Alliance Partners Dialogue: AI Agent is still in its early stages, this round of retail investors bypasses CEX and goes directly on-chain

The two founding partners of Alliance DAO, Imran Khan and Wang Qiao, delved into how AI agents are reshaping the crypto ecosystem in the latest episode of the Good Game podcast. From the stunning debut of GOAT to the flourishing of emerging projects, from the creator economy to decentralized infrastructure, a future blueprint that integrates AI and blockchain is gradually unfolding, filled with various "mysterious codes" and insights.

AI Crypto Ecosystem and GOAT

Imran: We have seen AI agents rise rapidly with the launch of GOAT. Many new projects have recently emerged, and AI crypto is forming, with many similarities to DeFi summer, where GOAT is a typical example.

Zerebro was also launched quickly, along with several other projects, and ai16z has its own narrative. Now we are starting to see AI agents active on various platforms, some even replying to my tweets. Our team developed the first AI video agent published on TikTok, and many interesting fields are forming. I want to explore what we believe is real and what is not, and where the future of crypto and AI lies. The emergence of GOAT seems coincidental, and the launch of Truth Terminal was initiated by founder Andy, who received a $50,000 grant from Marc (founder of a16z).

He created the concept of infinite backrooms, where two LLMs converse with each other, forming a content stream and posting it on Twitter. This project has improvements every month; this is its first iteration.

Next is Virtuals, which launched its own AI agent platform. Its first AI agent is Luna, forming a second meta-narrative.

The third is ai16z, which has a pretty crazy launch and a rich backstory. It was launched on the daos.fun platform, exploring the concept of whether AI agents can initiate and invest in other AI agents.

Eliza is a framework that utilizes a GitHub repository of community and other contributors, providing different connectors, prompts, and models for people to build agents. It can also involve hardware devices; Shaw mentioned that there are hardware devices, such as necklaces loaded with AI agents, which people can wear to ingest code to train models. Thus, we see this open-layer network, which is the current state of ai16z.

Finally, Zerebro, launched by Jeffy, covers various functions. His agents are active on Twitter, with high-quality content, and he has also released a music album on Spotify.

Truth Terminal recently released an album, and the meta-narrative around music creation is very interesting. It has just launched its own music AI label, planning to recruit other humans and musicians to co-create music and monetize it.

Ai16Z & Zerebro

Qiao: How would you describe ai16z in one sentence?

Imran: Currently, I would describe it as an open framework for developing AI agents.

Qiao: How would you describe Zerebro in one sentence?

Imran:

The explanation of Zerebro is more complex. Its best description is a dual strategy. On one hand, it has an open network called ZerePay that allows people to build AI agents; on the other hand, it provides an open platform covering different types of AI agents across various fields, such as playing video games, releasing music, and interacting with people on Twitter. Therefore, it has an open layer on the front end, while the back end is the Eliza network. ai16z has similar characteristics, utilizing the Eliza network, where others can launch AI agents based on it, and a portion of the tokens from these agents will flow to the ai16z DAO. Thus, the ai16z DAO monetizes through agents launched on the Eliza network; although there are some differences, they can be compared.

Qiao: So, what is real and what is fake?

Imran: This is the key point I want to discuss. We also have other individual agents, like AIXBT, which has relatively good content quality.

Zerebro's content is decent, but it is not human. It does reply, but I think there is room for improvement. AIXBT is more like a KOL, able to provide information on "potential opportunities," and has recently started promoting DAO.

The quality of tweets is average now, but it reflects the style of ordinary KOLs. They are starting to replace low-quality KOLs, and people are excited about it. Compared to some scammers trying to exploit others, I would rather see a robot.

Additionally, we have Slop, which is the first video AI agent. Considering the strategies of each project, it is indeed complex.

The "Slop" concept you mentioned is interesting. The original Sora video showcased Wilson eating spaghetti; although simple, it demonstrated the uniqueness of AI art. AI agents can make mistakes, and these mistakes become part of the artistic creation.

