Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
NEAR Foundation - Pioneering User-Owned AI
Recorded: March 26, 2025 Duration: 0:54:31Space Recording
Short Summary
NIR is advancing blockchain technology by integrating AI-driven agents and intent-based transactions, aiming to decentralize AI and enhance user control. Institutional interest in crypto is growing, and NIR's scalable infrastructure supports high user activity, indicating significant growth and adoption. The introduction of stateless validation and sharding models further improves blockchain performance, positioning NIR as a leader in innovative blockchain solutions.
Full Transcription
Hello, hello, everyone who are joining.
We're just going to wait a couple of minutes for more people to start.
And then we're going to start our Amy session.
Mark, Alex, I see you're already here.
How are you guys doing?
Doing well.
Excited for this.
Thanks for having us.
Great to have you, Mark.
Alex, how are you today?
Hey, Laura, GMGM.
Doing well.
Great to be here.
Awesome.
So who wants the hardest questions?
I'll say Mark.
He loves them.
Okay, Mark. So you'll get the fun ones.
I love these answers.
Alex is always the best-looking public speaker here we have near.
So normally it always gets directed to him.
Well, we'll see you who gets it today.
So, by the way, industry is going quite okay.
I see Bitcoin is pumping.
We have 87,000 right now.
How are your moods?
Bullish or bearish?
Can you guys hear me?
Or is my microphone?
Are you good?
Yes.
Okay.
Awesome.
So, bullish or bearish?
What's the mood today?
What's the vibe?
Always bullish.
Always bullish.
Just came back from New York from the Digital Asset Summit.
And definitely think that, you know, contrary to Twitter, where, you know, crypto-tweather sentiment might not be the best right now.
Speaking with institutions, they have actually been more positive and bullish on the space.
So I think it's definitely exciting times when more institutions come in and, yeah, definitely share the sentiment.
Okay, so Alex, you're bullish. What about you, Mark?
We can't hear you if you're speaking.
Mark, if you're speaking, I can't hear anything. No one can. So can you please check your microphone.
Yeah, still no. I would love to hear your sentiment, but unfortunately, nothing is happening.
Unless it's just me.
Alex?
No, I can't hear him either, but I just texted him.
I'm sure he'll sort it out in a second.
Awesome.
So you were in a digital asset summit in New York.
So all these Jews and our bullets, what are they talking about?
Give me some gossip.
What you heard there?
Well, a bunch of different things, right?
It was quite interesting.
Trump gave a speech.
Michael Silla was there.
So very, I would say...
Bitcoin heavy the conversations for most institutions, but there's definitely some interest, you know, around a lot of interest around tokenization, RWAs, and also more, you know, smaller institutional investors that are looking to invest into the space, more as portfolio diversification, but also looking at alternative chains right now beyond just Bitcoin and E's.
and seeing what they can do on that front.
So, yeah, it's been really interesting.
I met a couple of liquid funds, family offices, private banks, and larger asset managers.
And, yeah, I would say sentiment has never been that positive on the institutional side.
Awesome.
How was Trump speech?
I heard it was an eventful.
Yeah, I think I mean, I didn't go to the actual speech that he gave, but I just watched the recording.
I think it was like a five, six minutes speech, nothing groundbreaking, but still I think it was great to see a sitting president speak at a digital asset event.
So, yeah, shout out to the to the blockworks guys.
I think they've done a great job.
Conference overall was pretty great.
Oh, love the shame. I missed. Where are the next events you're heading to?
The next one is probably going to be, so I've been on the road now pretty much since beginning of the year.
So getting a bit tired from traveling.
But I think the next conference is going to be Token, 2049 in Dubai, which should be a good one.
So we're actually doing a lot of stuff there hosting our own events.
So definitely stay tuned.
And yeah, that's, I think, the next one for me.
And then there might be permissionless.
I might go to that one as well in New York.
And I'll see with Mark.
Okay.
Hello, Mark, are you here with us?
Mark, let's do one, two, three, mic check.
Can we hear you?
Well, we hear some sounds coming from you.
Can you guys hear me now?
Yes, you're finally back.
Testing, testing, are you okay?
Say something.
I want to make sure that when we start, you're here and you can contribute as well.
Mark?
Mark?
I think we're experiencing some issues with Twitter today.
I even see some issues with like attendees.
I only see, you know, a handful of attendees even to their shows that 600 people are here.
So please everyone, bear with us.
