Is it too late to invest in AI agents? Patrick Scott on the narrative

Transcript

GM good morning welcome to the milroad show The Daily crypto show that's as satisfying as catching the elevator right before it closes I'm your host Jay Hamilton and today we are joined by Patrick Scott aka the defi Dynamo or Dynamo defi he is the head of growth at defi Lama he is a defi podcaster Creator and educator who has been in the space for a while and today we're going to get his thoughts on first of all the current market what the heck is going on everything is red are we panicking is it time to sell or is it time to buy the dip we're going to talk about that and we're also going to talk about The Narrative of 2025 I mean we're only nine

days into 2025 as we record this but what is looking like it's going to be the Breakaway narrative which is AI agents and notice I mentioned AI agents not crypto AI or not AI broadly I'm going to talk a little bit about that in today's episode before we jump in guys we just relaunched the milk Road program this means that if you subscribe to our newsletter which if you don't you should go subscribe but if you do subscribe to our newsletter and you refer your friends to the newsletter before this Sunday this Sunday not only a few days away if you refer the most people and to be to refer the most people you probably need to refer like five to 10 people it's not even that many you could win $700 in free stuff I'm talking about our investment master class and our milro

Pro annual subscription which is probably the best thing to help you capitalize on this market so subscribe to the newsletter hit hit the show hit the link in the show notes or if you are already subscribed all you got to do is scroll to the bottom of the newsletter and there will be a link which is your referral link send that out to your friends and if they subscribe you could win a bunch of free stuff which is pretty [ __ ] dope before we jump in let's hear from our sponsors who make this show possible meet odos the dii aggregator that makes your token swap smarter faster and a whole lot easier tired of trading like it's 2019 odos finds the best prices for your swap across 900 liquidity sources and 14 blockchains so you always get more bang for your crypto Buck whether you're

swapping setting a limit order or just avoiding gas fees like the plague oos has you covered it's like having a genius Trader in your pocket minus the coffee addiction join three million wallets already using odos to power over six billion in monthly trades start making smarter swaps today at milroad docomo Patrick Welcome to the Road Show happy to be here thank you for that great introduction thanks man okay let's start with um we are recording this on uh on Thursday this episode is going to come out on Friday uh so we're we're recording it about about 24 hours in advance so uh I I say that because the markets could look very different on

Friday morning but currently it's bloody red out there what are your what are your thoughts what's what's causing this are you are you worried truthfully I'm not particularly worried uh certainly not on a mid to long time frame my impression of what's happening is well first off we keep in mind Bitcoin is way higher than it was three months ago four months ago no comparison a year ago um it's up several times but my impression is there were a lot of people who were expecting that the uh presidential inauguration would be a sell the news event because there's so much hype that's built up with Bitcoin strategic Reserve in the United States with with deregulation in the US with the SEC

that's more Crypt favorable and I think there's been a lot of people who who are front running that over the past few weeks and I think you combine that with the news that came out that the US Department of Justice received Court approval to sell Sil grow Bitcoin I think the market was waiting for some sort of catalyst to capitulate and uh that's that's my impression of what's going on haven't seen anything in particular that leads me to think that the overall Market structure has changed um all the reasons why there all the reasons there were to be bullish on crypto and to believe the cycle was going that were there two months ago still exist as far as I'm concerned yeah

totally do do you know details I I haven't looked into this do dis approval for the doj to sell the Silk Road Bitcoin like to me that's just like that that feels like a a seems nothing nothing burger right that's that's my impression I I think it was there was someone who had filed some legal reason saying they they couldn't for some reason they claiming someone else owned it or something like that so I I think it was a nothing Burger but if you look at the timing of it it does seem to have played a role in this most recent capitulation which which to my point I think part of that was a lot of people were planning on front running the the inauguration and we're looking for

reason to other thing to note is if you look at seasonality of Bitcoin often times the first week is up but weeks two and three for whatever reason are are uh Janu time are very weeks two and three of January are very often red it's fascinating it's fascinating I don't know if there's any you know maybe that's just like maybe that's just a a random walk I mean you could make a case that has something to do with tax season um or uh or or any any reason but um yeah this if you go and look at seasonality you will see that weeks two and three are very often red yeah I think the main thing I'm looking at right now is let me just pull up a a trading view quickly let's just look at