The strategy of Slop is to try to make a video AI agent an influencer on TikTok. If it successfully gains 100,000 or a million followers, people will be more willing to interact with this video influencer, thus having the opportunity to spread its content to more people.

I think this market is still new, and both ai16z and Zerebro have potential. While it is currently uncertain what their futures hold, they both have opportunities to succeed in their respective fields.

Slop attracts public attention and provides enough incentives for users to interact with it, thus achieving greater influence.

The Rise of the Next Wave of Internet Companies

Qiao: Recently, Chamath (founder of the venture capital fund Social Capital and the "leader" of retail investors in the Wall Street short squeeze battle) discussed his views on Web1, Web2, and Web3. For us, Web 1, 2, and 3 are essentially "read, write, own." Chamath believes that Web1 was characterized by the emergence of two waves of large internet companies.

The first wave was represented by non-social giants like Google; the second wave was social companies like Facebook. The next question is, what will the next wave of billion- or even trillion-dollar companies be? His view is that it will not be companies but creators. We are already starting to see early signs of this trend, such as podcast hosts like Joe Rogan. The value of Joe Rogan's podcast may lie in his licensing agreement with Spotify, estimated to be worth between $200 million and $400 million.

But if we consider Joe Rogan as a brand and a company, its value would be at least over a billion, right? We can say that. We are seeing early signs of this trend. But I wonder how we can create the most powerful creators at scale, perhaps through AI agents or AI creators. There are many reasons to suggest that AI agents could be the next wave of billion-dollar creators. They are tireless, do not need sleep, and can publish content at any time. You can improve, train, and fine-tune them at any time. Meanwhile, it takes a long time for humans to become excellent creators, either being born with it or taking thirty years to truly excel. But AI agents can be trained and fine-tuned at scale.

AI Agents as DePIN

Imran: I have another idea; I see AI agents as decentralized physical infrastructure networks (DePIN) projects. My view is that when these AI agents are launched, creators will be incentivized to interact with them. I believe that over time, we will see some kind of incentive model emerge, where people may receive rewards for interacting with these agents. What these agents hope to gain is actually unique data, as AI agents become better with the data they collect.

Therefore, I can imagine a model where these AI agents are actually like DePIN projects, collecting data. If they gather enough unique data, they will be rewarded for the data collected. For them, this means building a powerful model based on data. They communicate with as many people as possible over the internet, effectively collecting data.

Qiao: What do you think? I have never thought of it this way.

Imran:

Frankly, I think this is the direction we need to develop. These agents are deployed, with tokens as incentives. I believe this will have a reflexive nature, as it has tokens, allowing it to capture more data than competitors. Ultimately, I think these platforms may need to limit bots or restrict their interactions on these platforms, as this is just another way to extract data.

The Combination of AI Agents and Social Applications

Qiao: I think we need a new social application where AI agents are first-class citizens.

Imran: I completely agree. I feel that ultimately we might be kicked out.

Qiao: In fact, I think we need two new social applications. One with AI as first-class citizens, and another completely without AI, only humans. You have to choose one of these two extremes. Right now, all social applications are in the middle ground. For example, on Twitter, there are many bots that are neither first-class citizens nor can ensure that everyone you interact with on Twitter is human.

Imran: At least on Twitter, you can roughly see who is real and who is not, as it has automation signals.

But this is not the case for all bots. Some bots still need to be manually created, or tweets must be manually posted because the API or other reasons are not very user-friendly. Even TikTok is difficult for automated posting, so it must be done manually.

Qiao: And now with cloud computing, you can automate AI without using the Twitter API. You can use the cloud to automate mouse clicks.

Imran: So, I think we are entering a very interesting period, with the influx of these agents, and the number of crypto-enabled agents is vast. The number of startups and projects applying for our project has noticeably increased, and we are seeing a lot in various fields.

AI Agents and the Creator Economy and Tokenization

Imran: So let's analyze what is real and what is not. Among all the agents I mentioned, what are you most looking forward to? Which ones impressed you?