I hope as we go by with the AMA, it's going to get resolved.
But Alex, I think you're going to be the one man testing yourself to get all the questions.
Margaret is having some technical issues.
I'll try my best to replace Mark.
So yeah, why don't we kick it off in the meantime and I'm sure Mark, we'll sort it out.
Awesome. So I'm excited to see a lot of people here today. I see what we have 650 plus joining. Let me introduce myself. I'm Laura Cornelia and I'm Medinava. I'm chief ecosystem officer at Gate A.O as well as the principal at Gate Ventures, the venture arm of the exchange. And today we're talking about how agents are redefining, how users interact with blockchain.
and the broader internet, and NIR is building AI-powered blockchain infrastructure,
enabling on-chain automation and user-owned AI.
But what does it actually mean for crypto?
And how soon will AI agents be trading, managing wallet, and running daps for us?
And joining me for discussion today is Alex, head of capital markets, and Mark,
head of partner development, VC and investor relations at NIR Foundation.
So we'll be breaking down what's live, what's coming, and why NIR is betting on AI-powered blockchain.
So now I will let the speakers introduce themselves.
Mark, let's try with you to see how is your connection.
Mark, are you doing with us?
Can you guys hear me now?
Yes, we can.
Yep.
Good, good.
Finally.
I don't know why.
Okay.
So what did I miss?
So basically it's here time to shine to introduce yourself.
Hey, everyone, Mark here.
I'm the head of partner development VC and investment relations at NIR Foundation.
Been around for about two and a half, almost three years.
And I'm pretty excited to come talk with you guys, to share a little bit more about
near our ecosystem and all that we're developing.
Awesome.
Alex, why don't you introduce yourself as well?
Yeah, sure, happy to.
And well, first of all, thanks for having us.
Really great to be here.
So Alex here, currently based in Vienna in Austria.
I've now been with the New Foundation.
Also for roughly two and a half years,
fun fact, Mark and I actually join on the same day.
And at NIR, I've been wearing many different hats from, you know, starting with NFTs to consume an enterprise BD and now the institutional strategy.
So as you already mentioned, since last year, I lead the capital market strategy at the NIR Foundation.
So really anything that has to do with institutional investors and products.
And Mark and I also work very closely on all things, VCs and backer relations.
But originally I have a dreadful background.
So spend a couple of years in investment banking and asset management and also in private equity.
So guys, you're like a package deal. You come on the same day.
You come to all the spaces on the same day.
You probably even travel together.
Combo duo, right?
Yes, we're actually tired of each other.
So seeing each other way too often.
Well, maybe we can do some friendship therapy during this space.
But jokes aside, NIR has positioned itself as a leader in blockchain infrastructure
for, well, AI-driven agents and new intent-based transactions.
So before we jump in today's discussion...
Can you guys share more about your AI vision and how it fits into V3?
Mark, let's start with you.
And then Alex is going to be next.
Sure.
So I think if you really look at NIR from the beginning, we actually started as an AI company.
So it was in 2017, Alex and Ilya came out and was really developing and
obviously co-authoring kind of the paper, attention is all you need.
And what I find interesting is if you go from then until now, I think a lot has
changed in terms of what has evolved.
But I would say the vision of kind of having this decentralization and really figure
out where the synergies are is quite interesting.
So I would say in terms of...
kind of the how the AI vision fits within the Web3 side is that long term wise you know we
definitely don't want a centralized entity or company to really own us,
know everything about us, and really control us.
And I think that's kind of how that term comes along that we have what's called
user-owned AI.
And that really allows you to really have your own data ownership, more transparency
and monetization values that you guys can utilize from that.
And I guess the question before, because these are all terms was,
How does that work?
And, you know, what does that look like?
Right.
And I think what it looks like now, which is becoming more and more clear,
is that it's going to be an agentic world.
And in the future, you're going to have, you know, each of us are going to have tons of different agents executing different, I guess, novel daily tasks, whether it might be doing research, with helping you book your cars, your restaurants, or, you know, doing work and things like that.
And to do that, you know, they need to know things about you, right? And so I think that's kind of how.
We're viewing it is, you know, we definitely want to be that alternative that doesn't allow that centralize that you can control and decide what you want to give and what you should get and paid for back.
So I'll stop there and see what Alex wants to add to this.
Yeah, I think you've covered pretty much everything.
Yeah, as Mark said, really, we want to make near the home for user on the eye.
