lower the cost of kind of AI R&D again, whether that's
Web3 to lower the cost of kind of AI R&D again, whether that's crowdsourcing training runs or
distributing the inference that it takes, et cetera, and lowering the cost of it.
That all creates downstream effects that I think, unless you're in AI, it's sort of hard to
understand where the bottlenecks are. I want to sort of pause there. I'm curious if that
makes sense for everyone.
Yeah, I mean, driving down costs, definitely a big factor, but I would just highlight, so we've met with several people that are, you know, more knowledgeable than myself on the AI
side. Number one thing they all mentioned is data. Like data quality is the thing that is going to differentiate so
i think train training cost is definitely part of it and if you can lower training costs um that's
huge those though are probably going to go down anyways um data is right now the thing that is
the most centralized, I guess.
In terms of high quality data.
Once you're getting beyond what you can scrape.
And even that, you could argue that that could be decentralized more.
Desai kind of has some overlap there.
I'm not in that space, but when i hear people talk about it
it kind of rhymes yeah a similar thing anyways it feels like a low-hanging fruit i guess is what
i'm trying to say is that how do how do we get this data out there more quickly yeah i'm curious
on that point one run a group for deiI and DSI Builders and Investors.
So it has about 180 folks in it.
So know some DSI folks if you want some intros, high quality folks.
But second, on the data side, I'm curious how, I'm not sure what your background is,
I saw a little bit over your Twitter, but curious, how is that problem getting solved now, like outside of like data lakes and stuff?
Like sort of the obvious.
Very, well, either, you know, the edge that we haven't gotten to at least.
Yeah.
There's a lot of sophisticated scraping protocols and annotation and all of that kind of stuff.
But currently it's very manual, you know,
it's like you just pay people basically to, you pay people to go do these things,
you know, and that's where, you know, it's a loose take, but that's where recently I've
been thinking of my, man, what does crypto have down pat? It's like coming up with,
man what does crypto have down pat it's like coming up with do this thing and get rewarded
and do it at scale and with a lot of variation from people all around the world and so on so
can we harness that energy uh for something that's gonna have more utility you know, and maybe can still have like a memetic power. It can still be fun. Right.
But yeah, anyways, you can imagine like,
let's say we want to get a bunch of data around cooking. Right.
We have one approach is you could do the stable coin approach.
If you were going to do that, maybe you can do something like Farcaster,
right? Where you mix, you have a point system with USDC stables. And so that's another way
where you could get that, that cost down where each contribution is not necessarily getting
paid, but the top contributions are getting paid a bigger amount of money. Another one is you could
launch like a meme coin around it. Oh, we've got cooking,
you know, the tickers cook, right. You know, just soft ideas, you know.
But yeah, I'm curious. I'm really curious about what others have been thinking on that front.
Yeah, dude. Just even piggyback on that and we'll open it up. Um, on the, so
do you mean personal data is the most bottleneck? Like, you know, that, that last mile of data.
So people's private stuff is the most, not necessarily private, the more just, I mean,
personal data would be a subset of just data that doesn't really exist, right?
Like there's not a lot of it out there and that's publicly accessible.
In robotics, you'll see like a really common bottleneck is everyone's got these different approaches for scanning homes, you know, and getting like interior layouts.
So that's, you know, that's one example
of a niche. People are solving it in different ways. Some people are just paying people to like
wear Apple Vision Pros and, you know, do things inside of their house. Other people are creating,
you know, apps that have one stated use case, but actually are, you know, secretly a data collection play. Yeah, I think it's,
it's pretty, pretty varied. Love it. But I guess my thing, if you guys come across anyone, my,
my thing has been, can we somehow incentivize models that and data that doesn't exist,
as opposed to just decentralizing the stuff
that already exists.
I mean, I'm in it.
No, I just sort of subscribe to the notion
that people can get,
because it's like we're retroactively using the data
that we've already accumulated,
and then we're all fighting over these pools of data
to actually make our models function decently.
But yeah, it's not like we can retroactively reward people
for the original production of that.
Plus data brokers already sold it all away.
But I feel like the better future,
the way that I would see it
is that we're just fairly rewarded for those contributions, which I guess that's what you were alluding to.