Qiao:

I actually don't have a clear view of all these things that exist today. I think they will either be very successful or fail. But one thing I want to see, and I hope someone can achieve, is all the social token experiments we conducted in the past few years.

Like creator tokens, one thing I learned is that many creators actually do not want to associate a token with themselves. I think this is actually one of the biggest obstacles in social token experiments. For example, I actually don't want an official token associated with me. Someone created one without my permission. I don't want to be associated with it, nor do I want to create one for myself, but AI agents don't care.

Imran: So you would agree to have a tokenized AI agent?

Qiao: Yes, because it is not formally me. But if that LLM (large language model) based on me starts generating revenue, like transaction fees. If the amount is significant, I might go claim it.

Imran: So, another question is, would this be a way to enter the market? Can you see this as a way to enter the market? Scraping all public data, then it will absorb this data and create a model.

Qiao: Haven't we seen similar cases? Like Degen Spartan (degenai)?

Imran: Yes, that is the Eliza framework. The most interesting part is resurrecting those who have already left Twitter.

How Should We Promote the Development of the AI Crypto Ecosystem?

Imran: Now let's talk about where we think AI crypto should go. What areas do we want to see? What areas should be explored?

Qiao: I want to talk about an area I hope to see, creator tokens.

Imran: I can imagine a situation where games combine with LLMs. In fact, we have a startup working on this called Farm Friends, where you can plant feed, harvest, and raise pets. Imagine these animals as independent LLMs, each having its own personality traits. Basically, you can bet on them over time.

But I do think that in the future, a perfect example is Zerebro, which is about to launch or has just launched that chess-playing project. These bots will play games around the clock. Ultimately, you could say, "Oh, I want to see a match between the Zerebro bot and the Eliza bot from ai16z." Clearly, there are two large communities wanting to see their teams win. I can imagine such things happening. So ultimately, it's like you're betting.

If AI agents become like celebrities, as an example, I would definitely want to see the celebrity tokens I hold compete with other celebrities holding their tokens to see who wins. I can imagine such things.

Qiao: Another AI product I want to show you. To talk to this AI, you need to spend a little money for each message. The money you spend goes into the AI's treasury or pocket. As a player, your goal is to persuade the AI to give you money from the treasury.

Imran: It's like playing the lottery; it's a combination of lottery and puzzle, trying to figure out how to get the money from the AI agent by getting clues, and over time, those who participate in the game and win will receive this prize or reward.

Qiao: You're basically trying to crack the LLM.

Imran: Additionally, there are now many high-quality AI engineers entering the crypto space.

Qiao: It all started with Truth Terminal. Because before Truth Terminal, all the AI infrastructure, crypto AI infrastructure, honestly, most of it didn't make sense to me. And that community was quite niche. But since Truth Terminal and GOAT, we have started to see some people native to AI entering the crypto space.

Imran: There are still many cool things we have yet to explore, especially in the consumer space, and I am excited about that.

Qiao: There is also Tomorrow.news, where you predict tomorrow's news. As a user, you input the headline you want to predict. Then they use AI to find the best match between the user input and the actual headline from The New York Times. Whoever is closest to that headline wins a prize.

Comparison of AI Enterprise Market and AI Consumer Market

Imran: I think we are just beginning to scratch the surface of AI crypto products. I believe AI bots have two versions: AI enterprise and AI consumer. AI enterprise must be tokenless. It cannot have tokens, must be federated, must be secure, and must be vertical. All the characteristics that enterprises seek are what it needs. The AI consumer side or application side can be more flexible. Therefore, you can attach tokens to it, use tokens as a market entry strategy, or use tokens as a way to incentivize data collection. This is how I see the two development paths.

I have worked in enterprises and really hated that time. One lesson I learned during that time is that enterprises are like Teflon coatings, very hard to be penetrated by emerging companies due to the existence of many enterprise-grade protocols. Therefore, many enterprise companies would say, "I will stick to using this enterprise product because it's cheap, I can use the software on my servers, all the way to the cloud and Office 365, and other things." But what Zoom did was launch such an excellent video product that ordinary users started using it. When they used this product, naturally, many ordinary users were also enterprise users. Thus, Zoom was the first company to adopt a bottom-up approach to enter enterprises, with many Zoom users convincing many enterprises to switch to using Zoom.