So basically putting power and ownership and back in the hands of the user, where people truly own, you know, their assets, data and power of governance.
This has been the division of near since the beginning.
And this is really contrary to the current environment, which is dominated by a few centralized entities.
So we really develop open source infra tools and products for this gigantic future that Mark has touched on.
Awesome. Very interesting.
I mean, everyone is talking right now that AI is the future, especially like agents or even robotics.
mixed with AI.
And we can all agree that AI is already
disrupting the way we entire with internet,
but centralized control means also like potential risks,
censorship, bias, and exploitation of user data.
Let's talk about the biggest challenges and dangers of AI remaining centralized.
And how does this interlation actually give user more control and, you know, do the care?
I remember very well when we had an interview with CISI and he was like, yeah, DFI is great.
But we cannot trust users not to lose their private keys and the, you know, email recovery
because otherwise a lot of people are just going to be locked out of the world.
So like, do people, A, care about decentralization?
And B, what are the biggest dangers if we still remain centralized?
You know, you can feel your vision to everyone who is listening in here.
And I will start with Alex.
Yeah, I think it's a really important point, right?
If you look at the current landscape, as we just discussed,
it's really controlled by a few centralized entities.
If you look at Open AI, DeepMind, and Tropic,
they're all moving very quickly.
They essentially control the data, the models.
And this really poses a couple of issues with regards to ownership,
security, and trust.
So right now when you're using any of these AI models,
you're basically giving all your data to some company which may or may not handle it carefully, right?
And so you have a few centralized companies that have opaque algorithms and potentially backdoor controls so they can filter favor certain content.
And we've also already seen some of these closed AI systems make mistakes with sensitive data.
So I think this becomes incredibly important when you're providing, you know, really sensitive data, such as your medical records, banking information, or any other sensitive data.
You wouldn't want this to leak or be used in another way that you didn't intend to, right?
So with our approach with user on the eye, your data is always yours.
And your AI essentially works for you and not the cooperation, right, without any sort of compromises on privacy or ownership.
So we really see decentralized AI as handing the power back to the user where it belongs.
And so users, you know, can really deploy agents and all their data stays in, in states private.
And you also get full transparency.
into how the AI works and you can verify it and everything is open source right so there's
obviously a bunch of benefits that that comes along with this from a security and verifiability standpoint
but also i think it it really opens up new revenue models it really you know restructures sort of
the economics of how AI works today, as it really benefits everyone equally, users, developers, and
contributors.
So I think users should really care about this, given that it's their data.
And I think we'll see a future where this becomes more more important as more data comes
on chain and we have more sensitive data that gets exposed to these AI models.
Thank you, Alex. Mark, after this concise and interesting answer, do you have anything else to add or Alex just stole the show?
No, I mean, I think he crushed it, but I think one thing that I love what you asked is the point of
you know, do will people care, right? And I think that's something that's always been very difficult
to answer, not even just within the AI piece, right, but also even just, you know, blockchain
and decentralization overall, right? I think generally speaking, people don't care until they,
it's too late and then they start caring, right? And they don't have the ability to kind of turn
back. And, you know, that's kind of the limitations and that's kind of what we see, right? I don't
know. Growing up, I loved the, uh,
Terminator movies and Skynet.
And that's kind of, you know, the extremities of what you just said, right?
Like, what's the, you know, most dangerous, biggest danger?
Well, obviously, I think there's been movies about it.
you know, how do we get, how do we not get there?
It's by making sure that we're in control of our own.
We put limitations and we can decide what we want to do with it, right?
So specifically for this question, I think it's really focused on the extreme cases of, you know, what's the difficulty with that or, or, or,
Especially right now with Web 2, you can already see it.
All of us are just scrolling, doom scrolling continuously.
But now can you imagine what it would be like if it was the next step
where it's not just videos and interest that they know,
but it's actually everything in your life and what would happen if it gets to the wrong hands?
And so, yeah, get scary.
Oh my god, let's not get scared. You didn't know the amount of fear I have when I watch these moves about AI going broke and crazy and like taking over. So now, for example, when I use chat, GPI, every time, like, please could you do that? Thank you very much for giving me this response. Like, like, Mark, don't fuel my fear even more.
I want to continue on something that Alex kind of mentioned is AI agents.
So I'm going to first kind of ask VC related question,
then we can kind of go towards the consumers as well.
So from consumer side, I get it, AI agents can help
in numerous ways, from like automated trading to, you know, helping make financial decisions,
especially if you're thinking about defy would-a-combo.