But I don't want it just to be like everyone's just going in and scanning things.
And that's like the I don't know, I guess I come from a more creative stance.
And you got a lot of artists who are like, oh, they took my, you know, personal IP or that was my style and stuff like that and kind of seeing how that pans
out is where my curiosity lands and hoping that it's like fair uh in some sense does that make
sense yeah it does um i think for our you know i i'm just speaking mostly for i saw this come up
in my feed and i was like oh yeah i'm thinking about this a lot right now yeah we call it like duolingo
for everything so that's like one thing i'd like to see exist and i think is gonna require
yeah cracking how can you get everyone contributing um
to that's probably not one model maybe it is one model but you know how do you contribute to
that type of future where I can uh I can look at something that I'm doing and just be like hey
teach me how to do blank right um or put out a bounty even you know if the data doesn't exist
yet I think there's some people playing around with that part though you know if the data doesn't exist yet i think there's some people playing around
with that part though you know if you can get these these model creators to also more easily
be able to reward the contributors so i'll this is going to be probably the end of my constructive thoughts to it. But right now, Scale.ai is a very slow process.
One, you've got to be a really big company before Scale.ai starts to make sense.
And one thing I notice is everyone says, oh, we could really use more data on like x right and then they they kind of
they either solve that like we were saying they like brute force it in some way or they just like
stall out or they kind of like ship something mediocre using the public data sets or like what have you, I've just been thinking like, what's like, how can you
have some like genius AI person, you know, go, hey,
we need, instead of complaining, hey, we need data for X,
they turn that into a bounty.
And then they get that data very quickly
and they get it in the structure that they need.
And technically these things already exist like you've got mechanical turk and but why is that not serving
do you um you know you know sapien i'm pretty sure i do yeah sapien's pretty cool i think sapien's
like the closest that i've come across there's another one that's also in the drone
space, I think.
So, do you
know Gargoyle? Like Mike over there,
Gargoyle Systems?
I don't know. Okay.
It's sort of adjacent to that. I'm
interviewing them in a few weeks.
Nice. For the pot, but
okay, I'm checking out right now.
Yeah. Maybe I have the wrong one.
Is it security?
Yeah, security.
But it's adjacent to the crowdsource
sort of
approach to drones.
like data.
You basically have to do
crowdsourcing, I think, for any
at least in my opinion, like particularly for vision.
Yeah. You know, because you it's I mean, just going back to that cooking example, take a stove,
right? Stoves vary across every geography, times of day, know just all these um these things so you really want diversity
um in terms of the contribution um to it it's really relevant for robotics i mean i think about
it a lot from ar but but robotics is super super relevant too No, that's awesome. Look, I need to hop in like 10 minutes.
I want to kind of close it with this guy.
I was heading to a call.
Been hosting or starting to host these again
much more randomly.
So I appreciate you guys joining.
One, Russ and Colin,
I've had some great thoughts and comments from you guys.
So I appreciate you all joining.
And we'll be getting more people in as I start syndicating this.
Didn't even really tell anybody I was hosting it.
Just wanted to host it and see how the algo responds.
So I appreciate the interest in joining.
I want to close with this is something that's been on my mind and curious.
Conway just been pondering on it.
So can relate to your comments earlier. Also we'll send you the company that I, cause you'll,
you'll want to know this company. I forget the one that I mentioned earlier, but
they're one that's focused on the training run side. So I'll mention them, you know,
I'll DM them to you later, but the kind of closing thought I'm curious about is what do we think
folks from the, I think this applies for both DAI and D-Pen where we can brainstorm and have some overlap.
But what do we think folks from Web2 are like right about in their criticisms about DEI and D-Pen like as sectors, but then also are missing about DEID. But I say this sort of wasn't SF a few weeks ago.
It had a lot of debates with a lot of friends in AI
about how crypto is not even needed.
In AI, I'm still bullish, but curious,
what do we think people are getting right?
and then missing about why crypto are useful in these two areas.
And then missing about like why crypto
are useful in these two areas.
Russ, I'll give you a chance to monopolize everything.
Oh, no, man.
You definitely have some depth here that I have not ventured to yet,
so I definitely enjoy listening.