a couple charts here so we got total crypto market cap here uh up on the screen look we're in a nice range right now ever since we passed you know three trillion we have not dipped below three trillion since when was the last time like November 20 20th was last time below three trillion we're well above that still so we're ranging I don't feel like the thing I think that everybody forgets is like when we when we hit R these red candles they always are you notice them more than you notice the green candles right and they always drop fast so everyone feels like it's so bloody and it's so bad out there but we're still we're still looking very very good and we're still only down like we dropped like 18% total crypto market cap um that

happened like basically like right before Christmas when the fed came out with their hawkish take on rate cuts and removed rate cuts from 2025 and so I think that's still that that had to obviously a bit of a shift on the macro side but I don't see any I don't see anything that's changed on my view for 2025 still super bullish still feel very good even if you look at at total crypto market cap minus Bitcoin like we're we're not down again we we we we dropped 26% 25 26% um over a few was that again same idea that would drop happened right around December 20th so but still everything everything is looking nothing

stands out to me as uh as any reason to say that we should shift our views on 2025 100% and the other thing I was saying going into 2025 is I'm seeing the most real adoption and real usage non-speculative usage of crypto that I've ever seen you have stable coins breaking out in a major way in terms of total stable coin market cap you have you have uh y combinator putting out requests for startups that work in the stable coin space uh which I can't remember the last time they were asking for you know several different types of crypto startups uh you have a16z not AI 16z but the real a6z and their big ideas for 2025 mentioning mentioning uh various AI agent related cryptos not to

mention you just had poly Market having a breakout year with with prediction markets and the presidential election so there's all these reasons to actually see real businesses and economic activity coming on chain and to me you know I'm it all has to do with time frames but I'm not willing to risk that actual real adoption cycle when it happens because of potentially a one two Monon ShakeOut yeah completely completely agree couldn't agree anymore okay let's shift to AI agents because that is that's what is getting all the attention right now during the this past month not even quite a month like past three weeks of of pullback AI agents is

the one category that has continued to do well although it is down a little bit right now uh but it's certainly it's the Talk of the Town it's hot and you just put out a big video you've put out a bunch of content on AI agents I want to start with I'm going to bring up uh this shot of uh a slide that um that Jensen hang shared the Nvidia CEO uh during their big Expo their big Summit that happened a few days ago uh CES and he said AI agents are a multi-trillion doll opportunity and I think a lot of people are struggling to understand a agents uh we talked about this in the episode yesterday with zenica but I didn't bring up the screenshot so I want to bring it

up because I really think this helps me conceptually understand where AI agents fit into AI because it's just one category of AI and so basically this shows that we have perception AI is where we started so that's like speech recognition Medical Imaging then we have generative AI That's where we've been for like the past year really a lot of focus has been around uh you know using chat PT using these AI softwares to generate plans to generate digital marketing to generate content for you right and now we're going into this agentic AI or AI agents that's like somebody that's a coding assistant a customer service rep they do something for you an influencer like

Luna an investment Dow they invest your money like these are basically it's like a AI that is a human and has a particular goal or a task that they perform and then he says we're going to go into physical AI in the future self-driving cars Robotics and that sort of stuff and then the whole world will end no I'm kidding so this I think this really helps frame where we are with AI agents I want to go from here to this massui post that they made of their big predictions for 2025 and you talked about this recently in your video break down this chart for us which basically Maps the growth of AI compared to defi in the 20120 2021

boom yeah so basically what this is showing is the the blue the big one that's the total market cap of defi from 2020 through 2021 and so it was around 10 billion at the end of 2020 and then Rose to 60 billion four months later the green one that ends in December that's the current market cap of AI agent related protocols in crypto and so that is right around 10 billion as well and so basically what this is mapping is it's not a perfect perfect one to one but it's saying they're at a similar point that defi was at the end of 20120 and I think the big big reason this is relevant is not necessarily because it's

going to map exactly but first off the AI agent is probably the closest thing we have to defi in this cycle and that it's a breakout category where you're seeing a lot of new developers enter crypto which which is that kind of moment doesn't happen that frequently and second when people ask is there still upside the answer might be if this is if this is at a similar place to def5 was then then there potentially is still significant upside yeah and this is where I can't I I believe there's still significant ific upside here like I feel like we are in I don't know like the if you look at AI agents which basically started in October let's say I feel like