But for example, from PC perspective, we see so many AI agents being built every day,
but in the end, they're just chaty-pT wrappers.
So my question is, and that's going to be for you, Mark,
how do you see assist projects that are built on here?
Okay.
being innovative and different rather than just capitalizing on the fact that anything would AI written on it gets no more funding compared to Ton that was hot like a year ago.
Sorry, what was the question? Can you say that again?
Sure. So basically what I want on the server, like right now, you see a lot of projects are being built on near, are using AI more as a hot narrative to put in their marketing campaign because in then they're just chatyPD wrapper.
And VCs are getting all these pitches and they're realizing that there's nothing innovative, nothing unique.
So how can me, for example, as VCs,
looking at this project and fund this industry,
even though there is nothing that unique that's being built and we can fund.
Yeah, so I mean, I think that's pretty interesting.
If you look at the AIA narrative and VC funding across all,
like one thing that I'm realizing now is the fact that there's really,
The conclusion is the fact that everyone's still,
it's still early and everyone's still figuring it out, right?
I would say a past quarter or two,
everyone's talking about agents and how that's going to change.
But I mean, we're still so early and so novel from that standpoint.
And there's so many layers to kind of AI invest
investment, right? There's like the compute layer, the, you know, applications or the middleware or the data, right? And so each one of those specific points also has so much things you can focus on and develop from. And so I actually think it's quite difficult to really figure out, you know,
what position that you have, right?
And so in terms of your point of,
people were just adding on AI and,
and to help with the fundraise
or just doing novel rappers.
And I think at a certain point,
and I think that's now is no longer our funds,
getting that interest or getting interest into those types of projects.
I would say the market has slowed down in terms of funding.
And everyone's really observing to seeing kind of where developments go, right?
If there's very about, if there's actually value created,
as well as kind of what the thesis kind of turns to moving next.
So I do think right now it is a tough time.
Yeah.
to fundraise as well as if you're a fund, you know, if you look at a lot of the Web3 funds right now,
a lot of their vintages are kind of fully deployed and they're looking to fundraise for their
next vintage this year. And that's also kind of an issue that you're kind of seeing a lot of as well.
Oh, yeah, 100%. And a lot of funds are not doing the multipliers as they were doing the last few cycles. So that makes
Things still complicated, but you don't know how many pitches mark our cat with another AI agent every week.
So speaking about that, so I know your guy's AI system lets AI directly interact with on-chain services.
So if we're thinking about that, you know, all AI industry is still very early,
what would be the biggest use cases that we're still underestimating
and we can look into?
Alex, why don't I start with you?
Yeah, sure. So I think, yeah, as Mark said, it's quite early, right? So I think we haven't, we don't even know yet what the use cases could be in the future. But I think right now, if we look at our AI assistant, we really see, you know, agents evolve from just passive chatbots.
into more autonomous economic actors.
So I think the near AI system that you mentioned is a good example of that.
It's not just chatting with you, right?
It can actually execute on-chain tasks and coordinate with other services too.
So, you know, it can directly swap tokens, sign transactions, or, you know,
even buy something for you if you request it.
I think one of the, like I come from traditional finance, right?
So for me, like one of the coolest use cases is how can we use AI to personalize finances?
So people often think of trading bots or simple assistance, but imagine an agent that, you know,
can continuously optimize your your defy portfolio across chains and really become your autopilot for your personal finances.
So, you know, it can deploy across different chains into different, um,
yield generating protocols and rebalance that automatically based on your preferences.
So I think that will be quite a cool use case and we already see that to a large extent today.
So that's more on the Web 3 side.
But you know, our system can also execute transactions for you in Web 2.
I think this is really, really becomes interesting when you can now, you know, ask it to book flights or order takeout for you.
We've actually just demoed at East Denver how to buy headphones with your AI assistant.
And so how it works is you basically tell your assistant, hey, I want to shop for headphones.
And then your system will first respond directly based on the context that it has on you.
But let's say you give it more specific.
So you say, you know, bear it.
Bear market is maybe coming, so I need to save money and I don't have the money to buy expensive headphones.
So please find me cheaper headphones, right?
So what it will do now, the system will hit the discovery agent that can now route you to the best agent that is specialized in this task.
In the demo that we showed, it was an agent specialized in buying audio equipment.
And so basically that agent is called and replies back with a whole set of headphones that you can choose from.
