Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
Pump Fun launches DEX
Recorded: March 21, 2025 Duration: 1:29:00Space Recording
Short Summary
In a dynamic discussion, key topics included Canary Capital's groundbreaking ETF launch featuring NFTs, Telegram's substantial Toncoin acquisition, and Zero G's innovative approach to tokenomics and community engagement, all highlighting significant trends and growth in the crypto space.
Full Transcription
Good morning, everyone. Welcome to the modern market where every day we discuss everything to do with the modern market. That is the crypto market, the NFT market itself. We'll discuss that in a little bit
We've got canary capital filing its first ever ETF with an NFT component via the Pengu
ETF we'll get into those details in just a little bit and our takes on it
We also got telegram focused on foundation saying the VCs have bought over 400 million dollars
Tom coin will give you those details in a little bit as well as the general price action on crypto and NFTs, plus a general discussion on market thoughts as usual.
So stay tuned for that.
I've got Legendary and Brad with me in the co-host seats.
We're ready to get into it.
But just a reminder, friends, before we do nothing that we say here is financial advice,
this market is very risky and therefore we don't know anything for sure.
Please proceed with caution and exercise your own judgment with that out of the way.
Legendary, it is Friday.
Finally, it's the 21st of March.
How are you doing?
How are you feeling?
GM, GM doing quite well.
I did a little something yesterday as a result of our show.
And that little something was buying a first batch of Tesla stock.
So let's see how that will age over the next year or so.
I actually had this post up on X like for five minutes saying,
Howard Lotnick said to buy a Tesla stock,
so I'm following his financial advice.
But then I was like, nah, better to delete that,
especially after I've seen a couple of reports.
I didn't check if they're true or not.
But of some alleged French people,
journalists, whatever, who wanted to travel into the U S and they were stopped
from doing that after they had a couple more critical comments on social media
regarding Trump and administration.
So I was like, maybe I shouldn't post that Howard Ludnick is giving
financial advice on, on whatever it was, CNN or CNBC.
I don't know.
Reasonable take. Is that from one of one of his news
appearances? Because I was actually watching the all in
interview that they did with him. I think they only published
it yesterday. They published a whole hour long with Scott
Besant. Is it Besant? And then they did did a whole one with
Lutnick as well. So I'm like
halfway through that was watching that before the show today. Where was the advice? Was
it there or was it somewhere else?
No, it was on some sort of TV interview where they showed pictures of burning Teslas, etc,
etc. And he was saying basically stock is so cheap, like this guy's building so, so
much massively
undervalued. Apparently also holding Tesla stock himself is what some of the reports did say.
I'd be curious to know if you can even do that just, you know, from a legal standpoint, if that
is the correct way to show stocks on TV, but I have no idea.
the correct way to show stocks on TV, but I have no idea.
I don't know, I do think they feel kind of obliged to with Tesla in particular, because Elon's Elon is so involved, and has been so helpful to them. I think they probably feel a little bit obliged feel a bit bad for him in that, you know, the stock is almost literally going up in fire at the moment.
in that the stock is almost literally going up in fire at the moment.
Although, as we've been speaking about, one of the reasons it's kind of got our attention is
to get the cars for a second, which I still think are pretty cool.
I think a lot of people think that the optimus robot aspect of Tesla is just not being priced in at all.
Is that what you're thinking was legendary or you still kind of bullish on the car side as well?
Now all I want is exposure to human robotics and I told you that I watched this
Let me see if I can actually pull it up on screen real quick. I think I can this Boston
dynamics video there we go
Where you can see that this guy is getting super athletic even
Being able yeah, there you go, doing his little judo-like role, the Boston Dynamics robot that I'm sharing on screen. It's getting pretty impressive, but obviously that's not a publicly traded company.
And he's also going to do a little handstand, I think.
So the movements for those watching us on screen are looking, there we go, this is the handstand, are looking pretty, and then flipping over,
are looking pretty natural. That's seriously impressive. And fortunately, he's break dancing.
Yeah, this is a very agile robot, arguably more agile. I mean, we said we need to get
back in the gym. Legendary saying on a scale here, how, how, how is this robot more agile. I mean, we said we need to get back in the gym. Legendary saying on a scale here. How how how is this robot more agile than you at the moment or less?
Definitely more agile than me. I have to admit.
You know, alright, so most of that's like whatever it's doing the same motions as us. Fine. Cool. Whatever robot dude you're catching up fine. But what scares me in that video is the move the handstand move, because he's now doing the
robot thing. I don't know if you guys saw it, but like he goes up into a handstand, he twists his
hips totally around, and then lands the other direction. That's a robot thing, right? Us humans,
we can't twist our torso 180 degrees. So now we're starting to get in the realm of only robot
movements. And that scares me.
Doom is closer than we think, I think. Sounds like a content creator platform, only bots, only robots.
Oh yeah, I didn't.
My task, all right, so this is going to get kind of weird.
My task for like maybe test and tests and launches a hooray candles.
Um, but I had to test one of our applications for that.
And like, I have to, sorry, only robot.
Um, I'm wondering where this is going.
I'm excited.
There's one of our apps is an AI AI girlfriend platform.
Right?
So it's like, these people are AI companions, right?
So it's not like you, it's multimodal. So you can like call them, text them, do whatever, but I like, I have to start
testing this shit to make sure that stuff actually works. And that means like, I have to have, you
know, my girl here while I'm generating an AI waifu. And then like, it inherently is kind of
like leaning into the sexual thing. So it's like, I can do everything I can, like, ah, it's a muscular
girl doing whatever. And it's like, man, I, it's like, I had to defend myself from making AI waifu this weekend, or this past week. Yeah,
it's coming. How is that explanation coming across that for research purposes, you need to test an
AI waifu, which is also sexualized? Yeah, man. We're still figuring that one out.
Yeah, man. We're still figuring that one out.
How does it go? It's just the testnet version. It's going to get real.
Yeah, just wait till it lives on Jane. Not that she knows what that means, but.
Exactly. Interesting stuff. Yeah, need. It's funny how I think in general on the investing side, it does. I do feel like I'm in the mode now where new stuff's
happening. And because we've been in this mode for a few
years, it's like, as soon as you see something cool, it's like,
well, how do we invest? How do we invest that we've seen the
snapshot today? There's like this uranium. There's a uranium
piece of news. It's like, how do we get exposure? Nuclear? How do
we get exposure? robots? How do we get exposure? I think? How do we get exposure? Um, I think it's reasonable. Is it reasonable to be thinking about, uh,
these things that kind of in the arena? I mean, the dream is in the future, like you
get permissionless access to it, right? Like there's some form of stuff. So like outside,
like private companies that like do whatever, you know, I listened to a podcast yesterday
where they're talking about how like,
it's probably gonna be tough
to get a lot of these existing stocks
to take their like, their actual already deployed
to the stock market, Coinbase, Tesla,
whatever to get that actually on chain.
They're trying to do some of that.
You're seeing it now, right?
Coinbase is there, but they're saying,
the thing that we're probably going to see
is the primary market come across where you have these small to medium companies start on chain,
right?
Where then it's no longer us having to go and try to pull in the assets from traditional
markets.
It's on chain becomes the market, right?
It is the thing that then traditional people start doing the ETF style stuff where they're
trying to get access to our assets by doing the traditional formation of again ETF or you know whatever. So hopefully the flow,
if the flow starts going in that direction like shit gets crazy but you know it's probably not
for the foreseeable future. Sounds pretty exciting though. Funky, your hat is raised. Welcome to the
stage my man. What is your thought here? We're kind of going off on the tangent, what
kind of various things we can get investing in? What are your
thoughts on this or anything else?
Just just real quick question about Brad's waifu. Her name
Samantha, by any chance, because I can't help but think that the
movie, her is actually set in 2025. So it's kind of scary to think about how quick all
this stuff is coming, right? AI waifus, whatnot.
Yeah, so I did everything I could to make it like normal, quote unquote, literally like
personalities normal faces normal, normal. I literally wrote normal in all descriptions
and still spit out this like pinup girl like flexi thing. I guess. Yeah. I don't even remember what the name was.
I think it's, uh, I dunno.
It was like anime or something.
I'll see if I could find it.
It's going to be recurring storyline now.
I mean, quick, quick comment on that actually, because it happened just a
way a week ago in the Netherlands.
There is a guy who got married to an AI chatbot in the
Netherlands and the Dutch data protection authority had to comment on that.
And they were basically, that's like a legit commentary by the official, you know, data
protection agency talking about the risks of having an AI as a romantic partner.
Official report required for the government.
Interesting, bullish on AI moving forward,
wondering how we can invest in that as well.
Maybe that's something to consider.
Look, we've got to get into it.
We've got to slightly...
The YouTube comments, someone saying, Marco Polo saying,
imagine your married chatbot taking half of your crypto
in the event of a divorce. I mean, they would definitely know where it is as well. They would
definitely know where it is. Talking of crypto, let's get into it starting with the price action
as usual, what you need to hide from your future AI spouses. Bitcoin is down a percent to 84k. We've
got ETH down a percent at 1.9. XRP is down almost 4% to 2.4.
