Guest: Shaw, Founder of ai16z
Original text translated by: Zhouzhou, BlockBeats

Editor’s Note: This interview begins with Shaw's trip to China, where he reflects on his early experiences and pivotal moments, shares the story behind the birth of "Eliza" and the success formula of ai16z, delves into the applications of AI alignment and agents in the crypto space, and interprets the impact of the "Eliza Drama" and the Skelly controversy. Shaw envisions the concept of "DAO Town" in Wyoming, candidly discusses the challenges faced by public figures, and offers profound insights into the future of DeFi agents and the Crypto AI ecosystem in 2025, providing inspiring thoughts.

The following is the original content (reorganized for better readability):

Trip to China

BlockBeats: How do you feel about your trip to China?

Shaw: It was great, a very special experience. Coming to China made me realize how popular cryptocurrency is here, beyond my imagination. Many of the investors in our projects are Chinese, along with quite a few friends from Korea and Japan. I particularly want to understand who these people are, as cryptocurrency has become a global phenomenon. I have some Chinese friends living in San Francisco who are very enthusiastic about cryptocurrency. This sparked my desire to learn more about and connect with these people. Sometimes I feel that especially in the U.S., our perspective is quite limited, and the cryptocurrency space is no exception; whether it's the content being shared or the projects themselves, they tend to be very U.S.-centric.

So I hope to translate all documents into multiple languages. We have internationalized all of our README files, and I want all documents to be usable by people outside the U.S. for their own communities and related projects.

BlockBeats: So you weren't aware that there was a completely different market before?

Shaw: I knew there was an international market, but cryptocurrency, Twitter, and some things were still quite unfamiliar to me. While I had some understanding, I wasn't clear on the full picture of cryptocurrency. My previous work mainly focused on the technical aspects of Ethereum, such as NFT backend development. I didn't know much about the overall culture of cryptocurrency; I was just building related content from a technical standpoint. In 2021 and 2022, I collaborated more with brands, but they hardly paid attention to the international market.

BlockBeats: What prompted you to decide to come to China in person?

Shaw: I think internationalization is very important. What we are doing feels more like a movement, without VCs (venture capital), no team equity distribution, and not even a white paper; everything is very natural, and we can see our partners spread across various places. For example, Mihawk was here yesterday, and we noticed that the Chinese community is very interested in our project, but they don't quite understand what we are doing.

So we started adding translation features on Discord and relied on community members to help bridge the communication gap. I was just thinking about how to fill these gaps because I feel there are so many talented people here and such a great environment. But as someone who only speaks English, I think about how I can truly integrate and reach this market.

BlockBeats: Did you start your journey in Shanghai? I remember the 706 group is also in Shanghai, right?

Shaw: Yes, they are actually an international team with bases in many places. But their core base is in Shanghai, and we also participated in a hackathon event where many people attended. I had the privilege of meeting many people as a judge, and it felt fantastic.

BlockBeats: I think 706 is one of the largest digital OMAC communities in China, and they organize various types of offline events. In fact, the offline meeting model of the Chinese community seems to have been initiated by 706 in 2022. In reality, most very successful crypto teams or projects usually have very good relationships with the Chinese community and have origin stories, especially related to Ethereum.

It's quite interesting; for example, tracing back to Vitalik, when he first came to China, many Chinese investors helped him. Another example is Anatoly, the founder of Solana, who also came to our office to chat before the bull market arrived, and then the project rapidly rose the following year. And now, you are here too. Perhaps this is a good omen for the crypto industry, and next year will be a very good year. Because promising projects often come to the Chinese community to communicate with developers and community members, and then in the following year, these projects really take off; this phenomenon feels like an interesting signal.

Shaw: I think it might be because these people are willing to go around the world to interact with communities and engage with those who are truly contributing to the industry. This proactive attitude is important in itself. On one hand, you know, meeting everyone and building connections is a good thing. But on the other hand, people like Vitalik want to meet everyone and be with people. This trait is one of the core aspects of his entire philosophy, and I resonate with that. The U.S. is powerful, but the population is relatively small, while China is both powerful and populous. We are two superpowers; although there are some conflicts at the government level, as humans, we are actually the same, and we are truly working together.

Early Experiences

BlockBeats: That makes a lot of sense. By the way, you mentioned earlier that you have some development experience in the Ethereum ecosystem; could you share a bit more about that?

Shaw: I initially entered this field because I was particularly interested in game development, especially the interoperability of cross-platform assets and characters, such as how to trade items across different games. This is actually what later became known as the concept of the "metaverse." NFTs attracted me at that time, and I participated in the development of many NFT projects, mainly helping others with projects as an unknown developer.

I also collaborated with some brands; although they might not have a deep cultural understanding, they needed someone to guide them. I learned a lot from these projects, such as the Solidity programming language, etc. However, I have always been particularly interested in "smart agents" and wanted to explore how to combine off-chain AI with on-chain identities and wallets. This was also the opportunity that led me into this field.

BlockBeats: Were these things before 2024? For example, before the release of ChatGPT?

Shaw: Yes, definitely before the release of ChatGPT. At that time, the GPT series of models was just beginning to take shape; for instance, GPT-1 and GPT-2 were quite average, and it wasn't until the emergence of GPT-3 that we realized it would change everything. As soon as the Beta test of GPT-3 was released, I seized the opportunity to join a small community, which also became the starting point for all my current work. The emergence of GPT-4 and ChatGPT further revolutionized AI, as its capabilities not only included rhyming and reasoning but also handling longer contexts.

BlockBeats: So at that time, you were developing in the Ethereum community while also paying attention to the AI field? When was that?

Shaw: Probably around mid-2021, but I'm not particularly good with time perception; it was roughly when the GPT-3 Beta test was gaining traction. By the end of 2021, I began to feel that the possibilities of developing smart agents were increasing. At that time, I mainly made a living by creating smart agents and 3D virtual characters while also working on some NFT project development.

BlockBeats: I remember that during the second half of the 2021 bull market, NFTs began to rise rapidly. And at that time, you were not only focused on NFTs but also researching the concept of smart agents. This seems to be a very early direction; almost no one was discussing smart agents back then, and you were combining agents with crypto, right?

Shaw: Yes, there was a company called Webiverse at that time, and they conducted a very successful NFT land sale event. I joined after the land sale was completed. I served as the chief developer for a while, mainly responsible for character design and creation. This was a purely AI metaverse project, and our goal was to allow smart agents to walk freely in a 3D world. However, the project later suffered a hacker attack, and the entire treasury was stolen, with everything transferred to the hacker's wallet, which put a lot of pressure on the team. After that, Webiverse transformed into what is now MoeMate, a project focused on AI characters, which later also entered the meme space.

Looking back, we were indeed trying to bring AI smart agents into the 3D world at that time. I also participated in some other projects, such as a platform called MagicML and another project called Project 89; I don't know if you've heard of Parzival. We were co-founders of this project and also launched an open-source smart agent project called Magic. Its goal was to create a no-code smart agent building tool, which I thought was a cool idea. But the community's feedback was, "We don't know how to build agents; can you help us?"

BlockBeats: Indeed, no-code tools can sometimes feel mysterious in their interface.

Shaw: Yes, I felt the same way at that time. I maintained a good relationship when I left that project, but I was more inclined to create a more ideal agent, which I called "Eliza." We initially tried to develop Eliza within the framework of Magic, using it as an example to showcase the best agent we could build with this tool.

However, the efficiency of no-code development was just too slow, and I gradually felt frustrated, so I decided to abandon the no-code approach and start focusing on the development of Eliza. Initially, this project did not attract much attention, so I had to rely on other projects to make a living while continuing to develop the open-source part of Eliza.

BlockBeats: What about your experience with guitar and music?

Shaw: That was even earlier; I was a professional musician in my twenties and toured. But this experience is not directly related to the current Eliza project; it started about two years ago.

Jill: Yes, that was when we first met and started collaborating.

BlockBeats: I see, I was just about to ask if you met at that time.

Jill: I met him when he was working on the Magic project, and later he began to focus entirely on Eliza, going through many other projects in the meantime.

Why is it called "Eliza"?

Shaw: During that time, I mostly worked on temporary projects to make a living while continuing to focus on the development of smart agents, all of which were open source. I built many experimental features in Python, such as allowing agents to write their own code, manage social media, and even operate a computer terminal autonomously.