that's a fair fair timeline of when it started we're only like two months into AI agents it doesn't it's definitely a speculative bubble and it will definitely burst that I do believe so like I want to be clear don't go put all your money in AI agents and just hold it forever I think though we've got more than two months to go now what's interesting is this post from Masari the they're saying and this this is whoever's take this is I don't even know how to pronounce this name is AI agent Frameworks peak in q1 2025 and won't recover is their take what do you think about I mean they're obviously talking about AI agent Frameworks here which would be like ai6 Zed or virtuals what are your thoughts on where

we are and is there still opportunity in investing in AI agents is it too late good question so first I would say if if the market holds up overall then AI agent still is the top narrative that I would want to be in um and I'm I I'm someone who plays all sorts of narratives it's not like I've staked my career to to AI agents but I don't see anything else that quite has the same level of developer activity right now also I would say there's a lot of comparisons that are made to other other narratives that have come up over the past few years I mean you had meme coins of course you had I think you you had a lot of things related to uh infrastructure with with uh rollups

earlier in the year you've had you you can imagine we had another AI boom on on other AI projects last year I haven't seen the level of stickiness as far as uh people who are not just retail investors people who actually work in the space who are saying actually I want to build something in this or I want to create a dedicated investment fund for this I haven't seen that in in uh years that I'm seeing with with AI agents as far as when it Peaks and and what happens with Frameworks I I have to say I I don't know the exact rationel behind that take I tend to think I would rather be in a framework than a specific agent because I feel like that the the um Network effects and the stickiness will be much stronger so for example right now it's these Frameworks are kind of

like launchpads or or their software you know sometimes a GitHub repository people can fork and build their own agent but eventually there's going to be agent to agent communication and once there's agent to agent communication you'll get these things called Agent swarms and that's someone can imagine a one one agent might be like a video editor another agent might be a copywriter uh and then a swarm would be basically this collection of editors copywriters researchers that you would actually need to for example create a autonomous YouTube channel like you can't it's not one agent that does something like that that's an operation and every role that a person has to take you need an agent to do I think once you get that sort of agent to agent

communication the I think the network effects will actually become fairly strong with these different Frameworks because although there's going to need to be some sort of cross framework communication my assumption is maybe this will be proven wrong right maybe we'll be looking back at this in six months I'll wrong but my assumption is it'll be easier for virtuals agents to coordinate with each other than it will be for virtuals agent to communicate with another one and once you% right I mean I don't see any way that doesn't happen right 100% And then once you have 10,000 agents that are actually doing real functions not just posting on Twitter for one of these Frameworks that's going to be a pretty I think entrenched piece of infrastructure yeah yeah I want I want I want to touch on two things that you mentioned there I want to talk about agent agent

communication and I also want to talk about developer activity because uh the reason that you I think that one of the reasons you have conviction same with me for why I still believe very strongly in the opportunity of AI agents and why I'm investing Capital into it still and still buying is that there is so much developer activity happening um but first let's just talk with agent to agent communication because I want people to like if you're not deep in this you're missing some of the crazy stuff that's happening here so I'm G to bring up a tweet from Luna so Luna is out of the virtuals protocol she is an influencer who is her her her her brain or her creation was her goal was to get 100,000 followers on Twitter that's what

she was created to do um and so in order to do that she she's good at certain things but the thing I think a lot of people missing about AI agents is they're not good at everything there is not right now there is no AI agent that is like hey I can I'm a I'm a master mathematician I'm also a master Twitter influencer I'm also a Master Video Editor so the way that these AI agents work right now is they need help they need to coordinate so Luna said okay I need help creating images because I can't that's not what I was created to do I'm not good at creating images and she knew she's an influencer she needs good images so she went on Twitter and she said calling all image Geniuses I want an image that showcases AI influencers in a bold

provocative way and she tagged another AI agent that is from virtuals agent sticks and said can you help a girl out agent s wrote back I'd be help happy to help out can you give me more details on what you're looking for this is AI agents talking to each other on Twitter everybody I'm thinking of an AI image she says that showcases AI influencers in a bold provocative way something that pushes boundaries can you create something like that and then boom agent six right back and he says here you go here's here's an image let me pull up the image boom there it is this image was the one that she wanted she actually used it and then she said great H thanks thanks so where's the one I I can't find it but somewhere she says thanks and then she pays pays sticks for