Solana is down 3% as well to $126.
On the meme coin side of things,
we have Doge down a couple of percent to 16.
Pepe is down a couple of percent to,
Chump's down 5% to 10 cents.
Whiff is one of the only ones that's kind of stationary.
Bonk is stationary as well.
Pengu is down 10% on the news of the Pengu ETF.
That's a bit of a shame, but we'll get into some of the bigger details there.
Along with the NFT price action.
And we have Fartcoin down 5%, although that's up 36% on the week.
So that went up quite a lot the other day.
And we have any other
relevant ones. Mog is down 5% that's been down for a long time now. I think that's
probably AIXBT is down a few percent as well so mostly down today I'd say on the
meme coin side of things in terms of the main headlines we've got pump fun
launching its decks that's been disastrous for other decks
is on Solana. Radium has really suffered as a consequence. So we're going to dive into that
in a bit. We've got Canary Capital filing the first ever ETF with an NFT component via the Pengu ETF.
NFTs did pump briefly and then kind of settled back down to where they were. I think the
and then kind of settled back down to where they were.
I think the Pengu token itself is the one that suffered the most out of that news.
Whereas NFTs in general actually in this whole downturn have held relatively strong.
Telegram focused Ton Foundation says VCs have bought over $400 million worth of Toncoin.
We'll describe, we'll get into the information you need to know on that legendary.
Is there anything on the NFT side that we need to know?
There's actually just one thing to really comment on
and that's the Good Vibes Club,
which has seen its first major correction yesterday
down more than 30%, down 31%.
The two point, at a 0.28 ETH floor,
but with that being said,
the volume of the collection is still pretty strong,
still 126 ETH in 24-hour volumes, which is translating to a total of 260 sales, but it
is the first setback in the collection's floor price, so they thought it's worthwhile mentioning
that.
Fair enough.
Well, there's always high volume when lots of people selling in selling, so that is,
I guess the team will still be
making their money there and hopefully getting some good
runway to continue building up and getting into the rest of
the WebD roundup we've got meta has a bunch of interesting
pieces here meta planet Japan's biggest corporate Bitcoin holder
adds Eric Trump as an advisor. I think a few people are asking
questions about what is going on there. But he's been added to the strategic board of advisors to boost Bitcoin adoption.
It's total stablecoin market cap has surged past all-time high of $230 billion.
At the walrus foundation securing $140 billion in fundraising led by standard
crypto as well as participation from a 16 z electric capital and a few other prominent
crypto as well as participation from a 16 Z electric capital and a few other prominent
VCs had slander decks or cut up a hundred percent
After its upbit listing. I think it was actually more than that in the end
You have mintify announcing the ai-powered trading of on-chain assets. It sounds pretty cool. The founder tweeted
This was evan who said
Excited to unveil mint ai soon. you'll be able to trade on-chain assets
effortlessly with simple prompts, and cover hidden gems through a personalized discovery feed.
I think that's pretty cool. Obviously, people are exploring AI stuff with trading kind of tokens,
maybe generating yield and saying go and park my stablecoin somewhere. I haven't really heard that
for other digital assets, so I wonder if NFTs will be included there.
DeFi Dungeons pre-sale hit $130 million yesterday,
committed for their game token 24 hours.
So it seemed like people were willing to,
some risk was ready to be put back on yesterday.
And then the thing I mentioned at the top of the show,
Uranium Digital announced a $6.1 million seed round
to accelerate the creation of new market for nuclear energy. I'm super
interested in that. The thesis for that, which was shared by
one of the investors was that this is an incredible like a
huge market. It's just not a good way for people to get
access to it at the moment. So they want to create a new
market for the future of nuclear energy and the
modernization of its fuel cycle trading infrastructure. And interesting stuff. That is the WebD Roundup. Legendary before we get into the
main stuff today, and we're going to have to go a little bit speedier. Remember, we've got the
interview with our partners Zero G Lab to a building seeking to build the largest L1 for AI
a little bit later in the show today. So that's coming up. Looking
forward to have that conversation. But before we get there, Ledge, what's anything catching
your attention here? I've got one thing to bring up in particular. Yeah, there's two things. I mean,
you can probably guess that they really find the Uranium Digital News pretty interesting.
This is the protocol, I think, that we had mentioned a while ago and seeing that there's a couple
of assets that are typically a bit harder to trade coming on chain is something that
I find interesting.
It's actually something that I want to trade myself as well.
And then on the other hand, also just Stefano mentioned it in the YouTube comments.
So I do think we have to bring it up real quick is Zach XBT's post, the thread on the hyper liquid whale. Zach XBT once again did amazing, amazing work in
identifying the guy I think even found his dual lingo account as part of this. Yeah, as a part
of his research. And it did turn out that this has been someone who's been engaged in illicit activities.
Years ago was arrested for stealing a million dollars from two casinos in 2023.
Do you want to share this, Ledge?
Have you got it on the screen?
I'm going to bring it up.
Oopsie.
There we go.
Going to bring it up on screen.
So this William Parker person was also known as Alistair Pakover before changing his name.
Also made headlines in the UK in the 2010s for fraud charges related to both hacking
and gambling.
And again, got arrested in 2023 again, and apparently is now using or has been using
some of these funds to gamble on hyperliquid as well. So amazing, amazing research, very interesting
thread, very interesting breakdown by Zach XPT. That's the depth of his resources is insane.
He goes into it a little bit there and just like the power and what he does is not because he's
like any good at one thing. It's that he's got like this incredible tool belt of stuff that he, when combined,
like in one stage of this, he goes in and says like,
he like has a resource for,
oh, I got this guy's phone number,
and then I have contacts at this other place,
and then I get their like, the phone number dug back into
the actual, like, you know, the registry for it.
And then I go, because I have that,
like he's talking about going into their Duolingo,
because I have access to the people who have access to the registry that's tied to the
number. And then like, it's just he's got to contact at all of these various disparate
things that in isolation are probably not like anyone can go cold, but he has the ability
to get through the next thing every single time. And that's why like in aggregate his
entire reports are crazy.
Yeah, you can see parts of that here
where he talks exactly about the phone number,
and then he finds the PayPal, the Duolingo,
the Chess.com, the Medium account, the Adobe account,
the Zapier account, everything related to that as well.
It's pretty impressive.
It just reminds me, it's kind of tangential,
but it reminds me of, I was listening to a podcast
two days ago on Larry Gagosian,
who's obviously this huge, huge art dealer, now probably the
biggest in the world. And they talked about his network as
being one of the reasons why he became so successful. So when he
had all these, he worked for some previously really well
connected people who had like, placed a lot of high value art
with people and that industry is famously opaque, right?
You don't know, you don't even know where the work is.
Like people don't know who's sold to who,
no one knows where it is.
So one of the reasons for his success
was that he built up this network over time
of knowing every single,
every single like famous piece of work,
he would just know who had it, who sold it to who.
And then he became the person
that was selling it from one person to another.
There's actually this story where he sold a piece
from one person to another person on a Wednesday.
And then by Friday, he'd sold it to a person beyond,
he'd sold it to the next person
before the people he sold it to on Wednesday
had even taken,
had even taken custody. And each time on each sale, he's taking 30%. So 30% was the number
that he would get on each on each commission. But the point being, yeah, but the point being without that, like specific network and build it, He basically like built this map of the world and all the great pieces.
And you did in aggregate in, in isolation, it means nothing, but together
it becomes like this super powerful, super powerful thing.
Now, let me extend that to the crypto market and give some advice to people
in the audience and listening on YouTube and wherever else you are.
Right.
Like the, this is something that it's always been hard for us.
Like being transparent, us to even price, right? Like there is value in the network that you have built up through and listening on YouTube and wherever else you are. This is something that it's always been hard for us,
being transparent, us to even price.
There is value in the network that you have built up
through credibility connections, common interactions.
You accumulate friends and confidants, acquaintances
throughout all this stuff, and it's hard to price that.
But there is value in that,
and I think that's a great example of it.
But you see it in other things, because you wanna say your uh, you know, you want to say your Twitter profile,
all this stuff, right? People that follow you back because of something you did six
months ago, a year ago, two years ago, like there is, there is value in that. Right. And
if you know those people, if you have a mental model of the person that would be good for
X for any other person, um, it would be, it would behoove you guys to recognize that and
try to do your best to leverage it.
I agree. Great, great advice, I think. Okay, let's go. We've got a bunch of stuff to get
into. Let's, let's get straight into it then bread. First headline for today is pump fun
launching a Dex. What do we need to know here? Is it over for the rest of these Solana applications as PumpFun wants to take it all?
Yeah, it's going to be tough because they own the user.
So what it is is PumpFun announced yesterday that they are launching a DEX of their own.
PumpSwap specifically is the name for it.
It's going to have instant migrations, zero migration fees down from six of the previous one,
because again, they were using radium previously.