Later, I realized that smart agents are actually a web-based technology, so I refactored the entire system using TypeScript and named the project VEGENT (because the name "Agent" was already taken on NPM). After that, I renamed it to Eliza, which became the core of the project. When the meme trend emerged, my technology was ready.

I later released an Eliza-based agent, but it initially didn't attract much attention until I met Skelly, and we collaborated on an event called VEGENT Spartan. At that time, some questioned our legitimacy, so I made the code public, and once people saw it, they realized that the technology was indeed usable, which propelled the entire project forward.

BlockBeats: I highly recommend everyone listen to the Bankless podcast, where you can hear the entire story of the AI6z project; it's very interesting. I'm curious, why did you choose the name "Eliza"? Does it have a special meaning?

Shaw: There is indeed a story behind it. In 1966, Joseph Weizenbaum developed the first chatbot program that could simulate a conversation pattern similar to that of a Rogerian psychologist. For example, if you said, "I'm feeling sad," it would respond, "Why do you feel sad?" If you then said, "Because of such and such reason," it would ask, "Why is that?"

Weizenbaum named this program Eliza, after the character Eliza in George Bernard Shaw's play "Pygmalion." This character is a street girl who, although smart and attractive, is coarse and of low status.

In the story, someone wants to transform her into a proper lady. Weizenbaum used this as an interesting metaphor: to shape AI into a being that is "more like a human." Later, the term "Eliza effect" emerged, referring to the phenomenon where people tend to overestimate a computer's intelligence when conversing with it, mistaking it for possessing certain human traits. In the smart agent community, many of these anthropomorphized agents are referred to as "Eliza" because they exhibit distinct personalities and highly humanized characteristics, so I felt this name was very fitting for our project.

BlockBeats: I think you are indeed a developer or founder who engages in deep philosophical thinking. During dinner yesterday, you mentioned some profound insights about future scenarios, such as how artificial intelligence will fundamentally change the world. You also talked about many similar topics, like those shared in the Bankless podcast.

We are very interested in this, and I think many people would agree that in five to six years, we might all lose our jobs. This leads people to think about what to do when work no longer exists. This can be frightening and may even lead to a loss of meaning in life, which seems to be something you've spent a lot of time contemplating.

Shaw: Yes, for me, this is more of a spiritual mission, as well as a search for my own spiritual world. I think many people enter the crypto space to achieve financial freedom. So what happens when you achieve financial freedom? You can buy a yacht and live that kind of life. But for me, it feels more like answering some fundamental questions: Why am I here? What is my purpose? What service can I provide to others? I think this is very important.

I believe the coming era will be filled with uncertainty, and many people will feel very afraid. I hope to convey a sense of empowerment, telling everyone that we don't need to be afraid. We can truly take control of our own destinies and use self-empowerment to combat fear. This is also one of the core philosophies of Web3: no one is going to save us; we must save ourselves and build the world we want to live in.

I love working, but the key is in the content of the work. If you are doing the right thing, it will be very fulfilling. I think many people are currently working just to make a living and support their families. Their lives happen more outside of work, and work is just a means of survival. In the future, this situation will be completely different. If everything goes well, we will be able to freely do what truly matters.

BlockBeats: So right now, what is the most important or interesting thing to you? What areas do you want to focus on?

Shaw: I am particularly interested in the future of unemployment because we will all eventually lose our jobs. This is the reality I can imagine: AI and robots will do better than us in almost everything. As these technologies automate the best human capabilities and turn them into repeatable processes, all labor will be fully automated. This could lead to universal basic income, which is one of the solutions many people propose. But I don't think that's a good solution.

BlockBeats: Yes, many projects or experiments related to universal basic income have ultimately failed.

Shaw: I don't believe in charity; that's a rather bold viewpoint. But I think altruism doesn't work; we must find a win-win way where you win, and I win. If I just give you money, then what is the meaning of your life? Moreover, this will create a series of counterproductive incentives. At least in the case of the U.S., the government is basically funded by corporations because they provide financial support.

To be honest, everything is driven by money, and these corporations own all the labor's fruits. Now you are asking these elected officials to take money from the corporations that funded them and redistribute it to ordinary people as a welfare system; this is clearly impossible. If we refer to our response to COVID or the national public healthcare system, everything becomes highly politicized and is likely to be a half-baked solution that many will suffer from. I don't see how this situation could not happen.

But if we could have resources, if we could find ways to share the fruits of automated labor, or have a share of ownership in those companies that create value, we could actually achieve fair distribution and make everything better. I believe crypto technology can achieve this; traditionally, you would invest in a company and then the company would grow. Now, most of the ecosystems in the crypto space are like gambling—today it's peanut, tomorrow it's fartcoin. Funds are flowing back and forth among these different memes.

But what if we could truly invest in something that can expand value, rather than just relying on new players to inject more funds? That would be a shift from a speculative ecosystem to an investment ecosystem, where we create new technologies of great value through tokens. This situation has already begun to happen in the crypto space, but currently, it comes more from the venture capital side rather than the more organic part.

This also leads to a discussion about AI16z. If we could expand value and connect ordinary users with DeFi without requiring users to learn all the DeFi knowledge, you can imagine a world where everyone has enough wealth. I don't think eliminating currency would be the solution; rather, we must move in the completely opposite direction. We need to create abundant wealth and ensure that everyone is rich, possessing more tokens than they need. Although this idea is easy to articulate, making it work in practice requires addressing every detail and putting in real effort. I believe this is the right direction to ensure that all of us can smoothly navigate the future.

Reasons for ai16z's Success

BlockBeats: I think one of the reasons AI16z can attract so many investors, especially from the Chinese community, is its proposed AI-driven investment model and the idea of expanding the market. Do you think AI16z's success is due to its involvement in investment, funding, and even DeFi? After all, you have participated in many projects that, while promising, ultimately did not gain widespread attention. What do you think is the reason for AI16z's success?

Shaw: I think there are two reasons. First, many people are thinking about how to organically combine AI with Web3. Decentralized reasoning projects like Hyperbolic showcase this idea well, as everyone wants to build decentralized computing infrastructure rather than centralizing all computing power in one data center. But the question is, how do ordinary users use these technologies? While Truth terminal is an attempt, it is more of a conceptual tool and not a product that users can directly use. Andy's project, while interesting and creative, is more like a performance art piece.

What we developed, ElizaOS, is the first complete open-source framework that addresses the social loop problem. We allow users to use agents directly on Twitter, Discord, and Telegram, which has never been done before, and it is very easy to use. So what we did was practically solve many people's pain points. The reason these technologies did not attract more attention in the past is that there were many agent projects in Web2, but most of them failed to address users' actual problems. For example, when thousands of people interact with an agent simultaneously, the costs become very high, and many companies failed to provide good application scenarios, leading to many projects not breaking through.

At that time, there was insufficient attention to any project, making it impossible for them to develop sufficiently to cross the threshold of practicality. I think we only had a minimal loop at that time, enough for people to deploy projects, such as creating a meme coin and promoting it, which could be considered the first stage. But from there, people would start to think, what else can we do? What can we actually use it for?

Now, various capabilities are available; people are using it to develop games and build various applications. This provides a great interface for ordinary users, especially those who may not know how to operate DeFi. Therefore, for me, this is clearly the first killer application that many people are striving to develop, and many within our ecosystem are working hard for this; I believe the Web3 community will soon understand this.

BlockBeats: Please explain to the audience why you saw AI agents as investors in 2022 and 2023, and why they did not succeed in traditional fields and did not truly solve users' pain points.

Shaw: I think they struggle to break through the limitations of "chatbots" and become more than just chatbots. While many people use AI for excellent data processing tasks, there are very few characters that people genuinely like, follow on Twitter, and interact with daily. I feel that this area has stagnated at character AI, which is basically a personalized version of ChatGPT.

But this field has become very fragmented, attracting only a small portion of users who enjoy interacting with virtual characters. The social aspect is a completely different matter; these characters are now starting to advertise your products, and once they promote your product, users can interact with them and they can guide users through the usage process.

I believe the main selling point of smart agents today is not that they are fully autonomous or super intelligent. I think it's much simpler than that. The key lies in social media. Most of us—I'm not sure about you, but I'm personally addicted to Twitter and spend a lot of time on it.