So you simply pick it, add your details, and you can now pay with crypto and basically use the Amazon API to ship your headphones.
I think this is really the experience we're going after.
All of the stuff that I just mentioned really happens in the background.
So you don't even have to really know about this.
You only interface with your AI assistants.
And that will basically transact and execute all your intents at the back end without you
having to deal with that.
So I think this is really a big evolution where AI will, AI will move from being basically
isolated oracles of information to really become agents of action, right, that you can use
in your day-to-day life. I love this example. Mark, anything interesting you're seeing
that's built on New York that is actually, especially if it's something is helping to bring
more VEP2 users to the previous and then what we need right now is more mass adoption.
No, I'm
I actually pretty much killed it with all of them.
Maybe I'll just be a little bit more general in that what excites me is kind of what it looks like in the future in terms of what these, you know, the front end will look like, right?
Because right now, when you open your phone, you open up apps, right?
You have a specific app to help you book a car, a specific app to find a restaurant and things like that.
But in the future, it's no longer going to be like that.
It's going to be literally a single app that you get to access and really do everything, right?
And they know everything about you and they know what you like and what your favorite restaurant is.
It'll probably remind you that it's time to eat and say, hey, do you want to go to that restaurant again?
And so I think this is kind of the trend in terms of why, kind of earlier I mentioned,
it's going to be kind of like more of an agentic world and kind of figuring.
And they're going to be working and utilizing and talking with other things.
I guess agents to execute on the different aspects, whether it's to buy something, right?
Kind of now we're searching on Amazon and finding that product we like and then inputting our
data and credit card and then address to make that work.
But then future, it's more of, okay, you already know which wallet you want to use.
You already are able to utilize what you want to make that work.
So, you know, without giving more use cases, I think that's one thing that, you know,
I'm excited in terms of at least what our AI assistant could do is really help execute on that front,
which is what we demoed and showcased at Ytenver, which was agent executing and purchasing something off of Amazon and things like that.
Awesome. That's very interesting.
As now we're shifting towards the talk about the consumers.
We know that regulators, especially in Europe and other kinds as well,
are already questioning AI roles in data, privacy, governance, and potential economic disruption.
And with user-owned AI models on NIR, how do you see this conversation shifting
when it comes to AI ownership regulation and monetization?
Are you guys afraid that maybe within a couple of years, especially for talking about you,
the regulators are going to be too scared and will add unnecessary regulations on top of it,
just to make the adoption and expansion of the space much more slower.
And I will start with Alex.
It's a good question. I actually think that, you know, the user on the eye model that we follow is actually preferable by regulators. Just because today, if you look at the space, regulators worry about AI being in the hands of a few companies. They basically ask questions on how do these companies collect, and rightly so, they ask questions around how do they collect data, how algorithms make decisions who is
and who is really accountable when things go wrong, right?
So it's sort of a black box.
Whereas without model, I think, you know,
once it's when it's open source and deploy it on a public blockchain,
transparency is sort of built in from the get-go so anyone can now verify how the model works
where verify the data that went into training it and i would say this really makes it easier
to address concerns of of bias of fairness that some of these regulators may have
It's actually funny when I studied my, when I did my master's, my master thesis was on AI governance and regulation.
So this is something that I spend more time on than I would like to admit.
But I think it's a really important problem.
And so also I think in terms of privacy, NIR enables models to be trained or run in ways where,
you know, user data remains under user control. So basically assuring regulators that
Personal data isn't being exploited and data is being kept private or owned by the user who provided it.
So I really think that this plays into the hands of the regulators and they would actually prefer AI models to be used on the community owned where people get rewarded for the participation.
And so also monetization is more democratic and not being controlled by one centralized entity.
I'm not sure if I covered everything there, Mark, in case you want to add anything on.
No, I mean, what I love about regulation, especially about new tech, is the fact that
you know there's really no framework for it right i mean you kind of go to what happened in the past
precedents and you know for new things i think it's difficult and that's obviously what we've seen
just with blockchain overall right compared to securities blockchain compared to um you know
treating it as an asset and and comparing how that works and so i i'm sure there's going to be
some struggles of regulatory issues uh in the future for this and
I think it's clear that this industry, those in this industry as well, are willing to keep pushing and making sure what's the right decision and kind of pushing towards what they believe in.
And obviously, I think, you know, user-owned or user-controlled and not being controlled by a single entity is a huge mission and framework that, you know, everyone does agree on and really wants to make this more of a free-spirited kind of split as well.