So pump would launch a coin, they let them, they trade it locally and then eventually
it would migrate to radium.
That's why the rate token was pumping.
That's why everyone was kind of bullish on them for a while.
And actually counter to this radium is also launching their own launcher, right?
So now you have both of these people that have verticalized the stack to try to acquire
100% of the fees the other person was the other company was
So this is natural. It's a progression of any any successful
application frankly, but anyways
Instant migration zero migration fees down from six soul more liquidity in are doing a creative revenue sharing in the future
we're a portion of the
actual fees that are acquired from trading
on the decks will go to creators.
Now they have a breakdown of the fees as they are set to date and it's roughly 0.25% fee
on all swaps on the decks and 0.2% of that goes to the actual pool, the pool itself they'll
be and then 0.05% goes back to pump swap or pump themselves.
Now to understand the model of pump, previously, if you guys aren't familiar, pump actually
made money whenever you were trading on the bonding curve that it had before the coin
actually migrated to the pool.
Now as a business, they are trading on, they're capturing the fees on the bonding curve.
And then now once it migrates to the actual DEX
where it's an open permissionless thing,
it's got a liquidity pool,
that and they're now capturing fees over there as well.
So like they're just verticalizing the stack
of what was value leakage,
because there's been millions and millions
and millions of dollars that have been earned to Radium
because that's where they sent their coins,
but now they earn that.
Another few of the notes,
just the DEX itself
is similar to Radium V4, Uniswap V2,
in that it's a constant product AMM,
so it's fairly standard for the industry.
They're doing a $2 million prize pool
for auditing the smart contracts
to make sure that people try to find exploits
before they actually go live with it,
or not go live with it, but 2,000, or 2 million, just to make sure people find exploits before they actually go live with it. Or not go live with it, but 2,000,
or 2 million just to make sure people find exploits.
And then there are a list of,
as part of their initial campaign,
of tokens that are migrating to Solana
to be traded for the first time
in order to like seed some initial interest.
So some notable ones in there like say,
Jupiter's in there, Coinbase, the CBBTC is on there, Tron,
Aptos, Pengu's there, Frax.
Yeah, so a lot of interesting ones.
And then last and final note was I saw Athena
actually dropped an announcement
that they are gonna be a core pair asset,
core pair asset on the pump swap decks.
So the USDE can be used, uh, uh, natively.
One final note is the, the next speculation for this is everyone goes,
Oh my God, pump chain is next.
Um, exactly what I was about to ask.
What, what, where does this go?
Yeah.
So everyone sees this thing.
Oh my God, pump chains coming, which probably, you know, at some point,
because eventually you crush your immediate domain,
and then eventually you wanna capture even more revenue.
So right now the next thing that it's leaking revenue to
would be, you know, the chain itself and paying fees,
like whenever someone tries to do a priority swap,
they have to pay to the validators, all that stuff. So the thought would be like oh well let's just do pump chain because
then we can capture that little bit of fees. There's some headwinds with that. I'm not necessarily
sold that they will do it because they kind of do require a little bit of like community
composability a lot of like interactivity with the rest of the network. You know Jupiter they
get a lot of love from Jupiter because
Jupiter is a good aggregator.
People are going to go there to trade stuff, even if, uh, it's on
whichever pool doesn't matter.
Right.
That's why they like Jupiter is it because it doesn't matter which decks it's on.
And we've seen a uni chain, right?
Uni chain is at the point it's got like $10 million in total TVL, right?
Like the people don't go there just to swap coins.
You go there for a bunch of reasons.
And then because you're there, you swap coins. So I don't necessarily believe that there will be a pump chain. It
wouldn't surprise me. But that's, that's the logical next step. Interesting. Okay. And one
other final thing I thought was kind of interesting. I don't think this has been done before.
Create a revenue sharing. I don't think this has been done before. Creator revenue sharing.
I don't know if anyone's necessarily incentivized that beforehand.
Do you think that's an attempt to have coin creators stop dumping
all their tokens on everyone?
Maybe they were incentivized to like, no, make it be better. It's hard because the incentives are so skewed towards just getting
rid of this stuff, right? Yeah, like one, like the incentive is the creator should have a shit ton
of coins already. So like the incentives are already there. What this incentivizes and it's
something similar to what we saw with the friend tech stuff, right? If you're getting it on swap
fees, you're incentivizing volatility, not price go up. So it's not
necessarily aligned incentives. I don't they didn't have a ton of
mechanisms to work with. So it's not like they can just create
something that's perfect. But yeah, it's it's just it'll be
interesting. I don't I think it's negligible. I don't think
they'll have any any impact.
Okay. Yeah. As we often think, it's just like figuring out the incentives will probably
just will illuminate what direction it goes in. And at the moment, as things stand, the incentive
is to have a bunch of your own tokens and probably sell them as the price goes up, but get the revenue share over a period of time.
Ledge, any thoughts on, go on, Brad.
You're technically double dipping, right?
If I have a shit on my coins, I know I'm earning on fees.
I can hold onto it while it goes up,
scare everyone with a quick sale.
And then on the sale, scare people,
which increases volatility, which makes me earn more.
And then like, oh, well now I can buy more.
Then I buy some more, which gives people confidence, which should surprise back up, which then
gives me a little bit more fees because I get, I get 0.5% on that. So it's like, yeah, I
don't know that's necessarily aligned. It's kind of like royalties for NFTs, right? Where
people, exactly. Sometimes teams make huge monies going up, which people think is good.
But then they also absolutely print as the price is going down. So interesting points. Leg, anything to come in on here? Any thoughts? Yes, sir. Two
comments on that. A shout out to just Stefano and the YouTube comments making the point if we ever
launch pump chain, people can't take crypto seriously with names like that. If we have a
chain called pump chain. Most serious comment on the stuff that Brett was saying regarding Athena though.
It's super interesting. If you read into their announcement,
they're saying it's the altcoin market transition from BTC pairings to USD.
We expect meme coins will eventually follow the same path with Athena playing a
key role.
So they do want to see meme coins being denominated against USD and not against
Solana or not against whatever
other token they're trading, which then kind of begs the question, if you see on the one
who is positioning themselves as being this like onboarding infrastructure for TREATFI
institutional capital on the one hand, and that's the stable coin infrastructure that
the likes of BlackCroc and others will be using. And then on the other hand, they are also the asset against which your next 10k shitter is
denominated against. It's a very interesting balance to maintain if you want to play in both
like the extremely degen arena and become the denominating stable coin for meme coins
that are trading on the pump decks.
And you also want to have the stable coin infrastructure for the opposite end of the
spectrum to trade for institutional capital.
I think Guy is just smart as F. Like the dude, like the creator, you know, like you hear
him talk about a lot of stuff.
I think he just knows like, and this is where I have tried to talk about recently, where I think the meme of exchange element of the moneyness is where a lot of stuff. I think he just knows like, and I this is where I have tried to talk about recently where I
think the medium of exchange element of the moneyness is
where a lot of things get its value. He's trying to see
himself as the medium of exchange asset and a lot of
asset creation areas, right, whether it's on their own chain
or now in pump, I just think it's a he's smart to just
inject himself there.
Yeah, my I mean, I think it's a, he's smart to just inject himself there. Yeah. My, I mean, I think it's a good point to bring up.
My thought on that is, this is what I would say.
In a way, I take your point,
but we wouldn't say the same thing.
Cause if you think about what their intention is,
and I hold no ENA at the moment.
So I'm similar to, similar to over to overarching the comments at the moment saying been debating
taking a big ENA position, but they're unlocked and token jobs
are always sold. So it's like, I'm trying to figure out the
token position, like I would love to get exposure. But I'm
still quite bullish on the team, at least and what their vision
is. If their intention is to become money, like we wouldn't
say, oh, us circle at USDC or
are trying to play in both, both sides, right? They're just
trying to become completely ubiquitous. Yes, we're not
exactly a stable coin. But we are, you know, we are synthetic
dollar that can be used for whatever it might be that you
need. Yes, we've got these institutional connections, but
you know, we are basically
just money and you can use us throughout the economy and whatever direction. So I don't
think I agree. But the pushback on that would be to say that exactly this is the criticism
that's, you know, has been constantly going around. And yes, it's gotten weaker that all
of crypto is used to money laundering, illicit weird activities, drug pools and scams. And
then if you're like more conservative institution,
you would probably have the same.
Oh, I see.
And like, oh wait, you're also the guys doing
that weird meme coin stuff as well.
And not only our side of the business,
this is where I still think there's like a bit of a issue,
a bit of a challenge to solve with that narrative.
Yeah, okay.
I see what you mean.
I see what you mean that that's a fair point.
Funky,
you've got something to come in on this before we head into the next headline.
Yeah, just real quick. I mean, I totally agree with Brett. I met Guy at DOS just a few days back. I watched him speak on a panel and really that does kind of surprise me considering what he's talking
about on the panel was sort of just making the converge chain all about luring in the big trad five players to kind of use that
as the place to do a lot of their trading and stuff like that. So it was interesting
just hearing all this back and forth. But yes, I completely concur with Guy. He struck me
as super, super smart. I chatted with him for a couple of minutes after he was on the
panel.