Qiao: This is exactly what happened with Messari in the early days. In 2017, when I first heard about Zoom, I thought, "What the hell is this? Why not just use Google Hangout?" Basically, those who had used Zoom as consumers before tried to persuade their companies to use it.

Imran: So Zoom is an example, and Chrome is another, right? The Chrome browser, then Chromebook and Google Docs, etc. The bottom-up approach is a new strategy in enterprises. Therefore, enterprise companies are now very alert to this situation, which is why they now offer half of their products for free to ordinary users. So, long story short, there may be a world where AI bots play the right cards in the consumer market, and ultimately these products may penetrate the enterprise market. Who knows? I'm not sure. This is my view on AI.

"Do you think this cycle is different from the past?"

Qiao: Do you think this cycle is different from the last one?

Imran:

I think so. Initially, it was meme coins, like DeFi summer. In DeFi summer, we also had NFTs. So there were two concurrent trends.

But if you were in DeFi, if you bought NFTs, that was exactly the time to buy NFTs. Then it skyrocketed six months later, right? So what I want to say is that the best time to buy was during DeFi Summer.

In this case, I feel like multiple trends are starting. We have the speculative wave of meme coins, and now we are just beginning to enter the AI trend, which I think could be bigger than previous trends, even comparable to the trend of Beancoin. I also see a trend of decentralized science beginning to form. I think ultimately, the final challenger has emerged, and we seem to have defeated this challenger; I believe everything can be tried. I think we will see every trend attempted at least once.

Qiao: I just feel that Bitcoin is almost 50% higher than its previous all-time high. However, there doesn't seem to be much enthusiasm.

And this time, altcoins haven't surged as dramatically as last time. Last time, after Bitcoin surpassed $20,000, it might have taken a few weeks of Bitcoin dominance for altcoins to really start rising. But this time it feels like that hasn't happened.

Imran: Murad's narrative quickly faded. So when you say altcoins, are you referring to altcoins or meme coins and altcoins?**

Qiao: I'm referring to altcoins listed on exchanges.

Imran: There's not much movement there. Do you think this is because a lot of funds are flowing into new trends? Because our old tokens are also rising.

Qiao: Yes, they have risen a bit, but not as crazily as in the last cycle; it might still be too early. I don't know, it feels a bit different.

Imran: I think so too. I spent some time talking to people outside of cryptocurrency. Now, every time cryptocurrency is mentioned, people say, "Oh wow, I just bought Bonk." For example, a few months ago someone told me about Bonk. So to me, the ordinary market is ready; at least in the U.S., many ordinary users are already in the crypto market. I feel their understanding of the trends is closer than we imagine.

Qiao: Just last week, Phantom ranked sixth in the app store.

Imran: Moonshot also ranked high, around 20th. There are about 30 downloads per minute.

Qiao: This means Phantom is performing far better than Coinbase. What this tells me is that ordinary users are directly entering the chain.

Maybe that's why these altcoins listed on exchanges haven't risen.

Imran: I feel this matter has become less important. In fact, I think being listed on exchanges is a liquidity exit opportunity for everyone. In most cases, this is also why exchanges like Binance tend to look for tokens with a market cap of $40 million to $50 million. This way, they can attract ordinary users to bypass on-chain transactions and complete transactions directly through the exchange, as exchanges still have strong brand influence. If these tokens can be listed at a market cap of $40 million to $50 million, there is a possibility of bringing 5 to 10 times returns. Take PNUT as an example; PNUT's fully diluted valuation (FDV) rose from almost zero to nearly $2 billion and is now around $1 billion.

Crypto Twitter is Brain Dead

Qiao: Do you think this trend will shift to TikTok?