Whether it's needed, I think yes, especially when you're getting to the state
of decentralization and not letting people monopolize the infrastructure access. I don't
think it's going to be perfect. I always think that private funds are going to always 10x the other attempts to open source all those
things. They'll be sufficient. But yeah, I think it's necessary in that framework. If
you want to be decentralized, deep in crypto, it's definitely the way that you can cut out
the middlemen. And that way, it's the whole can't be evil scenario.
I think that was something.
Yeah.
Yeah.
So now, Conway, you go ahead and eclipse me.
No, I mean, this is, I think this one's a tough one.
You know, the solid ground that I go back to is just,
if you believe that open source matters in AI, then we have to figure out how to reward people for their contributions to open source.
And I think for all of its faults, crypto has made a lot of headway on how to efficiently reward people for completing bounties.
And I think when you add in AI's role in crypto,
I think AI has a greater role to play in crypto
than crypto has to play in AI, frankly.
I'd say they're probably right in that regard.
Mostly, this is slightly off topic, but I do think it's relevant to deepen also.
When we think about there's many jobs to be done, right?
And we have to verify whether or not someone completed that job to the degree that it needs to be completed in order to be rewarded.
I think AI is going to be really helpful um in that and i mean it's already um becoming helpful right um then on the flip side
and this is kind of a loose thinking it's like okay so if we if we extrapolate that ai can be helpful in reviewing you know uh contributions
then also couldn't an ai and or a creator of ai start to then create bounties to improve that ai
right so yeah it's like a loop um so i could see some hopefully near future where AI and its creators are just constantly kind of creating these jobs that can be autonomously verified in order to improve the AI itself.
And I guess the other thing is like to the critics, I don't know if it strongly says like, oh, this is why crypto matters, but it's like stable coins, man.
I mean, it's already here.
I mean, the biggest like, you know, the scale AI is doing tens of millions of dollars of stable coin rewards to the contributors in AI, you know.
stablecoin rewards to the contributors in AI, you know.
But then the thing that they're probably right about
is most crypto AI is just the narrative pump,
you know, to something that probably doesn't need to exist,
doesn't really add anything.
So that's my challenge to crypto again,
is like, can we use these mechanics to do things
that haven't already been done in a
centralized way and can we make like i i think what i want to see is that we create same thing
with dsai right like i want dsai to prove like hey because of these incentives we made something
that's better not because it's decentralized and not because it has crypto.
It's just better.
Yeah.
You know?
Yeah.
Yeah.
Dude, no, I think that's fire.
And Russ, even to go back to your point, like, can't do evil point. talking about like the commons for for ai etc that you know crypto has the people who are thinking
about how to you know build systems in trustless environments and that's frankly what ai is asking
for at least in web 2 so at least from the you, need perspective that there is, you know, although we could argue it's not useful, like not fully commercial or anything yet.
But there is that like kind of juxtaposition where it's like, you know, talking about open source use, et cetera, like verifying data and all that sort of stuff.
Crypto has all those thinkers and people who built nascent systems.
has all those thinkers and people who've built nascent systems.
So I think that's an important point.
And then Conway for you, just in general, yes,
agree with pretty much everything you said.
I think that's the challenge across just like the demand side problem in crypto.
Like crypto is great at aggregating the supply stuff,
but the demand for that supply has always sort of been a challenge. And I think, you know, building good products that are just that are full stop science and, you know, technical innovations before they're just, you know, just like a Web 2 version or Web 3 version of a Web 2 thing, like in actually advancing the space. It's super important. And look, I know a few teams doing some interesting stuff,
but love the challenge and call out.
So look, I appreciate you guys.
This has been fantastic, even with a small number of folks.
And hopefully you guys join again.
I'll be hosting these and growing them as we get going
and sort of building the community.
So thanks, man. I just wanted to leave one thing, which is WorldCoin, amazing growing them uh as we get going and sort of building the community so uh thanks man i'll uh
i just wanted to leave one thing which is a world coin i think honestly is a pretty
interesting example right it's something that didn't exist and probably wouldn't have existed
without the um crypto component yep no 100 look i appreciate you guys yeah thanks appreciate you
thanks man take care