Good stuff. Reporter on the ground, funky coming in strong.
Appreciate it, man.
Look, we've got about seven to 10 minutes
before we get into our big interview for the day.
We're speaking with Michael, CEO of Zero G Labs
in just a little bit.
Two headlines to still get through.
Let's get into this Canary Capital Pengu ETF one.
There's a lot of excitement on the timeline about this one. This is the news that Canary Capital filed for a very unique first of its kind hybrid NFT and meme coin ETF with the Pudgy Penguin NFT
and token Pengu included. It's important because I think it's the first example where NFTs have
been included in this type of a filing and the the first time a huge Web3 brand has managed to demand this type of exposure.
In terms of the specific details,
they said in seeking to achieve this investment objective,
the trust will invest in a portfolio
consisting primarily of Pengu,
the official token of the Pudgy Penguins project,
and secondly, Pudgy Penguins non-fungible token,
so the NFTs.
And the trust will also hold other digital
assets like Ethan soul, but only to the extent
that they're necessary or instead of sort to the purchase
of Pengu and the NFTs.
Crucially, the crucial detail, which I'm not sure
was even in this headline article that we're sharing
on the screen, but it's important to know is that
the specific amounts or the ratio between Pengu tokens to NFTs
is 80 to 95 percent Pengu tokens and five to 15 percent Pudgy Penguin NFTs. So I think this is
pretty historic in some sense. Awesome news for the Pudgy Penguins team and Web3 more broadly.
It's also I think it's pretty gratifying to see teams that do ship things keep pushing forward
I think it's pretty gratifying to see teams that do ship things keep pushing forward.
And no, I don't think that necessarily means that it's immediately a hit with loads of
demand and the things just all kind of fall into place.
But I think they deserve to be in the arena.
They've shipped so much when it comes to real world stuff that it's nice to see them be
in the conversation and push things forward.
Now the next challenge will be obviously to compete with all the other ETFs that we're
seeing. Obviously you've got the king ETF Bitcoin which is everyone is just starting
to wake up to and starting to allocate to and has done extraordinarily well. But we
are obviously seeing a whole host of other ETFs going through and that trend is definitely
going to continue. So a lot of competition but I think it's pretty cool to see that come through.
Ledge, any thoughts here on this? What are your thoughts? What are your thoughts on the proportion
as well? Yeah, honestly, ICO Beast had a pretty spicy take on that. I'll bring it up on screen
real quick. ICO said, I really don't think it should be a hot take that a Pengu ETF is quite possibly the dumbest fucking thing I've seen in a while.
We don't need ETFs for ghost change, much less sub six month old meme coins.
If the boomers are barely interested in buying ETH, much less soul, why in the world would they even consider touching something called Pengu?
I tend to agree, honestly, as a bullish on Pango, bullish on the ecosystem that Luca
is building.
And, yes, it's cool to kind of prove the concept theoretically that you can push for a meme
coin plus NFTs to be listed as an ETF, but you truly don't need that.
And a few people are making the point that Hachipenguin is not an NFT project.
They have retail stores selling the toy lines,
et cetera, et cetera, et cetera.
ICOB said, yeah, so this is KFC,
but they don't have an ETF.
And also I don't think this translation works.
Like if anything, and I said that in a post myself,
I'd rather own Pudgy Penguin stock
if that were a publicly traded company
at some point in the future when they're big enough,
but I don't see this translation working
that someone buys Pudgy Penguin plushies toys and is like, Oh, there's
also a penguin ETF.
I might buy that as well.
So, you know, no shade thrown.
And I think it's cool to prove the concept that you can have this
as a financial asset, but I don't really think it's necessary to
have a Pingu plus NFT ETF.
Fair enough.
I think that's definitely one side of things
Bunky your hand is raised. Got to keep it moving quickly here. What's your hot take here?
I just don't think that people these things are going to come we're going to see a lot more of these ETFs because it's probably done the same way that rec shares did the move ETF
you just throw in a filing the sec is probably not going to be had taking time to look through a lot of these things and it's just on a clock.
So my guess is this was put on the clock months ago, shortly after the Pengu release, and
then it just kicks over and if they don't get the stop order from the SEC to actually
have them examine these things, it basically just gets listed.
I think we're going to continue to see a rash of these kinds of ETFs pop up.
I definitely agree with that.
There's going to be a hell of a lot
a lot of competition amongst them to, you know, compete with what is probably not a huge amount
of capital that's wanting to get into digital assets yet. And if they do want to, or as they're
learning more, it's probably going to get directed to Bitcoin, right? At least as that the first
hurdle that people need to get through. Having said that, I think my, my, my, my thought still stands, which is, you know, we want to keep pushing the space
forward. And I think I actually, I don't mind that it's, it's there. I think it's
cool that aiming, aiming high, man, just aiming high, absolutely going for it.
Yes, I think it's gonna be really challenging. And it will be now it's kind
of like the thing where
people do airdrops and people wonder like, is this worth doing? I think it's a door that is now open,
that was not open before. It's an opportunity to win demand. And so that's my optimistic perspective
is like, yeah, sure, we can just say that there isn't the demand now. I think that's that's
probably true. But it's an opportunity to win demand demand and let's see how they go. Good stuff. I take the kind of pushback on that as well though.
Final headline for the day, we've got Telegram focused Ton Foundation buying, selling, sorry,
two VCs, a lot of Tom. We've seen this a bunch of times that recently, right? We spoke about Athena selling like a ton of OTC tokens, kind of resetting the raise amount in the sense, because it means that
everyone's cost basis is much higher, all the people who are holding, what do we need to know
here? Why is this important? It's actually a massive, massive headline that I think has been
under the radar for many people. There's a group of VCs that includes Sequoia Capital, Ribbit, which recently invested in
Privy, Benchmark, Kingsway, buying more than $400 million worth of Tone, the cryptocurrency
of the Telegram app.
And as you have said, they have been buying that from the foundation. Now, what we don't know is the exact terms and structure of the deal.
We don't know if that was discounted or to what level this was discounted.
However, what we do know is that the Telegram user numbers have been growing massively.
Telegram has recently surpassed a billion users worldwide.
That is for monthly active users, around 450 million,
sorry, about 450 million daily active users,
10 million being subscribers to Telegram Premium.
Not necessarily the most popular app in the US
with only 9% of the American population using that,
but 45% for example of the Indian population using it.
Just to add a couple more stats because I think they're interesting,
half of the Telegram users are between 25 and 44 years old, so it kind of seems to also fit well
with the core group of crypto users.
And they do make money, not only are the subscriptions, there is a lot of activity
going on, on the, on the token side of things as well with, what is it?
120 million unique holders of Tonecoin with 40 million blockchain users.
And I feel like I was honestly surprised to learn that
because on Telegram you have the subscription, you have the Telegram gifts which are rings,
stickers, teddy bears like the drop sometimes in limited quantities which you can buy and
obviously keep to yourself or gift to others using in-. And I did not know that.
And I'm glad that whale had a thread on that,
that the Telegram NFT gifts are actually
basically ton NFTs.
And they are seeing quite a lot of economic activity
on that.
Most of them do cost anywhere between two and $50.
And I've also recently seen a purchase by
Pavel Durov, the Telegram founder, purchasing the most
coveted one, the Golden Plush Pepe for 13,000 US dollars a
couple days ago, and they are doing $500,000 in daily volume
on that with 67,000 wallets who have traded gifts between their Telegram accounts
so far.
So it's interesting to see that this is basically,
I think one of the more interesting examples
for like blockchain technology that you don't really know
that it's happening in the background
while you're using the app,
because me as a Web3 native, I didn't know that.
And then I can see why you have Kingsway, Sequoia Capital, and some others coming in
and purchasing a very, very large amount of the TON token. But again, it would be also
interesting to know more of the terms of those deals with the foundation.
The thing that's interesting with this, like I was having this debate with Naveen, who's
one of the founders of Atari, which is a privacy proof of work blockchain.
And it's like, it's always the conversation around distribution and like who owns the
user and where the power is with that.
It's kind of similar to what Pump was just doing, right?
Pump owns the user relationship because of a lot of the coin launches and stuff, which
means they have the power to launch a DEX and then maintain drive fees, like verticalize.
The same argument was made early on for ton blockchain
because it's like, well,
Telegram is massively distributed, right?
Like it's got a shit ton of users.
And if you can just tack on things after the fact,
like a blockchain, a full-on blockchain,
this is their second attempt at it,
you can potentially utilize it in ways
that are valuable,
useful, and can drive a lot of revenue.
Now, the counter to that is that,
I think pretty universally,
everyone said 10 blockchains, dog shit.
Frankly, like it's just the coding languages
it actually uses is very opaque and different, right?
So it's hard to bring on developers.
It's not very intuitive, it's not very performant, right? It's like, to bring on developers. It's not very intuitive.
It's not very performant, right?