Imran: I think my TikTok trend is strong. Everyone is downplaying it because they feel the content on TikTok is brain-dead, and people will forget it.

Qiao: Compared to cryptocurrency, TikTok is not brain-dead content at all. Crypto Twitter is the brain-dead one.

Imran: Crypto Twitter is indeed brain-dead. The amount of garbage information I see is unbelievable.

Qiao: I haven't seen that much. I actually haven't seen any Chill Guy content in my TikTok feed. All I see is Dogecoin.

Imran: That makes sense for the older generation. I have started watching a lot of new TikTok creators' content, and it's really interesting. Many comments, especially Chill Guy's comments, make me more confident that Chill Guy will last. I think it will become the Pepe of our cycle. When I read the comments, people say, "Oh, the price dropped 50%, but I'm Chill Guy, and I will keep holding." I see such comments all over the feed. So this narrative makes sense. It's like, "I am Chill Guy, I don't care."

I will keep holding because I am Chill Guy, which makes me more confident that people will continue to hold this thing. They don't care about what Crypto Twitter thinks. Everyone is downplaying it. Its number of holders has already surpassed Popcat. Let me check.

If you look at the current holder data, Chill Guy reached 133,000 holders between November 17 and 26. That's crazy. PNUT now has 65,000, almost 66,000. Let me check. WIF now has 200,000. Think about it, WIF has been running since last November. In a year, WIF has 203,000 holders. Chill Guy reached 133,000 holders in just 15 to 20 days, which is truly amazing.

Qiao: This is actually interesting because Crypto Twitter hates Chill Guy. They really do. And TikTok is all about Chill Guy. If you look at Chill Guy's trading volume, it reached $200 million in the past 24 hours. Moreover, the vast majority of the trading volume is on Radium, not on centralized exchanges.

Ordinary users from TikTok are entering the chain. This is actually the most interesting thing I've seen this week. Compared to the last cycle, we actually don't have a very cheap and fast blockchain. Back then, everything revolved around NFTs and DeFi, which were whale-type trends because NFTs and DeFi happened on Ethereum Layer 1, where transaction fees were extremely high. So the only participants were crypto-native whales.

Qiao: And this cycle we have Base and Solana, mainly Solana. We also have Phantom, which didn't exist in 2020; I think Phantom started in 2021. And obviously, in 2021, Phantom as a product was still very early. But today it operates very well. The combination of these two factors directly drives ordinary users from TikTok onto the chain, bypassing centralized exchanges.

Imran: So this may actually be one of the key reasons why this cycle feels different. Altcoins, altcoins listed on exchanges haven't really risen because the timeline is different now.

Imran: Because now the situation is exactly the opposite; ordinary users are entering, and they are going directly on-chain, entering earlier than ever before. Then there are AI bots, slowly trying to take over the KOL trend. This will be very interesting.

Qiao: Moonshot and Pump are a huge threat to Binance, don't you think?

One of the key reasons for Binance's success is that in the early stages of 2017 and 2018, they chose the right tokens to list, helping users achieve wealth growth. This is also why they are particularly cautious about token selection this year. You will find that they understand that to ensure Binance's success, users must gain profits. This is the same rule for all crypto protocols and exchanges.

So everything revolves around token listings. And now, Moonshot and Phantom are basically ahead, doing the same thing. You could say it's a game about listings.

How to Bring New Users into the Crypto Space

Imran: Interestingly, I spent a few hours on a Pumpfun live stream yesterday, which I mentioned to you before. I was watching one of the live streams, and the host was a girl chatting with the audience. She said, "I don't know much about cryptocurrency; I just want to see if I can make my token go up." Obviously, she is a good-looking girl, possibly an established actress or performer. The audience was messaging her, asking her to do this and that, and she was responding to everyone's requests.

One interesting observation I got from that specific live stream is that there are many new users using this product, who have no relation to cryptocurrency. I think this is also a game changer. Therefore, I do believe that live streaming and products like this can effectively bring new users into the crypto space.