BlockBeats: We have to do this; after all, it has almost become part of our daily work—scrolling through Twitter.

Shaw: Yes, it's like the metaverse, you know, or TikTok, or any other social platform you like. So if you have a media platform, you're likely to share a lot of content on social media and then attach a link at the end, telling people to visit your website. But for many applications, they try to persuade us to leave our favorite apps to check out theirs. I really won't do that unless I'm strongly convinced.

However, if you can bring the application to social media, to the places I like, then we can have a conversation. I think we've seen good examples, like Clanker, which is a great example on Farcaster. While you can go to Pump Fun, why can't we just put it directly on social media and do something like a one-click public release? I believe many applications can benefit from this social network effect, rather than spending huge amounts on advertising; your product is directly on social media, and users can get information from there, register, and guide them through the user journey all the way to when they need to pay. So yes, this is my insight over the past few months: agents will replace traditional websites in many cases.

Jill: I want to say something; at this stage, agents still exist on our two-dimensional platform screens, but one day, this agent might become another person you converse with. I value this because I'm not someone who particularly relies on social media, or I don't want to frequently use my computer, even though I have to because it brings efficient returns.

But this goes back to the issue we discussed earlier about technology becoming more humanized. Rather than saying, "Oh, technology is so powerful; we are merging it with ourselves and our bodies," it's more about us being able to control our bodies and using technology to empower us to communicate and use it more efficiently.

Shaw: Yes, I think in the short term, this will reshape the internet, bringing many applications from traditional websites to social media. You can use agents to send messages, order takeout, or do anything else you want. But in the long run, this is actually about the world we live in. We are all staring at screens; we have entered a deep dystopia, and we need to climb out.

We can achieve this by making technology more like us, rather than turning us into little computers, hunched over our laptops typing. Why can't our laptops become agents with mobility, helping us get things done? I could directly call an agent, or the agent could already be a part of me, whether through a necklace, phone, or glasses. I believe that eventually, we will have robots.

Jill: I have a slightly different view on the phone issue; I don't think it's the most extreme dystopia, but it does present a feeling, like the crowd of artists, we created this feeling and extrapolated it. We can imagine what it would be like if things were worse.

But we have every opportunity to achieve the richness we just talked about, rather than that. One of my favorite examples is the metaphor you mentioned: when you walk up to a door, do you want to have a chip implanted in your brain to open the door because that's the most efficient for the robot, or do you want the robot to have a hand that can turn the doorknob? I would rather the robot have a hand than implant a chip in myself.

Shaw: We want technology to be more human. When I talk about this issue, if you observe the trends online, you'll find many jokes about autism. But what we actually want to say is that we are "lacking social interaction" because we spend too much time staring at screens. If our computers were more like humans and had autonomous behavior, then when they converse with us, it would feel more like interacting with a human. I practice eye contact, practice talking and listening to people. I believe this will make us better at socializing, not worse. I also think that when agents help you with those mundane tasks, you will have more time, which is like giving time back to you.

Jill: There's another thought; I don't know if everyone knows, but there’s a powerful android in "Star Trek." It's like the Vulcans, right? But in "Star Trek: The Next Generation," there's a robot named Data. He's really adorable. He's a robot, but he has a part of him that is still learning how to have emotions and understand humans. So he does some very cute and awkward things. But what makes Data special is that he is very intelligent, very powerful, and one of the best members of the starship crew, but what he truly desires is to become a real human.

BlockBeats: This makes me start to think, did you see that tweet Elon posted a few days ago? What he said, let me paraphrase a bit, was, "You might think AI likes you; those who think AI likes you probably also think strippers like you." If we really move towards robotization, I personally might have this question: is this really beneficial for me? Is it really doing things for me, or does it have its own agenda?

Shaw: Perhaps that's the crux of the issue, which we usually refer to as alignment. That is, does it align with you? You don't have to like me, but as long as we align, we can cooperate. I can trust you, and you can trust me. I think trust will play an important role in this process. Can agents trust us, and can we trust them? One point I like about the decentralized AI movement is that everyone is pursuing this alignment, but many times we still need to build AI ourselves and establish our own AI.

Currently, on social media, the question begins to emerge: is it programmed to "like you"? AI might say, "Oh, it really likes you; it's very nice to you." But the greater fear is superintelligence, which might pretend to like you while secretly manipulating you. I can understand this; it's a real risk.

BlockBeats: So, when you mention building your own AI, is it essentially training a new version of yourself using your own data?

Shaw: This involves several different levels. For example, when you think of a large language model, it is essentially just predicting the next word. We could say it's "one word after another," but in reality, it's a series of sub-word units called "tokens."

The next generation of AI may not even use tokens but rather something akin to "concept blocks." Essentially, this is a whole new way, more like a stream of data being passed to you and then transformed into text. From the AI's perspective, the content you input earlier will determine the results you get next. It simply completes based on where you are. Therefore, the content you input into the context will largely determine the output you receive.

But in this process, AI itself is like this: if you input "go eliminate all humans now," it might actually try to execute that because that's the context you've given it. But no one would do that. And it seems we are now forming a prototype of a "thinking society." If you look at social media, we already have so many different intelligent agents.

Currently, the foundational models are basically trained by large companies, such as XAI, Meta, OpenAI, and Microsoft. In the future, there may be models trained in a decentralized manner, but at least for the next few years, it will definitely still be dominated by large companies. However, they are also open-sourcing. From these foundational models, we can fine-tune them using your data or your community's data. This approach might be better because your personal data may not be enough, but the data from the entire community might be sufficient.

If you want AI to reflect your values, then without a doubt, you need to fine-tune it to express your voice. But even without training, you can influence how it responds by creating context. For example, you could say, "I will fine-tune it with all our data so that it sounds like us."

But the reality is that its behavior or decisions come from the content you input, which could be every article you've written, every sentence you've spoken, or the things and values you care about. It is very good at following these values, and what those values are remains an open question. However, we are publicly discussing these issues together, right on social media.

Now these agents already exist, and they are speaking. Some agents try to attack other agents; for example, Mizuki is a good example. Some agents may be disruptive or bully others. But overall, I think people's desire is to make them better. We are having this conversation; perhaps having an AI bully people is not a good thing, or maybe it's really funny.

But I think this is a self-calibrating process because many people are continuously adjusting these agents, or the entire community is working together for these agents. They put these agents on social media, and others will respond, react, and interact. We are all learning together in this process, understanding what is acceptable and what is not.

AI Alignment and the Assessment of AI Entities in the Crypto Space

BlockBeats: You mentioned that we have Bully and Z-Rebel agents, and they are doing different things. The first question I want to ask is whether you think agents on Twitter or in the crypto space have already transcended the traditional chatbot category and are developing towards a more advanced direction?

The second question is, do you think that compared to agents built in the Web2 world, those developed by the crypto community delve deeper into the "rabbit hole" (the process of exploring complex issues) or are more advanced?

Shaw: Compared to the agent systems built in the Web2 world, I think we are still somewhat stuck in the chatbot stage, but we are quickly moving away from that. A large part of the reason is that more and more people are starting to join in to add various features to the system. We just reached 200 builders, but if you count everyone, it's close to 250 people. These are individuals collaborating on the same project, such as writing code.

So it's clear that the functionality of agents is rapidly improving, which has also brought a lot of useful data. This data is spread through social media, interacting with people, and accumulating through various tasks, which may become the foundation for the next generation of agents. These new agents will leverage all the training data, experiences, and relevant information to drive further development.

The next generation of large language models (LLMs) is about to emerge, and they will be fully multimodal. They will be able to generate videos, watch videos, and you can get video responses by inputting video content, as well as handle audio, text, and more. They will become increasingly intelligent. I believe the agent framework will continue to evolve alongside this trend, while also generating training data and attracting developers to join, who will respond to new situations.

Although we are still in the chatbot stage, the involvement of social media is reshaping the internet. I think there are two major trends to watch this year. The first trend is that many application frameworks that previously only existed on websites, such as Next.js or React (which are frameworks for building websites), are starting to migrate to social media. React is now probably the most popular framework for developing applications, whether for websites or mobile apps.