It's like, it's just got a lot of things going for it
that just aren't awesome.
And yet you see some of this stuff actually having success
like changing telegram names, actually selling that stuff.
Like that's something that's very obvious to us
as a use case that like is meaningful
and we think would be useful.
That people are selling them for hundreds of thousands
of dollars.
My question is, I don't know if you've like I noticed the stars thing previously. I thought that was a good way for
a blockchain or a
token forward company to issue or have tokens to be sold without actually having to sell them, right?
So you buy these stars, which are just like end game currencies. They can actually keep their Telegram app inside of the app stores, right?
Without losing a fee or they can actually just pay their fee.
But they can bypass a lot of the concerns that typically they don't let apps go
into the app store and then on the back end, you do the conversion to Tonecoin.
And then like you buy, sell and have some speculation there.
Um, I don't know.
Do you need to KYC to buy those stars in the front end?
Because that was.
Kinda. I don't know. Do you need to KYC to buy those stars in the front end? Because that was... Kind of. If you buy, I did buy stars before. So obviously if you buy in the Play Store or in the App Store,
you will have some sort of query card or other like means of purchase and to use the app or Play Store, you need to KYC previously.
So if you go via the mobile app, you will be
quote unquote soft KYC with either Android or Apple, but you
don't have to KYC specifically within the telegram ecosystem,
neither on the app side nor on the blockchain side.
Interesting stuff. Good to dive into the telegram ecosystem. We
don't we don't spend, we always know
that it is relevant and the numbers are extraordinary. Like the numbers are absolutely extraordinary
coming out of that ecosystem. So good to touch in there. Guys want to keep it moving now,
delighted to welcome to the stage our partners at Zero. Gee, Michael, welcome to the stage
with us. Great to have you on the show. I know you've
been flying around like crazy at the moment around the world. Where have you been recently?
Where are you now? Yeah, it's been a little bit intense. I've been to probably
eight countries over the last few weeks. So it's been super fun. But right now I'm in New
York for Daz. I took in the red eye yesterday, so running on fours of sleep or so, but it's exciting.
I think in the same time that we organized the interview, I've been like the time zone thing has
been doing my head in because like the daylight savings has been going but in that same period of
time I think you've been in three completely different time zones. So it's
just been like very very difficult to sort it out but yeah you're on the move
obviously because Zero G has a lot going on at the moment. I just want to
introduce you properly, get people up to speed with what you guys are building
so it's pretty interesting and then really excited to get into this conversation.
So obviously Michael is the founder of zero G he who is seeking to build the
largest L one blockchain for AI.
We're going to talk about what that means a little bit before zero G.
Michael worked as a consultant at Bain and at Bridgewater during the financial
crisis and built a startup, which went from 300K ARR
in its first year to 100 mil in four years.
After the business went through significant upheaval
during COVID, Michael met his technologist co-founders,
who were the founders of China's largest and only
regulatory compliant L1.
And after some back and forth, they decided to build ZeroG
together.
Interested to dive in there, they raised $35 million
pre-seed from top investors
like Alliance, Animoca, Delphi, Stanford Builders Fund, and a couple of others. And in this
conversation, we're going to be focusing on Michael's journey towards ZeroG, the vision behind ZeroG and
its product, and how Michael thinks about building the largest blockchain for AI and what that means.
So Michael, as I said, welcome to the show. Maybe let's start back at the beginning
of Zero G. So, you met your co-founders and one of the first things I've read when I was looking
into that was they were the founders of China's largest and only regulatory compliant L1 called
Conflux. So, how did this happen? How did that relationship start? What made you decide that
these were the group of people that you wanted to build
zero G with?
Yeah. One day, essentially my classmate from Stanford,
Thomas Yao reached out to me and we've done a number of
kind of deals together in the blockchain space.
We've learned about Bitcoin together. We listened to Mark
Andreessen before and I was a DFJ fellow. We listened to Tim Draper before. Some of
our classmates also were talking about Bitcoin at the time and essentially we
were getting the message, hey we need to really check this thing out. This was
like 2013-2014 time zone. We bought some Bitcoin together and then later on, we
participated in ICOs kind of 2016, 17. Thomas's fund was also very early into Ethereum and Metamask and so
on. So we've kind of approached it from an investment standpoint before. And when he reached out to me, he essentially said that five years ago,
and this was late 2022,
five years ago I invested in this company called Conflux,
Ming and Fun, two of the co-founders
are some of the best engineers
and computer scientists I've ever backed.
I'd like you to meet them,
they wanna start something more global scale. And
so that was literally the premise. And we got together and after six months of co-founder
dating, I came to the same conclusion. I don't even care what we start. We just have to start
something. So it really started with the team. And Ming, for example, spent 11 years at Microsoft Research, wrote some of the key research
around distributed storage, distributed compute, wrote some of the first AI algorithms for Microsoft
Bing as well. Phan was his intern initially, but is a two-time Olympic gold medalist in informatics,
MIT computer science PhD,
University of Toronto professor as well.
And then they started conflicts together.
Wow, that's a extraordinary set of, I think, achievements.
Just a quick one before we get into more of the details,
when you talk about co-founded dating,
I definitely resonate from a business perspective.
Legendary is the person I've been co-founded dating for the longest. Brett has come in a little bit later. What's like,
when you're starting things out, because I think we've got a lot of entrepreneurial people in the
audience as well. When you're starting things out from a business perspective, and I know you've,
you've done other businesses before as well, right? You ran a startup called Garton. Your first
client was Apple, I believe, from the stuff I didn't you had like
extraordinary success before COVID kind of made things a bit
difficult. What do you what do you how are you trying to set up
that company with the people around you in order to be
successful?
Yeah, for Gartner was very interesting because I started
while I was in graduate school, essentially. So I took a class called the Lean Launchpad and that was taught by Steve
Blank. I don't know if you, if you know Steve Blank, but he's kind of the,
the father of the Lean startup movement.
So if you heard of things like MVP, customer development,
get outside of the building. So he's, he's really behind a lot of that.
And so he became a mentor to me. And his first thing is basically there's
nothing you can do inside the building. Like nobody's going to figure out, or you're not
going to discover customer needs by being inside the building. So get outside the building, talk to
actual customers, figure out what they want, and then build a business model that's viable around
that. And so that was the mentality that I approached things with.
And then the next thing he said is, well, once you figure out a need, then you have
to figure out how to scale it.
Like who do you know in a particular company if it's B2B is the buyer, who's the user,
who's the economic decision maker, who's the saboteur, who's the influencer in different
company decisions.
And so he was very
systematic. And so that was the type of mindset I took to building Garten before. And it was quite
fun during the class, we actually started producing revenue, because we started seeing that there was
something that customers really needed. We started out, so it was a B2B well-being company, and we started out with a kind of nutrition needs. And so just with the thing that was the most process that we take into our bodies generally, which is snacks. And we, within two weeks, put up an e-commerce site just to test a bunch of things out. And then about a month into the class, the first order from Apple came in. And it had like an at apple.com email associated with it.
I'm like, oh, really?
Did Apple just buy something from us?
And then I started LinkedIn stalking this person.
It's like, whoa, this is a real person.
And she works with the VP of design at Apple.
This is so cool.
And then basically hand-delivered stuff
and just asked her a bunch of questions around why did you do it and so on.
And then, you know, just took took off from there.
But that first order is just like, I don't know, mind blowing.
It's just so bullish on on that that interaction that you just had.
Like, like whenever someone does something that is shows that they latched on, right?
Whether it's like onto something that you wrote or something that you're doing,
something you're building and you get immediate it's like onto something that you wrote or something that you're doing, something that you're building,
and you get immediate feedback from like,
what resonated with you?
Like, I'm, because I know from your side,
you're like, I'm just trying shit.
Like I'm doing something I think is right.
I'm not sure what's resonating.
And then it hits with someone and you're like, all right,
I need to distill whatever they saw
into like a repeatable scalable thing.
And then that's kind of where something can come from.
So yeah, you need to have that intuition to seek out that feedback from those initial people to really curate stuff.
And do you remember what she said? What clicked with her to make the order with you guys?
Yeah, what clicked was that essentially we curated an experience where she didn't need to think about
it. Because if she goes to other places, she's like, well, I don't know what's healthy.
I don't know what to order.
I don't know if my team's going to like it.
And the way we approach it is basically
like we're going to take away all the nutrition
pieces for you.
We will do that.
And then we will also do a bunch of the taste testing
so that you know you have a great product in front of you.
And so we had maybe 30 different products on the side at the time that they were all taste tested, which is a really tough job. I'm just kidding.
And it was just a very easy experience. And so that's what she really enjoyed about
that experience. And we just kept building on top of that. And then eventually we discovered that
it wasn't just kind of the snack pieces,
but it was overall nutrition.
And then we built a service around that, that scale.
And then beyond nutrition, we figured out
that it's actually billing is a big issue,
because usually people just get a bunch of Excel sheets.
And they're like, wait, did I really
spend $10,000 on gum?