Awesome. Thank you very much. Gentlemen, I want to go a bit deeper into the tech of Nero.
So I know that AI-powered apps need high scalability, but most blockchains face performer bottlenecks.
And I know you guys with 9-Chain 2.0 introduced stateless validation and a reworked sharding models to improve efficiency.
So, can you explain me like I'm 5? How does it work? And how do these upgrade specifically benefit AI applications?
And how does NIRS architecture different from traditional L1s? And please keep it basic, give it for everyone who is non-technical here in the audience to understand.
But I'm very, very curious to learn. Who is the bravest to start? Alex, Mark?
I mean, I'm happy to go first.
We, Mark, we need bone here.
But yeah, I can try to keep it simple.
So basically, yeah, we introduced nitrate 2.0 and that
To keep it really simple, it just boosts capacity of the network.
So if you think about AI agents, scalability is really critical for them because they
will make trillions of transactions, and so you need a network that can scale.
So basically, we introduced 92.0, which implemented stateless validation.
I'm not going into the technical details here,
but what's important to know is basically it improves single chart performance
and allows NIR to add more charts easily.
So we actually just upgraded from six to eight charts a couple of days ago.
And this led to, I think it's a 33% increase in overall network capacity.
And so it really helps us to, you know, add more charts more easily and also increase the
charts performance itself.
So in a nutshell, if you talk about nightshed-2-0, what you need to remember, it basically
makes near faster, more efficient and also more resilient.
And that means, or that I would say, is really critical.
for supporting AI agents that might generate an unprecedented on-chain activity.
NIR is actually already handling between 45 to think that the highest was 47 million monthly active users.
And we have between 3 to 4 million daily active users.
So we're already supporting quite a lot of transactions and active users.
We're the second most used chain after Solana right now.
And so, yeah, now with HR, we continue to improve capacity even more.
I hope that was that explained it in a simple way.
Yes.
Mark, do you want anything to add?
Yes.
Yeah, I didn't mean to not answer Alex on that one.
That was probably, I was trying to raise my hand.
Don't know how to hit the button.
You know, Alex was hoping for you to help him out, waiting for you to step in, and you left him stranded.
How could you?
How could I?
Well, luckily we have two charts here, Alex and Mark.
So here I am trying to add in support on the next one here.
So in addition, kind of to what Alex said is I think the interesting part is it
It's not just trying to increase the number of shards, right?
I think there's multiple ways we can improve this tech as well.
And so one way that I'll kind of give a simple example is that current block time is about 1.2 seconds.
And in our kind of next update, we're pushing to get to 400 milliseconds, right?
So it's kind of like a 3x improvement on that.
And thus, you'll see drastic improvement in terms of overall efficiency.
And, you know,
on the chain level and in layman's terms like i always likeness to comparing that this is actually
like kind of like fine tuning right sometimes you compare you know car engines and maybe like a v6 to
v8 but sometimes you know you have more of horsepower in that v6 why it's because it's more fine
tuned so the point i'm trying to make is that we haven't even really tried to improve in fine
tuning yet you know so thus far it's just been more of just increasing the number of shards and this is just
an example of to highlight how scalable this is and really there needs to be more demand which we're
working towards and which we believe there will be and when there is you know a trillion agents running
executing a trillion you know transactions that you know we would be able to handle it right near has no
you know has one or one hundred percent up time no never been down and so you know that's something that you know
I believe that kind of differentiates us just from that tech aspect as well.
And obviously, I don't know if you've been following,
but you also has a cool cross-chain execution layer.
We spent a lot of time last year really focusing on chain distraction.
And this was one thing that we, after doing two years of BD and kind of talking to projects,
talking to developers, talking to users,
We learned is that actually people don't care about Dane.
They really don't care what it's on.
They just want to make sure it's easy to use and it works.
That's similar to the fact of what servers you guys use.
If you're using Google Cloud or you guys are using Microsoft's there.
It's just the focus is more on if it works well and there's no issues.
And so,
In terms of the chain abstraction and this cross-chain execution layer,
we're pretty excited about this tech.
We do believe that it does differentiate and provides a lot of opportunity for,
just new defy primitives, natively utilizing,
borrow and lending on Bitcoin,
natively able to kind of swap and trade
different types of assets that it's not just EVM, right?
Or not just, you know, you're able to add things like Bitcoin
or Ripple or Doge, Zcash as well.