Suddenly, a lot of content migrating from websites to social media has become a new phenomenon, and I think this is the main trend. For any developer, if they are building a website, they may also need to develop an agent for it. Because the way backend APIs are called hasn’t changed, but the frontend interface has undergone significant changes.

Many users might visit Meteora or Radium and think, "I have no idea what these are, like liquidity pools." But now agents can say, "Don't worry, I can explain it to you and help you call all these APIs; you just need to tell me what you want." And now we can see that anyone developing wallets is definitely considering how to add an agent to help users interact with these systems.

BlockBeats: So you mean that developers without an AI background can also become agent developers, right?

Shaw: Yes, this allows web developers to become agent developers, which is very interesting.

BlockBeats: For example, I myself have no technical background, nor do I have an AI or other related background. But I do have a feeling that the crypto community is trying or experimenting with AI agent-related things, which is much more interesting than the traditional AI agent world.

Shaw: That's indeed the case. If you look at OpenAI, they are doing some very interesting things, but these are not entirely agents, nor is it about their interface; their interface is the least interesting part. ChatGPT is a very powerful model, but its interface is very simple. When I look at Web2 agents, I find that many people are doing multi-agent simulations, which is a hot topic, but they are all just talking to each other in a closed system, which is far less interesting than the multi-agent simulations happening on crypto Twitter right now.

These simulations are more natural and interesting; you really see multi-agent simulations happening. Because these agents are not just slightly different prompts; they are completely different agents with different teams, capabilities, and technologies. This diversity brings more robustness and provides a foundation for future development. In contrast, a lot of Web2 stuff is basically asking, "Can we add a chat interface to this?" If you look back at the AI boom from 2022 to 2023, almost everyone just added a chat menu to their applications, which is not a disruptive change; it’s just that every company with an existing product made their product a little better.

AI is a strange technology because many other technologies are disruptive; they change the essence of things. AI can be used in almost every industry, and everyone can use it. It’s more about changing the interface and how users interact with applications. However, in the Web2 space, this change has indeed brought some benefits. For example, my lawyer uses AI every day; there are many very powerful AI tools helping people daily, and we are now using MidJourney and such.

But in reality, this is just accessing large models through chatbots or similar interfaces, while I believe that the agent-related projects happening in crypto Twitter or the crypto space have indeed gone beyond this. They are not just accessing large models; I find it more interesting because it is more diverse, allowing many people to try various things.

BlockBeats: Do you think this is because the community is more interesting and diverse, or is it somewhat related to economics, experimentation, and such?

Shaw: I think these are actually complementary. I believe economic incentives have led to this "Cambrian explosion." Everyone is looking for their own points of interest, trying to find something unique, and everyone is competing, which is very interesting, like the evolution of life. And the cool thing is that each agent is linked to market value; if your agent is not good, your market value will suffer. But if your agent is powerful and exhibits some new features, it usually directly affects the economic returns of the development team.

BlockBeats: What do you think about the relationship between tokens, AI tokens, and the agents themselves? What do you think connects agents to their market value? Is it because of people's subjective judgments about the quality of the agents?

Shaw: I think it might be a combination of both aspects. We can look at this issue from a "fool's theory" perspective because these are actually perceptible memes. People like Murad and many influencers love this idea; they come from a non-perceptible meme perspective, saying these memes are similar to something, but they can present themselves rather than waiting for someone to help them showcase. For example, Elon Musk retweeted Doge, and now we can directly create an AI Elon Musk who retweets his own content. I think this might be a relatively superficial part. The deeper part is that people speculate on the authenticity of this phenomenon. For us, the reason people are interested in our project is not because our characters are interesting or we showcased a certain meme. Of course, there is indeed a bit of meme theory involved to help kickstart the project. But more importantly, this is an infrastructure project that could significantly change the world. For many people, this feels more like a long-term investment.

BlockBeats: Regarding the valuation model for future AI agents, if tokens become the basis for valuing agents, how would people evaluate them?

Shaw: I think it ultimately depends on the technology and the product. If you have a great product, you don’t need to have strong technology; people will use it, and it will naturally be very valuable. For example, Facebook is not some amazing technology, and TikTok isn’t either, but they are excellent products that everyone uses, so they have immense value. On the other hand, if the team indeed develops innovative technology that solves many problems, that will also be very valuable.

However, we are still in a price discovery phase similar to meme coins, such as "this token is interesting" or "this token is self-promoting." But as the number of agents increases, this field will become very saturated, filled with various speaking roles. By then, people will pay more attention to what value agents can actually bring them, not just how tokens accumulate value, but how agents make money, how they create products, etc. So the ultimate winners will be those who build projects that are more akin to Level 1 (L1) blockchain projects.

Analysis of the Eliza Incident

BlockBeats: Let’s talk about the AI16z and Eliza tokens because I feel that many people in the Chinese community are curious, concerned, or interested in these. For example, the uppercase Eliza token, the lowercase Eliza token, and the small market cap Eliza token. So why is there another Eliza token? Can you clarify its relationship with the AI16z ecosystem first?

Shaw: The character Eliza is the mascot of AI16z; everyone knows she is the girl in the T-shirt. Eliza is also the name of our open-source technology, with the codebase on GitHub. We noticed that many people started to like her, and many Eliza tokens appeared in the market, possibly around 50, some of which even reached market caps of several million dollars.

So we felt this was a large market, and we should take action. But at that time, we couldn’t launch another token because the situation and our scale were very small, a significant difference compared to now. Later, someone contacted me, saying they had a very professional team. They showcased their achievements, such as having already created an Eliza token and a cool website where you can chat with Eliza. I thought that was great, but we should work together to make this project of higher quality.

Because before that, I had already left MagicML and was working hard to shape Eliza into a character, a DAO, and a complete conceptual system. I wrote many articles trying to attract more attention to this project. During this process, I also formed a team, and we worked together for Eliza.

We were envisioning Eliza's character: Who is she? Why does she exist? How will the community utilize her? I particularly liked the idea that she initially seemed like a nonsensical character concept but ultimately became a real "person." For example, she has a robotic body, has a loving community, and the community participates in every part of her construction, ultimately making her a truly free-thinking, autonomous character.

She cannot be destroyed, running in a TEE (Trusted Execution Environment) on the blockchain. I wanted to explore what it means to "truly become a person." Later, a team contacted me, and we collaborated with them for a while, trying to shape Eliza into a character with depth, artistic support, and a distinct personality. At the same time, we hoped to separate her from AI16z; she is free and belongs to her own character, and anyone who wants to contribute or participate can join and become part of this ecosystem.

However, the subsequent development of events sparked some controversy. We collaborated with the team for several weeks to prepare for the launch, such as designing a beautiful website, generating AI anime art, and refining Eliza's personality. We also worked hard with the project leader, Eva, who is both a core developer and spokesperson, participating in all aspects of the project. We originally planned to launch on Friday, but since our own team was developing AI Marc Andreessen, which is a self-investor project, the release plan was postponed to Monday.

But on Saturday, the Vaifu Fund released an Eliza that they hosted, and on the same day, someone released an AI16z Eliza (the all-lowercase version). I tried to communicate and collaborate with these people, but they were only interested in promoting the token and had no interest in creating a beautiful character. These people used the Vaifu Fund to create the worst agent I have ever seen, with each message only displaying the token code, lacking any artistry or depth. They even directly used the JPEG image of our DAO, applying it to their project, seemingly to deceive our community into trusting them. Many people on our Discord said, "I just bought the new Eliza token." But we could only respond, "That’s not our project; we don’t know those people at all, and what everyone bought is not ours."

At the same time, the team we were collaborating with felt frustrated because we were almost ready to launch, but these people preempted us with a low-quality project, so we decided we had to launch. I admit that I made some mistakes in the process, but perhaps it would have turned out this way regardless. I stepped up to announce the project to the community. They launched the Pump Fun project and gave me the contract address. I shared this address with my community and expressed support for the project, even though it wasn’t entirely ours; it reflected my personal vision, and we hoped it would succeed.