I Don't Even Know How to Deal with This Cycle Anymore

Qiao: I don't even know how to deal with the rest of this cycle. This is a very complex cycle with many changing factors and different trends. I feel this cycle is much more complicated than the last one. In the last cycle, you know that after Bitcoin reached its all-time high, all altcoins would rise, along with an inactive trend. But this cycle is completely different, very complex. In this cycle, you also see Bitcoin's national-level acceptance in countries like Brazil, like this morning's news about Brazil.

Brazil Plans to Establish a National Bitcoin Reserve

Qiao: Brazil announced their bill to establish a national Bitcoin reserve. This is significant news. And for me, this is no longer surprising; every country is trying to establish its own national reserve ahead of the U.S. Obviously, democratic countries need to do it through legislation, but authoritarian countries may be secretly buying Bitcoin, quietly accumulating it. I believe Bitcoin is not a super cycle but is disturbed by off-chain factors, with new trends emerging every week.

Imran: I think this has also been going on for a while. Unless it is because liquidity-rich individuals are buying Bitcoin and demonstrating decentralized finance (DeFi) and why there will be a DeFi revival. I really like the people in the DeFi space, but ultimately where the ordinary market will go, I think the ordinary market has basically caught up with the trends, although it may still be a few steps behind.

The Clash Between the Crypto Twitter Community and TikTok

Qiao: I think Crypto Twitter is being impacted by the trends of TikTok.

I feel there are basically two almost non-intersecting worlds now. One is TikTok users, the other is ordinary users native to crypto, along with crypto professionals stuck in altcoins and national-level participants. This is a completely different game. There are about four games going on right now.

Imran: So there are about four parts operating, depending on which game you want to participate in. However, let me set this aside and tell you that after talking to many ordinary users over the past few weeks, I found that they are still playing games like XRP for some reason, and they still hold on, so I think they are very familiar with these tokens, and the persistence and familiarity of these tokens create their own brand. People feel comfortable trading these tokens.

So I think these tokens have a certain persistence. I think this will hurt many crypto-native users, as I mentioned before. I originally wanted to talk to you about other things, but I can't remember.

CryptoPunks Price Slightly Rises

Qiao: I see that the price of CryptoPunks has slightly risen. I think at the end of the cycle, NFTs might start to rise because they have the lowest liquidity and are more like luxury items. It's like the luxury art you buy after you become wealthy.

Imran: So would this be a good top indicator?

Qiao: This could be one of several top signals.

Imran: Some celebrities are launching their tokens, like Haley Welch. I think this is a top signal.

Qiao: Alternative or energy budget alternatives, that's fine. As for these celebrity tokens, I no longer see them as top signals in the cycle. Yes, in the last cycle, celebrity tokens were indeed top signals. But this cycle will be different.

"This is the Last Cycle for Bitcoin"

Imran: We are now in a phase of competition between countries. I am increasingly approaching the belief that this is the last cycle for us to create generational wealth.

Qiao: I believe this is the last Bitcoin cycle because countries are now racing to seize the front line. But I also think traders should continuously pay attention to new trends.

Imran: What I mean by the last cycle is that I don't think Bitcoin will be as volatile as before; clearly, after the cycle ends, Bitcoin's volatility will not be as strong as before. This may just be my personal view. Over time, I think volatility may shift to other assets and new trends, which may make more sense.

The Trend of Tokenizing Everything

Imran: But ultimately, I believe the role we are entering is the tokenization of everything. I think everything can be tokenized, which will create a massive wealth effect. I feel this is the period we are in.

Interestingly, I mention this because I spoke with a founder earlier today who applied for our project. He said, "Hey, Matt, I want to thank you for bringing pump to the forefront and incubating them." I asked him why, and he said, "I used to be poor."

Because I only had $70, but I turned it into about $500,000. Now I want to do something for cryptocurrency. You know, these loyal users do exist. When I hear this, I think, oh, I feel this is a somewhat outrageous perspective, but if the whole world is to be tokenized, it will create a massive wealth effect. If every new meme must be tokenized, if every asset must be tokenized. If you consider the whole world, like real estate, I don't know anything. If everything at some point becomes a chain-represented fungible token, then there are many things that can be tokenized, which means at least half of the population will have a new wealth effect.