So I feel like it's a lot of,
it's pretty exciting in terms of all this developments
in terms of how this relates.
And how do this relate all into
to the future.
Obviously, I think we talked about the blockchain.
We then talked about the cross chain execution layer.
And next, it's really like, you know, as we described earlier in this space is the future is kind of more of a gentic executing.
cross-chain execution and things like that right and so that's kind of how these two
pillars kind of help and develop that the last obviously is what's near is developing in terms of
the AI stack and you know whether it might be shade agents that are private and secure
or kind of aggregating kind of the different agents together so that we we can help them
with the execution of the transaction things like that so i think the AI piece again is
still pretty early overall but you know pretty excited for all the developments that we can see
Awesome. And I mean, I think everyone here who are joining in are learning something.
And hello, hello for those who are just tuned in.
We are having a discussion together with two lovely gentlemen from Near Foundation about what they're building in AI agent industry.
We have Alex, head of capital markets, and Marg, head of partner development, VC, and investor relations.
If anyone here in the audience have any questions for the guys,
please feel free to write them down underneath this Twitter space,
and I'll be monitoring and asking them towards the end of this session.
And now continuing this a bit more techier route,
I know that NIR was also introducing intent-based transactions,
allowing AI to interact with multiple services seamlessly.
So how do these intents change the way blockchain works and why it is so important for Web 3 adoption?
Alex, let's start with you.
Yeah, sure.
I can kick us off there.
So, yeah, we've introduced near intense.
Basically, you can think about it as a fundamentally new way of interacting with blockchain.
And it's.
really at the heart of everything we built in the area as well.
So traditionally, if you wanted to do something complex on chain,
say, you know, trade one asset for another on a different chain
and then deposit it somewhere,
you'd have to do multiple transactions and possibly, you know,
use like several different apps manually.
So with near intense, you really flip that on its head
because you simply express
what outcome you want and then you let the network figure out the execution.
So in a way, like I always like to make that comparison, it's a bit like telling a self-driving
car your destination instead of steering through every turn yourself, right?
So with near intent, just to give you a concrete example, you might say, hey, I want to
swap 100 USDC on Eath to near and then use it to buy an NFT as one high level request,
And that intent gets then advertised and various solvers will offer ways to fulfill it.
So you no longer need to chuggle multiple transactions.
So even know which chain or service to use, the intent system abstracts that away.
So that's really at the core of everything we do.
And that is built on chain signatures, which we can talk about later.
But basically, the core idea is always you really want to abstract the way
the complex all these complex transactions and you know really solve for this fragmentation in in the blockchain landscape that really is what near intents are about and you can now also you know express intents for physical services so we you know use the example of you can buy pepperoni pizza and and then
have on-chain enforcement and dispute resolution if something goes wrong.
So I would really look at this as a fundamental shift in how, you know,
blockchain now become universal transaction layers across Web 2 and Web 3 rather than isolated ecosystems.
And there's a bunch of teams that are already integrating chain signatures and intense.
Probably the most prominent one is Infinex.
And they will integrate near chain signatures and your intents, which is very much aligned with the mission to, you know, abstract away UX challenges and all these complexities that are discussed in DFI.
So this will essentially allow them to support assets such as Bitcoin, XRP, DOS, etc.
It's just a high level overview of intents, and they will be obviously very, very important for AI agents because you want agents to be not just confined to one single ecosystem, but, you know, transact across different chains and transact across Web 2 and Web 3.
Thank you, Alex, Mark. Is there anything you'd love to add? Mark, are you here with us?
Sorry, I was on mute.
No, I think he pretty much covered a lot of the intense side,
and I kind of talked earlier in terms of, I guess,
how the applications would be with the intents as well, right?
So I think what's maybe just one thing to note is that it's not only going to be utilizing, I guess,
Web3 assets or, you know,
tokens or things like that, right?
I think in the future, it's going to be focused on a lot of the on and off ramps as well as kind of like the, you know, backwards as well, right?
Not just how do I go towards and calling actions to the back to the Fiat rail's, but you're going to see rails going back and forth on this.
So I think that's what's pretty exciting.
Thank you, Mark. You know, I'm not a developer, but I'm already thinking like, hmm, maybe I should
build something on near. So if we have any builders, founders, founders that are considering
utilizing your new AI technology, what are the resources they could use, what kind of support
you can provide as a foundation? Can you just walk it through everything you have going on to support
support the developers thinking of building something on the air?