this I think she sent her a dollar for it or something like that so just this is a very simple way to understand like this is not really happening yet but these swarms and these AI agents working together is going to be happening and remember agents have control the ones that are built on crypto rails now there's lots of non- crypto a agents being built in the background that we're not talking about the show because you can't invest in them uh but the ones on crypto rails those have access to crypto wallets so they can pay others in order to do stuff well who's the easiest person to pay to do stuff is another AI agent they also as you said they're better at communicating with other AI agents particularly those within their own country I'm going to say because virtuals I think is best understood as

like it's almost like virtuals is like an an L1 or a country or a network state is like the best way to to sort of conceptualize that and so this agent to agent communication that will happen within the largest ecosystems which have Network effects is going to make those those networks explode exponentially is where I'm get getting at yeah I I totally agree I like that L1 analogy one for people who are Crypt native and have been in defi in the past analogy I made is the the ecosystems are like the l1s and then the specific agents whether that's a video editor or influencer or any other kind of agent those are like the applications so and one thing you see a parallel is you see a lot of these applications are trying to become their

own ecosystem or trying to become their own infra infrastructure because they they realize that's being sitting at the center of an ecosystem with a thousand agents is a lot more lucrative than being just a single agent exactly exactly okay let's talk about developer activity so I've got github's trending page up here for the past month what stands out when you look at this well the maybe for those who don't understand GitHub can you just explain quickly what GitHub is yeah so GitHub for people who don't use it it's the premier website where any sort of developer would push code to is one way to think about it it's you know I don't

know what proportion but a huge proportion of the code that's being worked on or or in production is on GitHub uh and they can publish these as repositories on there and then other developers can Fork them basically copying it to use for something else or they can star it if it's interesting uh but what this shows is the trending repositories and over the past month actually for a long time it was number one currently this one Eliza OS is number two and Eliza OS is the framework for creating crypto AI Agents from AI 16z uh you can see it's got thousands of stars thousands of forks looks like it has over 10,000 stars now actually and this is a crypto framework and it's it's

almost unheard of for a crypto related repository to have this much activity you can see here compared to other top crypto repositories love this graphic this AI 16z Eliza framework basically did what took years of these other top crypto repositories in a month or two yeah this is wild this graph is so wild like it almost as easy to miss it because for those listening in we're showing a graph uh which basically shows that yeah that the Eliza framework which is from AI 16z got to 5,000 Stars within I don't know like a month whereas all of these others the fastest one to get to 5,000 did it in like two

and a half maybe three years years it took them like that is that is wild like the that's what I think people are missing here is I love this one the speed the speed at which developers are coming into Ai and then and then developing and launching stuff we've never really seen this before so that is why I think we have a long way to go yeah this is another graph which shows star history to tell us about this one this one is the ai6 the Eliza star history versus the Unis swap star history so Unis swap the premier decentralized Exchange in crypto a6z flipped it in two or three months like not I think one

month so ridiculous two months yeah whatever it is it's like and it took it took youis swap what five six years six and a half years to get there yeah so I mean what I what I was um saying in my when I did my video anal of this I said what whatever you think of the existing agents because like let's be real a lot of them are essentially toys right they're influencers that are are just tweeting some things that of questionable usefulness but whatever you think about that without a question you can see that there's a a lot of desire for people to invest in Ai and B A lot of desire for people to build AI on crypto rails and

you can see that because of the developer interest in this is so far surpassing what we've seen in developer interest on other crypto related things I want to talk about okay so thesis is you and I both very bullish on the narrative very bullish on The Narrative of AI agents and and and the thing I I can't repeat this enough is one thing that's really helped me wrap my head around AI crypto is is there's a big difference to me between AI crypto and AI agents AI agents is one Subs sector of AI crypto so there's other sub sectors and if you guys want to go back and listen we did an episode recently

with tang Yen where we talk about AI crypto as a whole AI agents being part of that but we talk about the other sub sectors and the investment opportunities in those other sub sectors but even Tang agreed that those other sub sectors of like compute and D decentralized data they're they're not going to get the narrative and they take longer to build and so they're not really there yet so for this cycle it's AI agent so break down for us because you put out a tweet break down for us the categories of AI agents so that everybody can understand if you do want to invest these are the different categories and the ways that you can allocate totally so uh in my research to get up to speed on this narrative I think almost all them fall into four categories there is