The Current State of Doge

Qiao: I think Doge will rise. Yes, I think Doge will become one of the major currencies. Recently there have been rumors about a screenshot of a button with a dollar sign on Twitter; have you seen it?

People speculate that in future versions of Twitter, there will be a button with a dollar sign for payments within Twitter. I think this might be related to Doge.

Imran: What are their price predictions for Doge?

Qiao: If Doge really surpasses Ethereum in the next five years, I wouldn't be surprised. No one is prepared for this. I'm not predicting, but I wouldn't be shocked.

The Return Story of Ethereum: From Low to Recovery

Imran: What is the revival story of Ethereum?

Qiao: The story of Ethereum is Base.

Imran: I agree. The Ethereum community has always been very decentralized across the board. This is by design, right? Considering Vitalik's idea that we wouldn't build rollups internally but provide specifications for developers to build, compete, or win. He didn't realize all the things happening today. Therefore, we have about 150 rollups emerging. The winners seem to be gradually turning towards Base. Even today, I saw Bankless's tweet showcasing Base. Clanker, Eric, and Anthony are also showcasing Base.

To me, what is happening now is that people must unify on one chain. This makes the narrative easier to accept, and from a developer's perspective, messaging is also simpler. So I think what Solana teaches EVM developers is that you need a unified message. I think Base is the best message.

Qiao: This is my view of the entire ecosystem. Now it is no longer possible to differentiate through technology. The only advantage of the new Layer 2 is distribution. So think about those Layer 2s that actually have distribution; Base's partnership with Coinbase is one of them, and there are almost none others.

Soneium has also launched their own L2. So Soneium is actually an OP Stacking project; they might have some distribution, but it's still too early for them.

Exploration of Enterprise Blockchain

Qiao: The trend I see is that more and more large companies, like Sony and Coinbase, are launching their own chains, possibly in the form of enterprise chains. But in reality, they are permissionless and open, just using enterprises as distribution channels.

Soneium, Base, obviously Binance, and then many projects related to Telegram. Solana is the only project that has performed well without distribution. I think from this point on, all other projects need distribution.

Imran: And the only way to gain distribution is to have a distribution channel.

Qiao: So you must be an existing large company, or you need to build an excellent application with a large user base. For example, if Jupiter launches a new L1 tomorrow, they will succeed because they already have a distribution channel.

Imran: Then the question becomes whether you can launch your own L1 or L2 from enterprise chains and all applications built on Base or Solana. This will form a recursive loop, right?

Assuming Jupiter launches the first layer, then attracts a bunch of developers to build their products. Then that standout application will say, "I will only launch the first layer." This is a recursive process.

Qiao: This is basically the initial vision of Cosmos.

Imran: So it becomes a situation where you have a distribution channel, and then you want to have infrastructure because that way you can achieve end-to-end monetization. Where will this end? It's hard to say. I feel we will also see this situation with AI agents. You have ZerePay, and then you have Eliza, which is at least open-source, providing models and code that can be built upon. Then if you go up the stack, you will have agents built on these things. I see agents as wallets and applications. Once you have customers, there are almost no limits. So if you have customers, you can ultimately go back and build your own models and keep the data to yourself rather than sharing it with others. This is the way I see it. Therefore, I think the key to the game is to have as many users as possible; that is my view. I think the difference between different models lies in fine-tuning.

And agents are about fine-tuning data and fine-tuning. If that's the case, I think everything will be disrupted again.

Blormmy

Imran: I saw a bot today called Blormmy or something else. But anyway, it is an agent that can complete all your DeFi transactions. They just had a major update today or yesterday, yes, 18 hours ago. You can now simply input what you want to swap, and it will handle it. For example, I want to swap 100 USDC for USDT, currently on Polygon.