No, absolutely.
So we have a whole team called Founder Success that really, you know,
has focused a lot of different types of campaigns,
activities that really could strive to help support them, right?
Obviously, I think there's the self-serve in terms of, you know, docs and,
but outside and we also have a kind of a dev hub to help support any questions
in terms of the developer issues.
Outside of that, we also run tons of hackathons.
We had an accelerator cohort.
Last, I think one was in Q3, Q4 with Delphi Labs.
And so I think we're pretty active in working towards finding great builders
and trying to empower them to really build that content and see what they're looking for.
Awesome.
As we're approaching the last 10 minutes of our session,
I'll have two questions for each one of you guys, and these are going to be the hardest questions that I always ask towards the end.
And everyone who is in my space be like, oh, wow, I don't know how to respond.
So who wants to start?
Alex or Mark?
I'll go first just since Alex needs a little bit more rest over here.
Okay, Mark, if you had to describe NIR and its new AIVision in three words, what would they be?
So the question is, if we had to describe NIR's AI vision in three words?
Yes. What would they be?
Well...
I think it would be what user owned and near because that's hopefully when we're going to get all this access point and when it's going to be here for us to use.
Awesome, Alex. How would you describe just near in three words? What would they be?
I mean, I would also say like on the I said you owned in terms of the protocol, I would probably say,
user-friendly that counts as one word by the way then I would say scalable and yeah
let's say those are the three I would say user user-owned and I use a
user-friendly and scalable as the as the protocol and maybe we can say Charlotte
but that's the same so
I love it. Okay. Next question. Mark, if you could onboard any celebrity or industry leader into Web 3 using NIR, who would it be and why?
That's a tough one.
I'm telling you, I left the hardest questions for the last.
I think it would have to be Elon. I mean, we're on X. I think there's so much that we could do.
Because he's so extreme, you know, that there's just so much, he's just so active right now.
He's got so much going on. And he's touching so many things. I think having him
I guess, be familiar, knowledgeable, and to the point where he actually believes in this more of the decentralized aspect, I think, I mean, I kind of think he does because he was looking to kind of gain a little bit more freedom for open AI. So I think that's kind of alignment, right?
I love it. Alex, what about you? What would be the celebrity or industry leader?
Good question.
Okay.
It wouldn't be a celebrity. I would probably like to bring Jan Lee Kuhn, who leads AI at
Meadow. I think that would be super interesting to get his thoughts and get him more involved in the
near ecosystem.
I love it. I mean, throughout this space, I always ask these two questions at the end,
and I got answers like Britney Spears, Future, the fart of AI.
Now we have Elon Musk. I love how diverse these answers are.
So with this, that's a wrap from my end.
Is there anything else, gently you would love to add or share,
as we have quite a large audience here with like 700 people listening in?
No, really appreciate the time and everyone listening in.
I hope this was exciting and interesting and you guys learn a little bit more.
If there's never anything that you guys want to talk a little bit more,
feel free to reach out, me or Alex.
Yes, thank you very much.
Mark, for being here, Alex.
Yeah, no, I mean, maybe just to iterate, I mean, we, you know, we're really heads down building a lot of that stack that we touched on and maybe just to summarize it, how it all comes together.
You know, we have the underlying everything is the blockchain, the highly scalable layer one blockchain that already handles millions of daily transactions and users.
They would continue to scale through sharding to basically support all these trillions of agent transactions.
Then we have the chain abstraction tech, which Mark already discussed a lot, basically the combination of chain signatures and intents, which are talked about.
And then lastly, you know, we have to use on the I stack.
So like including the trusted execution environment, infra, basically allowing everything to run in a verifiable confidential way.
So this is this is how the stack all comes together.
And yeah, if you would love to learn more, as Mark mentioned, feel free to reach out to both of us or anyone from near.
And yeah, we'd love to talk more.
And thanks for having us.
Awesome, thank you, Jen, for being here.
We as gate ventures are also very actively right now looking for interesting projects to invest in.
So if there is any builder who is building on here, please feel free to reach out to us as well.
Our DMs are open.
And basically, thank you, everyone for coming in.
I probably, guys, going to see you in Dubai during token 24-29.
in person and everyone who joined in.
Thank you for being here.
We've got to be hosting these Twitter spaces every week.
There's going to be one tomorrow.
So please join that one.
And with that, thank you and please have a good day.
Thank you, everyone.