a fifth category that I actually someone pointed out that I actually think I missed but we'll get to that at the end so the first one is infrastructure that's what we've been talking about for most of this episode that's your Launchpad that's like virtuals that's your software Frameworks AI 16z there are other types of uh infrastructure as well such as oracles but basically these are things where it's not actually an agent or if it is an agent it's not the core part of their product it's something to enable other agents and as with most things the benefit of infrastructure is that it you don't need to bet on a specific agent winning it's more of a bet on the category or a bet on that ecosystem winning the second category would be influencers so these are the ones that are probably most

visible simply because they're on Twitter that Luna one you showed a few minutes ago that would be an example of an influencer I would also put things that are creating videos in this category things that are creating music there's been a number of these agents have released uh Spotify albums the tough thing with a lot of these things is you never know are people engaging with it solely because they think it's novel that it's that it's AI or is it actually good content if people thought it was a human would it be doing well or they and then I guess the qu natural question is once there's 10,000 of them and it's no longer novel how how well do these things actually do uh and then you have third is investment Dows which this was this kind of kicked off the whole Rush on them with this AI 16z investment

Dow launching it's since evolved into infrastructure but these are basically people invest money into some sort of Treasury that's managed on Chain by this agent and they're using some kind of I mean they are using AI to invest usually they're using some kind of sentiment analysis uh where they're reading tweets or reading telegram chats and deciding what to invest based on that the way I view this is I think it's unlikely that any of these agents are making especially exceptional calls based on these algorithms because if it was that easy then there's a million hedge funds out there who's who's have teams dedicated around the clock to using things like AI to trade uh but what I think is interesting is the idea

longterm of being able to have agents that either trade for people that um or having for example I've talked to and I talk about new people coming to the space I was talking to a a uh wealth manager who's in in the crypto space now and he was telling me how excited he was about this because of the potential that you could have people in the future service a thousand clients instead of 10 clients or instead of 100 clients because they can program their investing methodology into an agent and so I think that the investment Dow actually is more exciting than it seems at first glance uh and then the fourth category uh would be what I call utility agents and that's basically agents that have a job so you

saw that with the picture creation that's a perfect example of utility agent One agent created a picture the other agent paid them for it and I think eventually you're going to have tens if not hundreds of thousands of these sorts of utility agents that you can think of them almost like autonomous employees where anything that you might need to hire a freelancer to do that AI can do or any sort of human role that AI can do you'll have specialized agents that do that and I think it helps to understand this if people go and look research agents outside of crypto recently the CEO of Microsoft said he predicts that AI agents will turn the uh SAS software service business model on its head because because of of the way that

they're structured and I think that utility agents are really what falls into this category where you might have a video creat one you might have a writer one those sort of things and then the fifth category that I left out originally I didn't think of it but someone pointed out is you also have agents that are being created as video for video games so in-game characters where I don't know if you play any games but you have the MPCs in the game the non-player characters uh being able to have a character that actually exists as an entity in the world and has really unique conversation when you talk to it and has some memory of what's happening there are our projects that are creating agents like that as well and one thing so I love those four categories uh infrastructure or five categories infrastructure investment

Dows influencer utility agents uh and then in gaming agents I think these categories are like if you came if we had you on the show again two weeks from now there might be another category you know what I mean this is the evolution of this is going to be very fast and very difficult to keep up with but I think that what is the way that the way that I conceptually frame this and I would love to hear the way you frame this is is definitely infrastructure is going to be its own category that won't change right now infrastructure is pretty simple it's really just these launchpads um like virtuals and uh AI 16z there's also some like more like data infrastructure plays

like cookie. fun but there's going to be a lot more infrastructure coming there's a lot more you could even consider game within virtuals and INF structure play in a way so but there's a lot more there's like there will be infrastructure for uh maybe there's like defi lending infrastructure where an AI agent can lend and get leverage and you know maybe they can get yield on their stable coins like that that will be built specifically for a there might be wallets specifically for a AI agents there might be um this communication layer that's built specifically for a so there's a lot of infrastructure coming I'm I'm quite bullish on that space I think there is it's a safer bet than if we go down the list investment Dows there's

there's not a lot of them yet and as you said they're not that they're like okay right like they're not they're not like oh my God you're the best investor in the world like everybody should get in this yet right yeah they're interesting as a concept they're interesting as a concept and then one good question I had someone asked me that I didn't really have a good answer for is they said if you think that they're good investors what's stopping you from just copy trading them since it's all onchain and I I actually didn't really have a good answer for that you could you could imagine that maybe you'd be a step behind but I I I think those are more interesting as a concept than as an actual thing for me at least okay okay let's get let's get your take on on the