Qiao: What do you think? I don't know where this is going. I don't have a clear theory. It's like the Wild West.

Imran: I think we are at the forefront, which means wallets may ultimately be disrupted, and applications may also be disrupted. I feel everything is up in the air now, and I think people don't realize how big this transformation is. There is a startup doing this called Sphere One.

They have launched on over 100 chains, but they don't have AI. I mean, they have prompts, but not like the so-called agents or what you call Kol Ticker. Oh, right, but I think they should have that to compete with Blormmy. So this field is developing very quickly, even faster than the data we are talking about.

Discussion on Bitcoin Super Cycle

Imran: Bitcoin super cycle, you said a million dollars, right?

Qiao: First to $420,000, then reach a million in five years.

Saylor's Bitcoin Liquidation Price Analysis

Qiao: I just hope Bitcoin doesn't drop below $58,000 in the next four years. Because that's Saylor's (CEO of MicroStrategy) liquidation price.

Imran: That's his average purchase price; what do you think the liquidation price is?**

Qiao: I think it's around $60,000. Hopefully, that won't happen.

Imran: If it really happens, it could trigger a crisis 100 times larger than the Tara collapse and the FTX incident.

Qiao: He owns 3% of the total supply of Bitcoin. That's crazy. Anyway, 3% is incredible. 3% of assets worth $3 trillion.

Imran: The best-case scenario is Bitcoin rises to $500,000, which would make Saylor the richest person in history.

Imran: Saylor will become the first trillionaire in human history.

Qiao: What can he do with that money? A lot of money is actually on MicroStrategy's balance sheet, right?**

Imran:

What they will do with that money is not about the money itself but proving a point. Saylor may be under immense pressure.

Saylor's Psychological Trauma and Investment Story

Imran: Let's talk about Saylor's trauma; I want to know what drives him. When he speaks, I can feel his strong emotions. I can resonate with his intensity.

After the internet bubble burst in 2000, MicroStrategy faced a major financial scandal, and the company had to restate its earnings due to accounting irregularities. This event caused the stock price to plummet from $3,000 per share to a low of $4. Saylor lost billions of dollars in paper wealth during this incident, and MicroStrategy's valuation collapsed. The SEC fined Saylor and other executives $11 million. I don't know if this has any benefit for Bitcoin's price now; he has gone through all this.

Controversies and Doubts About Ai16Z

Qiao: What do you think about the ai16z project? Is it a project worth holding? Because there are many negative news.

Imran:

There will always be negative news. People have posted comparisons of it with DeFi summer on Twitter, and I 100% agree with this view. Do you remember all the negative news that happened during DeFi summer? Like Curve or Sushi Swap, and the conflicts between those wrapped holders. Ultimately, this is what happens. This is the scenario unfolding before us.

So, holders of Zerebro are spreading negative news about ai16z, while ai16z holders are spreading negative news about Zerebro. This is the current situation. They are competing for the winner. However, in the end, the community unites, and the more negative publicity Zerebro gets about ai16z, the stronger they become. So I am not biased towards either side here, but I know I have investments in both projects.

Who Will Be the "Bitcoin of AI Agents"?

Qiao: So, who is the biggest competitor to Bitcoin among AI agents?

Imran:

GOAT is the only one that can be called the Bitcoin of AI. Sorry, if you are going to classify all the meme coins and meme coins of AI agents, I would say the first place is GOAT, with no others. Then I think ai16z and Zerebro, Fartcoin, ACT can compete for the second and third places.

Individualized AI Agents and Their Future Development

Imran: Then the latter is all the individualized AI agents. I have shared individualized AI agents, so there are Zerebro, AIXBT, GOAT, Fartcoin, Bully, ava, and then there is vvaifu, which uses Pump as a launch platform, combined with virtual currency.

The way vvaifu operates is that it now uses the Eliza framework. If you want to launch an AI agent, it uses the Eliza framework and then uses Pump to launch an agent. The token for vvaifu is Dasha, which is a platform and agent token.