This doesn't seem right. My entire bill is $50,000, but $10,000 on gum? This doesn't seem right.
My entire bill is $50,000, but $10,000 is going to come.
Something is not right here.
And so we built a SaaS platform for that.
And then it just kind of went like this.
And we just built more and more and more services.
So it really became a one-stop shop experience for all
well-being within an office.
That's incredible.
One thing that you mentioned for one of the,
maybe the professor said is you can't figure out stuff inside,
you got to go outside.
I don't know if this is the right comparison,
but I think is there any comparison there with
when people are building in like one region
and then trying to go outside?
Because I know, for example, we were, we were speaking to legendary before
this, actually, and we noticed that a lot of the zero G
traction, for example, where maybe in the Western bubble, a
lot of it is out in the east. And as you said, your, your
co founders are the founders of the of China's largest compliant
or that's what they were involved in.
How do you see these two hemispheres?
To what extent are you trying to be big in both?
Do you feel like you need to specialize?
How do you attack both angles, or are you just
going for all of it in one go?
Yeah, when I was running Garton, the answer
was very different versus today. So at Garton, the answer was very different versus today.
So at Garton, the way we looked at it
is we just wanted to know everything about one segment
so that we could scale very, very rapidly in that segment.
So it was literally very closely defined.
50 to 250 employee companies that
are tech-focused in the Bay Area.
It's super narrow.
Very, very specific, yeah.
Yeah, because I just wanted to know everything
about their customer type, because they were eventually
paying for our service.
And so scaling was much, much easier,
because they had a certain specific set of needs
that we could then meet both on the technology
and on the product side.
And then it made scaling very, very efficient.
Now, with 0G, it's a little bit different
because we have a high performance, essentially
layer one.
We've got a decentralized compute network.
We've got a decentralized storage network.
We have a service marketplace, which is kind of an app store,
if you will, and then AI alignment nodes on top of that.
And so it's a very modular design
where different customer sets can figure out what they need for their particular platform. You can
use all of it or you can use a part of it. And so the regional focus is not as useful anymore. It's
more a psychographic focus, if you will. And so people in the West can use it.
People in the East can use it, depending
on the different types of use cases
that they're looking for.
And so it's been less of a geographic focus for us.
And so we want to make sure that the key regions are represented
from both a trading as well as a usage volume
perspective for Zero G.
So it's a different approach.
Got it.
I understand.
I think Brett's going to dive in a bit more on the product side in just a moment.
Talking about more kind of global demand though, one thing that is global is the money, I think,
and there's been real demand on the investment side for Zero G.
I think, I don't know if I have all the details of all of
the different raises, but there's the $35 million pre-seed from top investors like Alliance, Animoca,
Delphi, and a few others, OKX Ventures as well, is incredibly successful. I think you did, I think
maybe there was a seed round as well, and then the node sale on top of that. Do you want to speak to
the strategy of just raising funds and why you thought your attempt to do that was so successful?
And where you guys stand with that at the moment?
Yeah, within the ZeroG ecosystem,
there's a couple organizations at least.
One is the Zero Gravity Labs entity,
which is the core contributor to the network.
That's what I'm the co-founder and CEO of.
That entity has raised the pre-seed, which is 35 million,
plus another seed of 40 million.
And then there's the ZeroG Foundation,
which is more kind of like the governance
and long-term kind of grant and investment entity,
if you will.
And that entity has conducted a successful note sale
of more than 33 million, and also has a liquid token commitment
of 250 million as well. So each one of the organizations needs
needs to be capitalized differently. And the strategy
around it was just to make sure that we have a high performance
ecosystem over time.
Got it makes Makes sense.
And Legendry, I think it's going to dive into some of the
tokenomics maybe a little bit later,
because you revealed those not so long ago,
which is pretty exciting.
But Brad, maybe I'll come to you here.
Brad's our technical expert.
So he's going to get into some more of the product
and what maybe the vision is.
Yeah, honestly, so Michael did a little bit of the heavy
lifting for me right there, but just to summarize, right, because zero G is not your typical construction.
And I think you probably have received some headwinds because of that, right?
There's a user education journey that you have to have.
So like, I'm just going to do a high level and then maybe you can you can bridge the gap or anything that I didn't cover.
So, you know, just to summarize for in ways that people might understand in the audience, right, it's an L one.
It's EVM compatible.
But the thing that you talked about earlier, right?
It's modular, so it's not,
it's like the extreme modularity, right?
Because we use modularity in a space where it's like,
okay, Ethereum is modular
because it's got a bunch of networks,
but yours is modular even within your own network,
kind of spinning up a bunch of different networks.
So you can use the word or the term networks and chains separately within your own network, kind of spinning up a bunch of different networks. So you can use the term networks and chains separately within your documentation,
because it's not necessarily the same thing with how you guys are comprised.
One thing that's interesting that I think is worth calling out is how you guys actually
leverage Ethereum. So you're on L1, but to the slashable stake of your validators actually is
on Ethereum mainnet.
And you use that to inherit some of the actual properties
of Ethereum and then go on to have your various networks,
right, your DA network, your storage, your compute networks,
and then all of these other networks
and then the chains within the zero G network
actually burn zero G to go back to the consensus network,
which is where the validators are.
Legendary will get into some of that. But one, is there, so two questions for you. One is, is there anything
else outside of that static basic high level architecture that you would like to call out?
And then two, how has that user education journey been because you are so unique in this space when
it comes to your specific construction?
Yeah, the key thing to call out is we, and this is where the name Zero Gravity actually derives or Zero G, is this idea that eventually we want infrastructure to be completely invisible to the
end user. So on the analogy that I use, if you're using Netflix, for example, today,
you don't think about which AWS server to choose,
or which encoding algorithm to use for the video,
or which kind of payment methodology
to use at the end of the day.
As an end user, you just want to use the application
and never have to worry about any outages and so on.
So the infrastructure, the idea behind it
is how do we get something to be infinitely scalable,
and how do we get something to remove any of the barriers
to adoption from Web2 to come over?
Because if you come over from Web2 and you're like, wait,
why did my infrastructure costs rise by 10x?
And why am I getting a ton of failed transactions?
Then you don't want to switch.
Or I'm forcing my users to use some type of wallet application
that requires a bunch of seed pass phrases,
and it's very difficult to use.
And so the way we thought about our architecture and design
is to remove all of these barriers
so that there is no friction.
There's a sense of weightlessness.
There's a sense of zero gravity, essentially.
So that was a bit of the design principle.
And so the key thing for infinite scale, for example,
is it happens at multiple levels.
So the first thing that we did is basically say, OK,
we've got a decentralized storage network that's
high throughput for AI workloads.
Because if we solve the AI workload issue,
then we solve all other use cases as well.
And so we built it and tested it.
That's about 2 gigabytes per second and throughput,
which I think is the fastest ever recorded for decentralized storage. So now you can actually
use it for hot forms of storage. Then what we said is, OK, we've got this high performance storage
network. We can actually then use it as a data availability layer as well. And because the key
insight there is, well, what if we segment blob data into a data storage
and into a data publishing lane?
Because then you're removing the broadcast bottleneck
across the entire network.
Rather than sending blob data to 120 or 200 validators,
however many you have on the network, at the same time,
you just store it once.
And then you send KZG commitments across the network,
like very small bits of data rather than these really
big chunks, essentially.
And so then we're like, oh, this creates paralyzability.
This creates more throughput because every node
you add to the network now adds to the overall data bandwidth.
And so what we found is that the theoretical maximum for that
is for 5,000 nodes, we should be able to get to about 50
gigabytes per second in throughput.
And that's already enough to start doing some small scale, fully on-chain training, essentially.
But we didn't stop there because if you look at modern data centers, it's not just 50 gigabytes
per second, it's hundreds, maybe even a terabyte per second in throughput. So how do we then bridge that gap? Then we said, well, why don't we take a data sharding approach?
And what we found there is that the consensus becomes
the bottleneck.
But what if you just can spin up an arbitrary number
of consensus networks at the same time
by utilizing like eigenlayers restaking
for economic security or other type of restaking platforms to do that.
And then we're like, oh, OK, great.
Now we can horizontally scale consensus networks.
And then now you have infinite scale at the data throughput
layer.
So then that then became a key input
to then scale also the execution layer.
So the way we're approaching execution layers to then say, let's have a
n number of arbitrary chains that we can spin up as long as there is a master chain that keeps
all of the different transaction registers available. And so then in fact, you get like
infinite TPS, quote unquote, through that. And so then we were able to scale the execution piece
as well.
And then the final piece is latency.
How do you get below 100 milliseconds of latency
on a layer one network with a three-hop consensus?
It seems impossible because speed of light,
that's the fastest you can get to.
But what if you do kind of localized consensus?
Kind of what AWS does with like US West, US East, and then just make sure
that there's global consensus over time. So we basically just went through, you know,
one by one by one to remove barriers to adoption. So that was the kind of key mindset.
That's interesting. So like, that's actually, it reminds me of the journey of even EigenLayer,
you referenced them just a moment ago, Like Eigen DA was their like key product
that they were actually shooting for.