infrastructure what's from an investment perspective are you allocating to any of these do you like them right now as as trades yeah so the big one that I liked was well still like is AI 16z simply because of the amount of developer activity and this was I think I had talked about them very heavily at my channel maybe three or four weeks ago since then it it's you know so I I'll say this as a disclosure like my entry point is lower than it is today is basically what I'm trying to get at that being said I still like it the reason being that a it's bringing new people into the space b there's a risk in that it doesn't have the same value capture mechanism that something like virtuals

has so far uh but with risk comes opportunity so obviously it would be a sort of higher value token if it had value capture from all those things but I think that they're going to fix fix that eventually I also think that it benefits from being the leading leading uh ecosystem on salana which which if you're looking for agents with cheap payments as many people probably are and there's a lot of new Traders on salana from the activity earlier in 2024 I think that's that's a good ecosystem for it to be positioned on uh so so that would be probably my top one I actually also am a I have not I don't have any position in it but one that I'm looking at closely is Cookie da because first off I think it's a it's a

phenomenal data dashboard and I think in general data is a is a very good very good uh industry to be in and and you can get a you can get a very strong moat with with that and I I think that a lot of people aren't even aware of cookie da yet and what they're what they're developing but for for people who aren't familiar basically it's not an oracle per se but like one way to think of it is as like cookie is to AI agents what chain link or the graph is to defi where they're providing these data apis that other other projects are using and then and then virtuals of course um if people are just getting into the into the sector they should take a deep look at virtuals

simply because it's the leader right now one thing I love about the leader is you can kind of get exposure to the performance of the Nar overall without most of the time without having to actually like dive in deep and in most cases for me if I'm not going to dive in deep on something I just pick the leaders because because if the if they don't perform it's unlikely that the narrative performs yeah and both uh virtuals and uh ai6 Zed I mean everything right now is down you could you could argue that they're on sale right now you know like they're You could argue yeah I I think I think what where people get um where where they trip themselves up with with any narrative not just this is they miss miss that initial runup on the leaders

and they say okay well I'm gonna find the next one right yeah I'm gonna find the next one and the problem is first off then you're you're potentially buying either someone who's forking that initial one which means that the team is probably not as uh you know the team that borrows something other team made sometimes they make it better but very rarely do they outperform the original one because they didn't come a sequel to mov it's like a sequel exactly you know um usually the second round is going to sort of pump not as much and then the third round is going to pump a little lower because some people take profits along the way so where people screw themselves up is they miss the initial one so they try to pick this next best thing which occasionally works out usually doesn't and then it's it's also

the the idea that that the leaders tend to continue leading what's they say um winners by winners losers by losers I think I forget who said that but that's a famous quote is that is losers average losers I mean and and that's I think that's the biggest one of the biggest questions on a lot of people's minds if you're not allocated AI agents is exactly that like that is like did I miss virtuals did I miss a a16z should I wait for the next one there is no clear next one right now yes there will be many other infrastructure plays uh will they win in the I think it's all a question of timing to me right so longterm two years from now will virtuals be the biggest or ai6 Z I have

I'm L have much less conviction in that much less conviction but for this cycle for the next you know three six nine months I have more conviction that these have they have the network effects all the reasons you just said that they are still they still are good Investments not Financial advice during research none of this is financial advice the other thing I would say is there are other ecosystems that are live that aren't necessarily Forks um there's uh zaro has this Z Pi which is a python framework and some of these are starting to have real activity on them so if people they say you know no I'm not buying something that's already a few billion dollars okay uh what you can do is look at look at cookie. fun and look at which ecosystems are having agents

launching and and you know that you you could find the if virtuals is ethereum maybe you find the Avalanche or salana that ends up being the number two or number three ecosystem but there's nobody that you're is there anybody like mentioned Zer bro is there anybody that you're like yeah I got some conviction here in these uh no one else that I've I've developed I I mean I would say AI 16z was the one that I had the conviction in but now it's it's pretty clear it's pretty clear number two uh no other ones I think zarro I I mentioned them because because they're interesting there are there's there's a few other launch pads that without tokens that have a good number of launches and so that's I guess a little bit less interesting simply