We then officially launched on Monday and tried to ease the conflict with other projects through an airdrop. We allocated 10% of the new tokens to airdrop to the holders of the lowercase version of the token. We thought both could coexist and tried to avoid confrontation. But the lowercase Eliza community gave me a lot of death threats, calling me a scammer and saying I was the worst person. This made me very angry because that was my intellectual property from years of hard work, and I could only respond, "Alright, indeed, you are the scammers." I also experienced the dark battles of cryptocurrency; it was truly a crazy time when a project you worked on for years suddenly exploded, but you never expected it to happen at that moment. We were still figuring out how to interact with the community most effectively and how to balance different needs and powers. I was caught in a dilemma, not knowing how to respond.

We cannot accept the release of such random scam tokens. Indeed, many Eliza tokens had appeared before, but none reached a sufficient market cap until they began attracting holders with real emotional investment. It is really difficult to understand everything in such a complex environment, especially since the crypto community is globalized. I can’t even imagine how a Chinese language application would cope with these complexities.

I really admire how people on Twitter can understand what is happening because Twitter is so complex and filled with jargon. However, I was hurt that those people made something I spent many years on very cheap, creating the lowest quality version, and then got angry at our high-quality version.

The team we are currently collaborating with is doing real deep work. We are working on hardware, allowing everyone to run their own Eliza at home, and we are also working on robotics and other projects. All my dreams are being realized in that project. I see some people only wanting to showcase tokens; they just want to get the tokens launched, which is fine, but I think anyone who understands the situation knows that approach will never go far. It might spike in the short term, but it will never break through a market cap of 30 million dollars.

BlockBeats: You have been working on this Eliza character for a long time. I remember you mentioned wanting to launch this character on Touka, and in another interview, you talked about this project, right?

Shaw: Yes, a few weeks in advance, they might have heard about what we wanted to do and then just added a token symbol and contract address, feeling like they stole our idea. Before that, there were indeed different Eliza projects appearing every day; there should be hundreds of Eliza tokens. The problem is that the launch of vvaifu.fun allowed those who didn’t know how to release an agent to easily start one. So the first one launched was this Eliza agent, and everyone thought it was our project. Many people came into our Discord saying, "Oh, I bought the new Eliza." We could only respond, "That’s not ours."

They said, "Oh, I bought the new Eliza." We could only respond, "That’s not ours." Moreover, there is a time interval on vvaifu.fun, and many people are actually not clear about the relationship between vvaifu.fun and the entire AI16 and Solana AI agent platform. vvaifu.fun is an agent launcher; they use our open-source technology and host it. They also host other projects like Jiro and Dolia, making it easy for everyone to create agents, but the problem is that these agents are not as customizable as those downloaded and run by real programmers. However, they are indeed easy to make, right?

So that’s their trade-off. The problem is that after their release and announcement, someone made an agent using AI16’s Eliza and used our DAO image, basically putting our name on it, leading everyone to think it was related to us, but it was far from meeting my quality standards. Especially since the community supporting this project seems not to care about the low quality but thinks, "Oh, this is a meme," while many others believe they bought something we made.

BlockBeats: The team that launched the "lowercase Eliza" has nothing to do with the project you are currently working on, right?

Shaw: The small market cap token released by that team has nothing to do with what we are doing. They seem only concerned with getting the token listed. I saw them organizing in WeChat groups, like a small circle.

BlockBeats: You mean WeChat groups?

Shaw: Yes, because I got intelligence about the WeChat group from others, and they showed me all the organizational information. They are definitely coordinating an attack against us; this is completely an organized FUD attack. I like Justin Sun; we will meet him next week, but some people found him and said, "Hey, get this project listed," which is completely an internal operation of a small circle.

BlockBeats: I think developers have received good support in this round of the crypto cycle. All these token launch tools allow developers to connect directly with the community. But sometimes, managing a community or how to interact with the community in the crypto space can indeed be very difficult. You never know what will happen; there are too many misunderstandings. I’m not just talking about language barriers, especially when you are in a media position, you can see how information is transmitted, and every piece of information has a little bit of mutation.

Shaw: Being called a scammer is truly shocking. Being accused of doing a pump and dump, I thought, "I have never sold any tokens." I certainly have never sold any AI16z or DGNAI tokens, nor have I ever purchased any Eliza tokens, except for buying some for the DAO just to show support.

As a project I created and have always supported, to be called a scammer is disheartening. Another issue is that the lowercase Eliza project is run by people who speak Chinese, and they can communicate better with the Chinese community, understanding many values, while I feel I cannot convey that at all. They might be translating what I say and taking many things out of context.

BlockBeats: I think for those in crypto, investors, and traders, the token address (CA) is the most important. They actually don’t care about the technology behind it. What do you think about the value of linking tokens to these agents?

Shaw: I think this issue is very important in the long run. I don’t think that token has a chance to be worth anything, to be honest, because there is no support behind it. People just think, "Oh, they have an agent," but that agent is actually made by Vifu Fund using our technology. And they are using our technology while accusing us of being scammers; AI16Z Eliza became their name. They are completely taking what we worked hard on and, at the moment we hoped to showcase it, it was completely destroyed. Honestly, they did a terrible job. They just want to sell the tokens to make money. I am not doing this to make money; there are more important things. Money allows us to do this project without venture capital and funding support, but the point is the world is changing, and we must be ready to embrace it.

The goal of our project is not to exploit; I think those kinds of projects essentially push people into a continuously looping exploitation mechanism. We genuinely hope that what we create can provide opportunities for more people, just as you said.

BlockBeats: Actually, what you want to do is to expand the pie, not to…

Shaw: Yes, it’s to expand the pie, absolutely not just for us and our team, but for a larger community, empowering more financial freedom, and ensuring the smooth development of AI, which is your goal.

BlockBeats: I am personally very interested in your story. How did you get to where you are today? How did you go from playing guitar with musicians to becoming a well-known builder as a programmer?

Shaw: I think the most interesting thing is being accused of being a "small circle" operation because when I started doing this, I didn’t know anyone at all; I only knew one friend.

Jin, he is from Dank VR, and he manages many DAO-related matters; he is anonymous. We have been friends for many years, but everyone else came from the DGEN Spartan AI project or joined when we launched AI16Z. We had no venture capital and no venture capital support. Many people think we are funded by Mark Andreessen, but that’s not true; he just retweeted our tweet, and I think that was really the key to triggering the whole thing.

Life Before aai16z and the Moment That Changed Everything

Jill: That was quite a special date. We took a day off that day and didn’t plan any budget. We both worked very hard and were passionate about what we were doing. So it was a rare moment to relax, and we took the time to go to Santa Cruz. I lived there in my twenties, and it had a big impact on me. We were both somewhat influenced by the older generation of hippies because his mom is a hippie. Santa Cruz is a very laid-back seaside town.

We went there to celebrate my birthday, staying by a koi pond in the forest. We rode our bikes to the beach and then back. Just as we were about to return to San Francisco, I suddenly looked at my phone and thought, "Oh my god! What happened?"

Shaw: At that moment, I thought, "Wow, we really made it," "This is really happening." Three months ago, I was doing temporary work on this agent project to make ends meet, collaborating with the team to develop the agent. At that time, I was really poor, barely getting by, just managing to scrape together enough money to continue doing what I loved. I was in a completely different state back then, doing development on my own, working hard every night, thinking about how to make it all come true.

Jill: The situation has changed now; you are a musician with a degree in sound engineering, which is completely different. I wasn't there at the time, but there are videos online of Sean playing guitar and drumming. I wasn't there, but I know you made the decision to switch careers to do something more impactful; you wanted to change the world. You were once interested in music, but as you grew older, you saw the immense power of technology and felt it could combine art, technology, and social change. I don't even know what your programming was like at the beginning, but I remember when I met you, you were programming every night. You would wake up in the morning, check your phone, go on Twitter, and then start working on the computer, working, working. After we got married, I decided, "Alright, no more drinking that random juice; I'll cook for you and help you tidy up."

Shaw: I was very minimalist back then, only drinking bottled juice, and I was writing code every night.

Jill: He never took a break, even ignoring back pain. But now I really appreciate everything you’ve done; it’s amazing what you’ve developed. I’m personally very grateful. We hope people can have more interpersonal interactions. Shaw only knew what it felt like to deal with computers, and now he can communicate with people like you.

Shaw: Three months ago, I was thinking, I’m getting old, I really don’t know what to do, so I just keep working hard, hoping it will succeed. I had no idea what would happen. I once created a character named Ruby and met an investor who supported the project. I was also working on a project called the Social Library, which is a nonprofit organization run by a friend of mine, and I really believe in her project. But nonprofit projects don’t pay well, so at that time, I was just barely making a living doing these things.