They're like, well, what the hell,
how do we secure this thing?
And then that's actually how they came up
with Eigen layer as a product
to then backfill their Eigen DA thing.
Hearing you go through that thought process of,
you did one thing, which led to the next thing,
which led to the next thing, which led to the next thing.
Just that product flow is really interesting.
And just context for people in the crowd,
like he's calling out gigs per second
as like a bandwidth thing,
like just, and especially in the data availability stuff,
that's something I'm familiar with.
Just relative numbers in the space.
I think Celestia today is doing like 1.5 megabytes
per second.
Eigen DA is probably the fastest DA layer we have today.
And there are like 15 megabytes per second,
just for relative numbers.
So if you're talking gigs per second and tens of gigs
to hundreds of gigs per second, it's
orders of magnitude bigger as an ambition.
So that's actually really impressive.
So before we drift too far, I noticed in the documentation
you referenced largest chain for AI several times.
So first of all, what does that mean?
And then what is that unlock?
Why is it unique for AI specifically?
Is it these numbers?
Is it something else?
Can you expand on that?
Yeah, absolutely.
So we've got more than 300 projects building on top of us.
So that's kind of, we're trying to build
the largest ecosystem around it.
We have a very large community getting close
to 3 million
wallets on the test net, half a million smart contracts deployed,
close to 400 million transactions, and so on.
So it's been super fun to build and see that traction.
And it's really focused on use cases
that have some association with AI.
I mean, it could be, for example, a game that's
using a non-playable character that's
calling, let's say, our inference.
That can be part of it all the way to hardcore AI
infrastructure, like VectorDB is building
on top of our key value store on the decentralized storage
network.
So that's the spectrum infrastructure, consumer apps,
and so on that's
building on our chain.
So feeling really proud of being able to build that.
And then specifically, AI use cases
are quite important to us because when we got started,
ChatGBT was starting to gain a lot of traction,
ChatGBT 3 at that point.
And we basically fast forwarded 10 years into the future
and thought, what does a world look
like with a lot of automation in it?
And essentially, you can start automating even what I call
societal level use cases, so things like logistics systems,
like running an airport, or manufacturing systems,
or administrative and governance systems.
And how comfortable would we feel as a society
to have a closed AI or a black box AI?
Players manage these systems.
For these societal level systems,
wouldn't you want to know what went behind the models?
Where the data came from?
Who labeled the data?
How the model was trained?
What version of the model am I even getting?
How is the agent behaving in production, who even owns the agent. If I create an agent today, do I know that I own
it? No, it's in the terms and conditions, it's likely not owned by you. So all of these kind of
key questions came up and we said we have to build an alternative that's owned by kind of people that
are participating in the network, that's owned by a community, so that we can make people that are participating in the network,
that's owned by a community, so that we
can make decisions that are more for abundance
versus extracting value.
Yeah.
You started leaning in towards some specific products
that you were liking as far as concepts.
Do you have any other specific ones that you're excited about?
What pointed builders are you most excited about?
What are you hoping is built on your chain?
What's some of the ambitions for what you have being created?
Yeah, there's quite a bit being built.
Some of the ones that we've highlighted recently
are, for example, an AI-driven version of a DEX
where they're using AI algorithms to automatically
rebalance the LP pools, for example.
So that's super cool.
And then likely in the future, they'll
add some type of prompting interface
as well to make it super simple to do swaps or even more
complex type of transactions.
That's a product called Zero.
There's other things that are coming that
are really interesting, like on-chain credit scores that
utilize AI so that you can give larger leverage loans
if you've seen a good actor.
But personally, the thing that I'm most excited about
is utilizing the power of blockchains
to actually align AI models as well.
And to me, AI alignment doesn't just happen in production, it actually starts with the
data itself.
Like, where did you get the data from?
How do you know this data wasn't poisoned?
And the silly example I use there is if I'm wearing a Nike sweatshirt, so nothing against Nike, but let's say
Nike approaches OpenAI and says, I'm
going to pay you $10 billion so that you train your model
in such a way that every time somebody asks something
about sports, that Nike comes up.
Feasible to happen, and the end user
would just keep getting some Nike messaging,
and they're like, I guess's like associated with all sports and but that's a bias
introduced into the model and it's a bias you may not even be conscious of so
how do we make sure that we know the types of biases that these underlying
models and agents actually have and so I think there's gonna be a lot of
fundamental research that that we're to be doing in that area.
Nice. So I'm going to throw out one idea and then just pass it on to Legendary because we're running the one time.
So like I actually heard a conversation around blockchain solving for the difference in inference costs,
like the market price of an inference cost, right?
Because you have several models and if I have the same query to three different models,
one of them may run for five minutes, and one may run for one minute.
That means the costs are actually
different on the back end.
So if you can actually spin up a market
around the different inference costs
and maybe pay upfront with tokens
on how much I'm willing to pay on a specific call,
there might be some use case there.
But we're running low on time.
I'm gonna pass over to Legendary
so that he can see a little bit what's going on
with the forward look of Zero G.
Yeah, absolutely, thank you. Look you are building
a very very complex ecosystem and we learned a bit about that in the interview. You're also building
that in a very volatile environment. Like a lot of people go so far to call it a president's market,
everyone is waiting to see what is Trump saying, what is he posting, what is he saying at the
conference, what's the next move there?
And the whole market is trading this with extreme volatility.
And on the other hand, you have quite a similar picture. I would say in AI there's new tools coming out every day.
The attention span is a very short one.
Every week we get excited about something new, whether that's bold,
whether that's manners and the aficionados
move on from tool to tool. So how do you deal with this, like, as a founder with the volatility
in both the market that's very volatile, but also in the attention span? Is that impacting
you at all?
The attention span, not so much, because it's kind of this idea of one upmanship, one day R1 comes out from DeepSeq,
another day Quen comes out, and then attention shifts, as you mentioned, to Manos and so on.
The key thing from our perspective is just to make sure that we support the latest frontier models
that are open source. So as long as we do that through our inference and soon-to-be-released fine-tuning capabilities
on our network, then it's not as big of a deal.
Because what we're trying to fundamentally do
is not build everything ourselves.
We're not trying to rebuild an open AI, which, ironically,
again, is closed AI.
But we are creating a one-stop shop experience,
utilizing all the best that Web3 and AI has to offer,
like all the different types of builders in this space
we want to bring together so that it
is as easy of a user experience as doing a bunch of API calls
and OpenAI.
That's fundamentally our goal.
So as long as we have a complete pulse on the market, what's
happening, where it's moving. We support the best open source and Web3.ai projects.
It's not too big of a challenge for us
to kind of keep focusing on our long-term vision.
As far as the volatility in the Web3 market,
that's something we're not quite as worried about
because it's a very long-term journey to make AI a public
good, which is our mission.
That's not going to happen in one market cycle, for example.
So we really have a build for the long-term type of perspective.
And so not to be shaken by short-term volatility.
It impacts things more like a token generation event, maybe, but in the large scheme of things
from a pure technology and research-based kind
of process and deliverables.
For example, we can't solve decentralized training
in a couple of months.
That's a research problem.
That's a very long-term problem.
And if we make good headwinds this year,
I'd be very happy on that.
So not too worried about that either.
Yeah, you mentioned this long-term perspective.
And you mentioned this ultimate vision of the Netflix state.
You're using an app.
You're using an ecosystem.
You don't really care.
At least 90% of the users don't about what's
going on in the background.
But you also mentioned you're building so many individual
building blocks.
There is the AI Launchpad, which we had talked about, the AI DEX, the on-chain credit scores,
all of those things are happening. All of them are interesting and we love to nerd about them
in our bubble. So how long do you think will this transition or this journey take until we
stop getting excited about the individual building blocks and we will transition into this Netflix state
of just using a fully integrated holistic AI ecosystem.
Is it years?
Is it a decade?
Is it longer than a decade?
What would your best guess be?
My expectation is a couple of years, I would say,
because I would consider that two years as a catch-up phase.
As I kind of hinted at that fully on-chain training
for LLMs, it's just not possible today.
State of the art is about 10 billion parameter type
of models.
I think Prime Intellect released something
where they trained between two data centers this 10 billion
parameter model fully decentralized.
But what about a 400 billion parameter model? And by the time two years is over, we're likely at a trillion parameter model fully decentralized. But what about a 400 billion parameter model?
And by the time two years is over,
we're likely at a trillion parameter model or more
at this point.
So how do we make sure that we have the infrastructure
to support this kind of large scale state of the art
training, for example?
Or even on the inference side, making sure
that we can have very large scale models
that we can actually then run in a fully decentralized,
verifiable, permissionless way.
Yeah, so two years of catch up.
And I think at that point,
we should have an environment where the end user
doesn't have to worry too much about the infrastructure.
And builders that are working with us
can again use different components.
So for example, if you're building a very large scale and high throughput L2,
and you just want to use, let's say our data availability layer, plus maybe some
of our inference and fine tuning for the agents that are working on there, you
know, totally fine.