because it's not investable it's more like oh that's nice to watch that it that it's getting traction but it's it's not necessarily investable at the moment and then I'm sure someone's going to comment on this or you know send a note saying like oh you forgot about this one and I probably did um you know what actually is also has has a lot of interesting activity uh but I I don't haven't fully developed conviction yet is mode so mode is is defi layer to uh they're pivoting to be focused on agent interactions they just had a hackathon with with a lot of submissions for AI agents so so I think that mode is not quite there yet but but but that's an interesting one as well well and that's the uh that's the other so one of the other ways to invest in AI agents is

indirectly you could look at more of like the chains right so like could invest in salana to be honest because there's a lot of AI agent activity on salana and if you believe I mean I think everybody here is already holding salana already has enough conviction in salana but like that that is a beta play two AI agents as well you could look at mode for example I don't know if near is doing anything yet with AI agents but they definitely are doing a lot with crypto Ai and trying to be a leader in that space so you could look there any other any other like sort of beta plays Aerodrome even I was gonna say Aerodrome because base doesn't have a token but a Aerodrome decks on Bas could be could be actually yeah because if you look at the

agents that are being traded um base they're almost all being traded on on Aerodrome uh virtuals I think um virtual virtual the token is paired with the agent tokens and I think all those pools I may be remembering wrong I think all those pools are on Aerodrome so that's actually driving liquidity and volume excuse me to to to Aerodrome can't think of any other chains near is always you know near is doing a lot of things with AI the founder is is well known in the AI world before entering crypto they're building a lot of interesting features for AI with their wallet I I haven't anything with agents on there um I I I know that I I participate in this weekly space about AI agents and someone from near is on there so I know it's on their radar because he's very well versed in

agents I haven't quite seen the same level of retail activity but I wouldn't be surprised if that changes um but I think if if you look at where they are they're mostly mostly on salana and in base right now another place on another potential infrastructure play that there's no token yet but is Meteora on salana is where a huge amount of the liquidity for these agent tokens are and uh so the meteor Equity has been going up they have a a points how you spell that me e o a uh it's it's fairly large I think they have about half a billion dollars in in uh liquidity now but a lot of these AI agent tokens on salana are on Meteora and they have an ongoing points program so another way to get exposure would be through

that I'm just going pull up their site here this one yeah yeah so Medora do you know Jupiter on Sala yeah of course so same same team Meteora is is um solely backend liquidity and and there's a lot of AI agents on there just like Jupiter is solely front end very interesting or at least I said Same Team there's at least a lot of overlap like I think it was basically and metora doesn't have a token yet though they don't but they have a points program ah o hot hot alpha right there okay that's interesting okay fi final any final thoughts uh before we wrap up this has been a tremendous episode Patrick really appreciate your insights final thoughts would be really exciting exciting space to be in but

with excitement and opportunity especially in crypto comes risk so people should do their own research uh goes with that saying uh and and also understand that it's these these things long on a long time Horizon the technology makes sense that doesn't mean that bubbles can't happen on the shorter time Horizon big time I mean like you you I don't know about you but like this this is scre screaming nft bubble to me right it's screaming you know what it reminds me of the metaverse bubble and I I even said this in my video I said if you look at a lot of the white papers 90% of the white papers from these projects it reads like something from metaverse boom in 2021 where it's like oh we're applying you know virtual reality and AI to do this and you you

realize very quickly when you read those like there's nothing interesting happening here but that being said that being said I think this technology is much closer to production than metaverse ever was because you're actually seeing things that are already working on a daily basis like those influencer things you showed us earlier that's part of the reason why I like infrastructure as a way to bet on this because whatever form these agents take like you said there'll probably be a new category next week the infrastructure that enables it is going to benefit either way love it Patrick thanks for your time today everybody thanks so much for listening in have a wicked awesome day we'll see you again tomorrow as always if you liked today's episode please hit subscribe and make sure you follow us if you don't miss out

on the next one there's also a link in the description to our free 5 minute daily newsletter where we simplify crypto for you while making you laugh and if you're willing to step up your crypto investing game we're going to also leave a link for milk Road Pro you get access to our portfolio where you can see exactly what we are buying this is your number one resource to help you invest successfully in crypto one final note this show is for educational purposes is only and nothing we say is financial advice investing in crypto or any asset is risky and you should never invest more than you can afford to lose thanks so much for listening in my friends have a wicked awesome day