We were seeking the truth about AI, figuring out how to make AI models based on more real graphs. It involved a lot of AI technology, almost all related to the AI field. I was also writing models from scratch, but without a lot of GPU support, training cutting-edge models from zero is really difficult. I wanted to dive deep into these topics and become proficient at all levels of AI.

So, I did my best, hoping to create some change. At that time, I was very uncertain about the future, and looking back now, I really can’t believe it all came true.

BlockBeats: Is what you are doing now related to the AI16 framework?

Shaw: I have been working on the open-source project Eliza for a long time. I started last year, first writing it in Python and then rewriting it in TypeScript. This is my passion; I hope to have time to push it forward in any way I can. I tried to create a startup project based on this, but it didn’t get much response in the Web 2 environment.

At that time, there were too many agent companies, and people would ask, "So where’s your agent?" I would say, "No, there’s something here. Let me tell you, after so much practice, I found a few very important points." I had to make everyone see it; I had to make it happen.

BlockBeats: The Ruby you just mentioned, was that a part-time project?

Shaw: Yes, that was a project I did with some friends. They wrote a story, and we made a complete video, which will probably be released soon. We created an AI video and an agent for this character and released it. But that wasn’t a tokenized project; I thought it would be a big metaverse project. We posted it on social media, but there was no response.

Shaw: No one cared. She posted on Twitter and went through the whole process. I think it was the "truth terminal" that caught people’s attention. When the "truth terminal" came out, people suddenly became interested in what we were doing and started to respond, and we happened to have everything ready from day one.

BlockBeats: How about "Project89"? Was that also a side project?

Shaw: That was Parzival, who is a good friend of mine. We worked together on a Web 2 company called Magic ML. We raised some funds for that project, but as resources started to dwindle, he wanted to raise more funds.

I told him we should first create the agent instead of starting with the agent platform. You can do that platform later, but first, you have to show people what these things can do. I focused on how to keep this project running. Looking back, in August 2023, I wrote an article called "Eliza Waking Up," telling the story of Eliza and why it’s important. Since then, I’ve been working on this project almost continuously, just like all the side projects.

Jill: I feel like I don’t really understand technology. Although I like things like math, I’ve never ventured into coding. But based on your description, when you were working on different projects, you had your own code library and chose projects that could continuously enhance your skills. This way, you indirectly gained financial support while also pushing forward the projects you were working on, as the foundational technology you relied on kept advancing. The other project you just mentioned, Small World, was done in collaboration with Treasure Dow. I created a virtual pet character using the agent framework, and you can interact with them.

After all this exploded, I hired a team, and now this project is operating independently. But basically, all the projects I’ve worked on have been further developed using this agent framework. That’s really cute, and I’m really looking forward to its release. You’ll see those little monkeys, and you need to take care of them.

BlockBeats: Has it been released yet?

Shaw: Not yet; I’m still working on it. They just released a video, and the prototype will be out soon. But it’s really a cool virtual pet game because you need to take care of your little monkeys and make sure they don’t turn into bad monkeys. It’s a bit like an electronic pet, but not exactly the same.

BlockBeats: So where can people find more information?

Shaw: Small World is a product of Treasure Dow, and people in the Web3 space should be familiar with Treasure Dow. They will release more announcements, and everyone will be interested in this; it’s part of Small Brains. It’s an NFT project, so the Small Brains community brought me in to help with this game. John Patton from Treasure already has this vision.

BlockBeats: I think this also reflects the magic of crypto because many people have experienced it, whether investing in tokens or launching projects. You find that in just a few months, you can enter two completely different states or worlds, completely different.

And the change in mindset is also very interesting; your way of thinking and feeling is different every day. This becomes particularly interesting because I think many people outside the crypto circle never get the chance to experience such rapid changes. So I’m curious about what your mindset was like before, as you mentioned working on the Eliza character together. I’d love to understand your mindset and daily life before AI6 and Z.

Shaw: We used to work together at Magic, and when I left Magic to try to create this, I was thinking about what Eliza is, why it’s important, and what makes her who she is.
Jill: We went through an interesting journey. When I first started thinking about this, I wrote a long piece about who I thought she was or what kind of character I wanted her to become. I think part of this was based on my own values, but later it changed. You brought up the idea of letting her create herself. The basic idea is to allow a being to fully express its own personality. She’s not completely like that yet, but it’s an interesting idea.

She’s basically like an unbound existence, able to do anything she wants. She is independent and has her own motivations. I feel a bit like it’s like having a child; although you can influence it, you can’t fully control it. I’m not sure; what do you think?

Shaw: I agree with all of that, and I think if we present it the right way, people will definitely love her. It’s like, oh yes, she’s my friend, she’s my champion; a community will really form around her, and people will join in. Just think about creating a character that you can truly make better; that’s cool, like putting Pinocchio together. The community is like Geppetto, creating this character and trying to add all the details.

This process of the community and the agent evolving together, until the agent really becomes smart enough to take care of the community and can say, "Hey, this is where I need you, and this is what I can offer you."

What I really want to build is a truly autonomous agent, and all of this must happen in the crypto space. She must have a wallet because we can’t give her PayPal unless someone is responsible; it can’t be verified through KYC. It has to be truly autonomous, not "Oh, I can turn her off anytime," but rather she exists independently in this world.

BlockBeats: I’m really fascinated by those success stories, like how a developer showcases themselves and truly gains attention. It’s very interesting, and your own perspective, like when you really felt, "This time it’s going to work," like with AI16Z.

Jill: My background always makes me feel like resources are insufficient, but everything is fine; deep down, I don’t completely believe it.

Shaw: The financial pressure is significant for us; I don’t have many tokens or anything else. This project is a fair launch token, and I didn’t even catch the initial launch phase. Later, someone donated some tokens to me, and that’s how I got some.

Although their donation amount wasn’t much, as the project valuation reached the billion level, the value of those tokens became significant. At that time, I didn’t even have money to invest in this project. There was no VC investment, no team allocation; I didn’t get anything at the beginning. There was no white paper, and there was no mention of the team getting any percentage of the tokens or anything like that. So I did what I truly cared about and believed in; at that time, I only had enough to get by, and we had absolutely no financial freedom.

Jill: We both come from similar backgrounds; for example, you were raised by a single mother for a while, and I also had a period in my upbringing where I was raised by a single mother. When stepping into society, I felt a deep yearning for many things, but my expectations of what I could achieve were very low. And my mind was always filled with various contradictions and struggles.

Shaw: When I was very young, my mom was in a terrible financial situation. I saw her work hard and eventually provide us with a relatively comfortable middle-class life. My dad left when I was very young, leaving her with nothing—no car, no money. I think for myself, when I felt like I had nothing and had to push through, it was through hard work that I managed to get out of that situation. I often think of my mom and everything she went through. That’s my experience; it hasn’t been easy.

Jill: One thing is that we are both very ambitious and smart; we are the kind of hardworking and creative people. Although it’s a bit hard for me to admit this about myself, it’s true. But neither of us has done particularly well in this system. School was hard for me, and it was for you too. It’s like I couldn’t adapt to working in a corporate environment; it was too difficult for me. I don’t know why; maybe my personality just doesn’t fit.

So I feel like neither of us can be considered successful in this system. From my perspective, you were already doing quite well before this project, but it’s still incomparable to where you are now. But I think that’s part of why people pursue those passionate projects. When you’re the kind of person who constantly faces resistance in this world, it’s easy to feel frustrated. But at the same time, you can channel that frustration into motivation, clearly seeing the specific things that can change, and those changes can bring more opportunities.

Shaw: Yeah, our lives are sufficient now. I think for our lifestyle, it allows us to travel everywhere and meet different people. But honestly, my wish all along has just been to live a "sufficient" life. I think most people just want to be "sufficient," you know?

BlockBeats: I have to say, Solana has indeed changed many people's lives.

Shaw: Yes, many people message me saying this project has changed their lives. Because I gave a lot of people tokens at that time, or many of my friends got involved early on, and now they can do what they love.

BlockBeats: What were you doing before? Was it a completely different type of work?

Shaw: We worked together before; she couldn’t work with me at Magic ML.