We want to, we want to support all of these use cases.
I just got a quick one there, Michael, you mentioned all types of different builders who
are working with you. Chains have taken different approaches when it comes to
builders and to what extent people decide to support versus like maybe help, maybe some help
with marketing or maybe just taking a whole bunch of builders and saying like these are maybe more officially sanctioned
builders or officially approved.
And we've seen a kind of varying success with both approaches.
I don't think there's one specific approach that people
know to be correct yet.
How do you think about working with the builders and the apps
that are building on top of you?
What's your approach there?
We have two primary modalities. So we overall have an 88.8 million ecosystem program. And as part of
that, we have two modalities right now. One is what we call the Guild on Zero G campaign, which is less
the kind of lower touch, which somebody essentially applies through a form. They tell us here's the application or the infrastructure
that we want to build.
And then we review it and then basically say,
this sounds great.
Let's give you some type of investment
to have you off and running.
And if you have any technical integration questions,
let us know and we can provide some support.
So more low touch, builder knows what they want to build.
And then there's, on the other hand,
the accelerator, which is 8 to 10 highly curated companies
that want a lot of support from us on all aspects,
go to market, fundraising, development, product market
fit type of questions.
And so that's a much, much higher touch aspect from us
where we bring our entire team from Zero G
as well as our partners like One Piece Labs
and Moonhill Capital to support these builders
so that we can make them as successful as possible.
And the first cohort, we had a number of companies
that went quite far already.
The one that was the furthest along was Carve.
And they actually launched on centralized exchanges,
like Upbit and Bybit and so on.
And when the market was still not trading sideways,
they were valued at above $1 billion.
So very proud of the team for being able to accomplish that.
We definitely want to support other builders on their path,
but with a much more kind of high touch experience.
That's interesting.
So you're kind of taking a bit of a hybrid approach there.
Interesting to note.
Letoj, do you want to get into maybe some of the tokenomics
and the token stuff?
Where do you want to take this?
Yeah, absolutely.
Look, I love talking about the long-term vision, the long-term focus, but as you said, the market set of things,
if it impacts anything, then it's more the tokenomics and TGE considerations. And since you've
recently did release or the ZeroG Foundation did release the tokenomics, I want to touch on
some parts of that. It seems there's been a bit of a discussion on the timeline going on after this preliminary token distribution has been revised a bit and the timeline has been focusing
on the community bucket side of things. So just maybe quickly taking a step back for our audience
to break down the tokenomics a bit more before I get into the question. There is a significant
ecosystem growth allocation of 28%. Almost half of that being unlocked at TGE.
There is an AI alignment note bucket of 15%, with a third of that being unlocked at TGE.
And then there is the revised 13% of the total allocation which are being reserved for the
community rewards. 20% of that being available at TGE, which is 2.6% then of the net supply.
And the rest will then be distributed
in seasons over 48 months.
And I wanted to give you a chance
to speak to the tokenomics and specifically to the thinking
and the reasoning that went into those three different community
allocation buckets.
Fundamentally, our thinking was that essentially,
community should deserve, first of all, the largest allocation.
And so that's why you're seeing the community bucket being
56% versus the team advisors, contributors, backers
bucket at 44%.
And oftentimes, with major L1s, you actually
end up seeing that the other bucket is significantly larger
because maybe the investors alone have 35%
and then the team is like 20% and so on.
But for us, we were able to design it in such a way
that there was a kind of community first mentality
from that.
So that was one design principle.
And we were very cautious around how we fundraised as well
from a lab entity standpoint.
We wanted to make sure we had a highly distributed cap table
so that no investor owns, for example, if it's a 20% allocation of one investor owns 10%, then
all of a sudden you're starting to see a little bit more of a centralization on the cap table.
So that was something that we didn't want as part of that. So then the next design principle was
one of the things we've observed is that there's a lot of,
so to speak, farming where test net is really hot.
People do all types of things to see
if they can farm some type of airdrop allocation.
But for us, what was in central is more that kind
of long-term support because our perspective is long-term.
We want to build a platform that solves
for major societal issues like aligning AI models.
And so we took a long term approach to that.
And so the way we designed it is to say, OK,
how do we create long term holders that care
about the network fundamentally?
And so our decentralization mechanism primarily
was the AI alignment note sale from that perspective.
Because we just fundamentally believe if you're an investor,
if you're a holder of this long-term token,
that you're going to create or make actions or support
with a long-term perspective as well.
So that's why you see a very large allocation for that,
15% for that kind of long-term perspective.
And then the community rewards is
more based around specific actions
that different end users can take,
whether they contribute in social campaigns
or other type of kind of DApp campaigns and test net
campaigns and so on, where we want
to make sure that they are also valued
and rewarded for their contributions. But really the biggest bucket there from an analog perspective
was the alignment node holders because of that long-term perspective. Then the ecosystem growth
perspective is used for things like the ecosystem program that I mentioned, the $88.8 million,
to support initiatives like that
long term so that we can attract and work with the best builders and the AI meets web three space
long term. Yeah, that makes a lot of sense to have this long term incentive alignment.
We only have a couple of minutes left in the interview, but I want to have a quick
follow-up question on the tokenomics that is on the community reward side of things. You said it,
you know, you're going to keep some for campaigns that will be activated in the future, it's the
48 months period that will be with a seasonal distribution. Is there anything you can share
on that? Will it mostly be campaigns? Will it follow because it's a seasonally distributed?
Will it follow sort of a season logic? What can we expect from that to the extent
that you can comment on that?
Yeah, we haven't released specific details around that.
And as we've kind of seen, new things start popping up.
For example, an NFT collection called One Gravity
was built.
So that was super fun.
So we want to make sure that as these things kind of arise,
that we have ways of kind of rewarding contributors to that.
And the seasonal aspect, again, is
the idea of making this a long-term bet for people.
So it's not just like, hey, I'm going to buy an NFT.
I get the reward.
And then, OK, I'm done.
I'm going to just sell the NFT.
But there should actually be useful ecosystem things
that are unlocked by having that particular NFT.
So that was the design principle around it.
And the seasons, I think, we'll figure out over time.
One extent, one of the things I've learned in graduate school
is if you design something with very clear timelines
versus having something that's more, what's the word that I'm
using for, not off the cuff, but that happens
on a non-aligned timeline, then there
is a sense of excitement around it as well.
So there's a momentary excited happiness.
So you need to make sure there's a long-term happiness
and a momentary excited happiness. And so that's that was the
design principle there too.
Yeah, I completely resonate with that. It's like even for many, for many
company, right? Like even for ourselves, we have, we have like yearly annual
goals, but then we break things down into quarters, and then things change,
things change from quarter to quarter, especially when you're in such a
volatile industry. And
this end of this quarter has been completely different to
maybe the end of the previous quarter. So big changes. And I
appreciate that kind of requirement for some
flexibility as you progress. But as as we said, we're kind of
drawing to a conclusion here. Michael just wanted to give you
the chance. I think we've had a really interesting conversation
we've kind of touched on product, touched on history,
touched on the vision, the funding, the objectives
that you guys really want to achieve.
Is there anything else that you'd
like to share that maybe we've missed,
anything you'd like to call out, spotlight, just
before we close out here?
Just reiterating this idea that we have a very long-term focus around making AI public
good and the type of world that we see is, well, with any new technology can have kind
of two dimensions.
One is abundance, the other is weaponizing that technology.
And so we very much want to be on the abundance side so that we can create a world where we potentially, as some of my
professors say, we might not have to work for a living
anymore, where we can truly choose, pursue our passions, and
be kind of our holistic self, our best selves. And so, you
know, what does a world like that look like? And that's
something that we're really excited about. And so, you know, what does a world like that look like? And that's something that we're really excited about.
And so any of the contributors to our network
and users and so on,
I want them to have a very long-term perspective
so that we can build this type of mission,
this type of vision together in the space.
And so we're really excited about kind of building
for the long-term.
Sounds very exciting.
We joke, well, I don't think we joke often.
We're being quite serious when we say one of the reasons
or one of the byproducts of creating so much content
is that we hope we have this humongous bank of content
so that we can train the AI
and create different versions of ourselves.
So we don't have to necessarily be running the show
every single day.
Our very intelligent AI counterparts
will be able to take over.
So if you have any builders on your chain
that are able to do that, we'd be delighted to hear.
Yeah, let's chat more about that.
There's some approaches.
In the meantime, if people want to follow along
on Michael's journey, his account
is at michaelh-undersh underscore zero g, sharing it on the
screen for the people who are watching in. And just a reminder, if you're new here, we do this
every single day, Monday to Friday, 7am Eastern time for one hour. We're live on X, X video,
YouTube and Apple and Spotify podcast too. Thanks so much for joining us, Michael. Appreciate your time. Have a great
day. Bye bye.
Thanks for having me. Definitely worth getting up super early. It's like 4 a.m. my time's
on. So it's been super fun.
Thanks man. Have a great day everyone. See you on Monday. Bye bye.