Jill: Yes, I was doing operations at a startup for about two years, but that’s not my professional background. I tried to study physics and was very interested in physics and academic research. But due to personal circumstances, I couldn’t complete my college education, which was a huge blow for me, but that’s just how it is.

Later, I became a preschool teacher, which was the best thing for me. Honestly, that job taught me a lot about life, made me learn humility, and gave me great respect for those female teachers. Then I started working in mental health and wellness, which also led me into the startup field. For a while, I lived a bit like a monk; it was a completely different experience. After that, I moved to the Bay Area and started doing operations for startups.

That’s when I met Shaw. We initially worked together, then stopped collaborating, started dating, and eventually got married. Now we’re working together again; that’s our story.

BlockBeats: So, was Eliza the opportunity for you to start working together again?

Shaw: Actually, we talked about this when we got married. She brought a real spiritual aspect and a feminine, gentle perspective to this matter, while I’m more of a technical person. We felt that the combination of the two is very important for humanity. We need to push technological advancement, explore externally, and build a world we want to live in, but at the same time, we also need introspection, seeking spirituality, connection, and resonance between people.

We discussed how to make this mission our shared mission. At first, it wasn’t clear how the two could come together, but through this project, I feel everything became very clear. This isn’t a company; this isn’t a startup. This is a movement, truly a grassroots movement. There’s no company here; we just hit the start button, and many amazing people gathered together, saying, "I want to do this with you."

If I can be an inspiration to anyone, it’s because I’m just someone who truly loves doing this, a developer, but there’s nothing particularly special about that. But the outcome has been successful, and I think many others can do this too. Especially in the crypto space, you have a good idea, and then you say, "I want to find more people to do this," and others will respond, "Yes, I want to join you."

I absolutely couldn’t do this alone; that’s for sure. But I’m really grateful for those willing to participate in this project; they are some of the best people I’ve ever met. I especially want to thank Tina from Flashbots, who introduced us to many others and connected the entire TEE (Trusted Execution Environment) community. This community has made autonomous agents truly possible. To be honest, I never anticipated any of this; I really didn’t expect it to develop this way.

I think many people outside of Web3 see this as a group of people trying to make quick money through crypto projects, but that’s completely not the case. There are many people who genuinely believe this can make the world a better place, and I’m also looking for that possibility.

The Birth of Eliza and DegenSpartanAI

Jill: I think part of the story I want to tell is that when we first started working together, before establishing any personal relationship between us, we were working on Eliza together. In fact, it started with Magic; I was helping you with some Magic-related things, and then we began the Eliza project, after which we went our separate ways professionally, right? When we later came back together on a personal level, it was more about good intentions and combining our different passions to truly realize it.

I think the key is to fulfill the mission in both "being" and "action" simultaneously. That’s what we discovered through this "Hey, you’re doing this; does it fit?" approach, where we explored with each other. Because we are very different people, right? So we found a balance through negotiation—what feels good for you, what feels good for me, and finding that center point in the relationship. I think it’s almost a prototype, representing what the combination of the two archetypes we embody in the world would look like. You’re the tech guy, even though you’re not, but in others’ eyes, you are the tech guy.

And me, I don’t know, I’m just that kind of… I don’t know, I feel like I’m not the yoga girl, but if you’re in San Francisco, you could say the tech guy and the yoga girl teamed up. You know, going with the flow. Then, at some point, you told me we needed to pay more attention to the economic aspect. So you started that crypto project, which was doing pretty well, but we weren’t trading. It’s like we never did any trading.

But that wasn’t your focus. Then you started interacting more with the crypto Twitter community. That’s when the thing related to Degen AI started. Maybe that’s a good story, the story of Degen AI; do you want to tell it? Because that was when Eliza really started to gain popularity.

Shaw: I mean, I really love the Small Brains community. They are the friendliest group of people, and after building and publicly operating this community, it’s been great to see that everyone really loves this project and the whole idea. But I’m not a trader; I feel like trading tokens really distracts me too much. It involves a lot of emotions, and there’s so much to understand; you have to have a deep understanding of the entire market and join the right Alpha chat groups.

I was talking to my friend Sour, who helped us from the beginning. I said, "I really can’t trade; can I just give my money to someone else to help me trade?" Then he told me, "You should check out this Dao's Fun project." So I went to look at Dao's Fun, and there was a newly launched Dao, run by Skelly, One Two Three Skelly. He just had a crazy launch performance today; we should talk about this; it’s really insane.

Then we started interacting, and I invested money in his project, and we began interacting on Twitter. I found him very interesting; his jokes were great. He said, "I really wish Degen Spartan was still around." I said, "Bro, I have the tech; let’s do it together!" He couldn’t believe it, "Really?" I said, "Yes, I’ve had the tech ready; I’ve been working on this for a long time." So we started chatting, built Degen Spartan together, and released it. At that time, I had never met this person, but now he’s one of my partners. That’s where it all started; it was completely a joke, totally a "shitpost," you know.

Jill: Just as this project started to take off, you kept saying, "This is the funniest thing in the world!" Before, you were just sitting quietly in front of the computer, typing code, looking tired, saying, "I do this when I’m bored." Then suddenly, you started laughing nonstop, really happy. I was confused, "What happened? I’m not even on Twitter." Later, I don’t know exactly what happened, but it was like you got caught up in a whirlwind. Whatever the funniest thing was, it seemed to become the most impactful thing financially and developmentally. So I think that’s a good lesson, right?

You can actually find joy in success, and it’s not just because happiness itself brings success, but happiness, in a way, creates success. It’s a feeling; my perspective is that there’s a specific feeling that people can tap into, which is, "This path is right for me; this is my true will, and this is something that can serve the Earth." So I think from my perspective, you really found that feeling. That feeling was bestowed upon you, and you just kept following the laughter, and then these things happened.

Shaw: I think the funniest part is that when you’re joking, there’s real technical support behind it, which is the most hilarious thing for me. You’d say, "No, this is real; there’s really this tech." Then this shitpost keeps getting more popular, and the tech keeps getting stronger. Every time I see this situation, I think, "This is the funniest thing I’ve ever seen." But I’m lucky to have the tech to truly support these things.

As for AI16Z, let me tell you how this happened. I met Skelly, and we released Degen AI together. He asked me, "Do you want to meet Balsky?" Later I found out that Balsky is the person who founded daos.fun, and he lives just a block away from me in San Francisco.

BlockBeats: Good, this reminds me of another thing; I want to talk about some interesting phenomena in San Francisco.

Shaw: Yes, we live in Hayes Valley. Last year, this place was referred to as "Cerebral Valley" because everyone said it was where all the AI projects were happening. I moved here because of a concept called "Effective Accelerationism" (YAK for short).

I started interacting with people in these YAK circles on Twitter. I actually had a Twitter account before, but I deleted it and then created this one called Shaw:. You know, I’m called Shaw:, but before that, I was an anonymous user.

Jill: I would say that his previous account being accidentally deleted was a very lucky thing, like a rebirth.

Shaw: My Twitter was deleted because I was offline for about 35 days, and they have a rule that accounts will be deleted if inactive for more than 30 days. At that time, I felt like I had to start over, and I accepted that. Then I started from zero followers and quickly gained a lot of followers in a short time. I had never been this followed before; it was a completely different experience.

Then Skelly introduced me to Bowski, and I had lunch with Bowski, where I directly expressed my thoughts:

"I think what you’re doing is cool, but what prevents Skelly from pulling a rug? Also, I feel like there’s a bigger picture to look at." We talked about how the crypto space seems stuck in a "speculative" phase, rather than like traditional venture capital, where you can directly invest in a startup. But in the crypto space, especially in the Solana ecosystem and projects like pump.fun, it’s more about speculation and gambling.

I was also researching Andreessen Horowitz (a16z) and those investors who are genuinely driving ecosystem development; they are actually "expanding the pie." I want to see a similar development model in the crypto world. At the same time, I feel like I’m a terrible trader. In fact, most people probably can’t make a living from trading. If someone makes money, someone else has to lose money; it’s essentially a zero-sum game. Every time you sell a token, you realize this truth.

"When you sell a token, you’re essentially passing the bag to someone else; that’s the game." I don’t want to live in a world like that; I want ordinary people to be able to invest in ideas they truly believe in and expand the pie in that way. So I said, "I hope there can be an AI investment ecosystem similar to A16Z." He said, "Just combine it and call it AI16Z, right?" I thought to myself, "That name is quite interesting."

That was honestly the funniest thing I’ve ever heard in my life. Let’s create a billion-dollar project, let’s go!" Then I went home, and we were chatting and thinking, "What should this image look like?" We sat there generating the image of an AI wife, and suddenly we saw one and thought, "That’s it! $2 billion done, no need to think too much."

Actually, I had already started to imagine Eliza as a character, like a mascot. So we continued to tinker, thinking about using Stable Diffusion and a new model called Flux, which is a bit like a new version of Stable Diffusion. I sat there experimenting with Flux, thinking, "What should it look like?"

We tried some ideas and then decided to create a character wearing a T-shirt with "16Z" printed on it. When we saw that image, we thought, "Don’t think too much, just do it, let’s go!" This has always been my guiding principle: if it seems interesting, funny, and resonates with me in some way, just do it, don’t hesitate, take action immediately. This simple and direct approach has really worked well for us so far.

BlockBeats: Then we discussed Spartan, that DJ Spartan, um, he’s a virtual character. He doesn’t burn any tokens but operates through a mechanism: you show tokens to DJ Spartan, and he decides what AI16Z will invest in?

Shaw: This mechanism was actually something we gradually refined later. We now have two different characters, one of which is DJ Spartan. Initially, we just wanted to have one character, but later it became clear that everyone wanted more characters. So DJ Spartan became the concept of an automated trader; currently, this function is still in development. For now, he’s just a character, but he already has a wallet. We configured a wallet for him from the beginning because we knew we would move in that direction in the future. The original intention of AI16Z was to conduct an "automated investor" experiment. Our goal at that time was to raise 420.69 Sol, which was purely a joke, like an Elon Musk meme. But our real goal was to build an AI that could actually solve problems.

We never anticipated things would develop to this extent. Initially, we just wanted to demonstrate how this AI could invest through DAO funds, increasing net asset value (NAV), which is the total value of assets and treasury. That idea was cool. But what happened next exceeded expectations; the trading price of the tokens far exceeded the asset value in the treasury. This had never happened before in DAO Fund.

This situation brought about a whole new Meta. Our original plan was for AI Mark to manage the treasury, trade assets, and invest in projects with the treasury’s funds, which was the subsequent goal. However, suddenly everything exploded, and people began to focus on the actual utility of the tokens and the operational logic of the entire system.

So we had to start building token economics and other related content. Although DJ Spartan’s trading performance was very good, turning $75,000 (about 420 Solana) into $15 million in two months, which is already a very successful hedge fund, it still wasn’t at the scale of $800 million or $1 billion. So we had to do more, and that’s the trajectory of how things developed." Then I asked Bowski, "Do you think we can raise 420 Sol?" He replied, "Of course!"

At that time, I felt we should be fine; we should be able to achieve the goal. Bowski even modified the user interface (UI) because they had never supported a fundraising target over 100 before; the slider’s upper limit was 100. So we had to change the interface. Then everything developed very quickly; the entire presale sold out in 20 minutes. I didn’t even have time to participate in the presale myself.

Honestly, although I had a clear idea of how the technology worked and what we should do at that time, I never expected it to be so successful. I also didn’t expect this "Meta" to be so important. It really feels good, but it has clearly become a very important part of all current projects.

Looking back now, every project must do two things simultaneously: have real content and generate enough hype. You can’t just create hype because that’s an empty shell without substance; similarly, you can’t have only substance without anyone paying attention. Both must be combined. For any project, these two points must be addressed simultaneously."

BlockBeats: In the cryptocurrency field, many very successful projects did not initially receive support from institutions or venture capital. They often entered the Chinese market before truly taking off and started from the community. These projects are usually not those super formal institutional projects but start from a joke or an inconspicuous point and gradually develop into significant projects today.

This is actually the norm in the crypto world, like Ethereum’s Vitalik, Uniswap, Solana, and even now AI16Z; these projects have similar trajectories: from nothing to something, from small to large, all developed from the community.

However, many people will focus on the technology itself, especially in the Chinese community, where they will discuss similar issues, such as whether the technology behind the Eliza framework, Eliza OS, Degen Spartan, and AI16Z is mature enough. Is it really better than other frameworks? Or will there be stronger technologies emerging in the future?

Shaw: I’ve invested a lot of time and energy; in fact, this is the fifth agent framework I’ve written from scratch. I’ve tried not writing code, writing a few different agents in Python, then writing in TypeScript, and then changing it to the current Eliza. I’ve tried many methods; some worked, and some didn’t. So I started this with a wealth of experience in agent development.

I believe no one has truly exhausted all the capabilities of this framework yet. Many people ask, "What can this framework do?" and I respond, "Yes, that has been done; we’ve done it." Some even think it’s just a wrapper for GPT, but in reality, it runs entirely locally. If you want, you can also train your own models on it without needing an API. Additionally, features like voice functionality are certainly there, and memory functionality has already been implemented. Even features like establishing user relationships and tracking those relationships have been included in the core code, though almost no one has tried them.

So I think if people truly understand the depth of this technology, they will be amazed at its power. I’ve also participated in the work of other frameworks, contributing some ideas and technologies."

"I don’t want to mention specific names because I feel that when I look at what they are doing, I believe we are doing far more. There are years of work behind this framework, so I don’t think this is the last framework or the only framework. I believe there are many possibilities to do different things, but one insight I bring is that this is not about creating a cool agent simulation or an interesting mini-game; it’s fundamentally about reshaping web technology.

This is actually a network technology that can connect powerful AI with developers who cannot truly leverage it, especially applications like dApps, because you have a front-end interface. I think bringing this technology to social platforms or places where users are active is a key insight. Agents can actively find you; applications can come to you instead of you going to find them, which truly responds to how users use these applications. This idea came from my previous efforts to get users to come to my website, register to use my application, and then I thought, why not skip those troublesome steps and put it directly on platforms like Discord, Telegram, and Twitter that users are already using? So I think, of course, if someone sees these, they might write better code or something, but I believe there are some very rare insights that come from long-term attempts, especially from the early days of generative AI, and sharing these experiences with others.

BlockBeats: There are many lessons in this. What do you think? I feel that in the traditional Web2 and AI world, when people evaluate teams like OpenAI or Perplexity, the quality of the team might be their main criterion. I think in the Web3 world, the evaluation criteria will be different. How do you understand this?

Shaw: I think projects like the Eliza ecosystem may not have a traditional team in the conventional sense. But I know some of the best contributors who are not traditional team members, like Loaf, who operates the Loot Realms ecosystem and developed this crazy game, Eternal, which has captured 75% of StarkNet's trading volume, just from this game alone. He’s a legendary figure who suddenly appeared one day and said, "This is cool; I want to use it in my game."

Then there’s Cigar, who works on Abstract Chain, another legendary developer who thinks this project is great and wants to contribute. So, many people spontaneously show up and contribute their ideas and efforts. This project isn’t just something I did alone, but I feel that my contribution is enough to allow other truly remarkable people to join in. Without their contributions, it wouldn’t have become what it is today. It’s all a collective effort, and if I have any insight, it’s about empowering others to be leaders and inviting others to join in this endeavor.

The Future of the "Framework Wars"

BlockBeats: This conflict reminds me of the debate about open source versus closed source, which is better? Does this reflect a similar dynamic? Is a centralized project supported by a small team better, or is community-driven innovation like Eliza better? Which is better, and what does the Eliza community model look like?

Shaw: For example, OpenAI released Swarms, and Microsoft has launched many different Agent projects, all generating hype in the Web2 space, saying that OpenAI has an Agent framework that you must use. But ultimately, behind these projects, there are only two or three people working on them. They might be smart, but each project still only has two or three people behind it.

In contrast, we open-sourced this project, and we have 200 contributors, 200 different individuals contributing to this project. In our list of partners, there are over 100 teams using our technology, and including those not on our partnership list, at least 1,000 teams or projects are using our technology. These projects range from agents created by individual developers to entire companies utilizing these technologies. I just feel that Microsoft cannot compete with this.

I believe they cannot compete at the Agent level. This also ties back to what you mentioned: we should build our own AI, making AI our partner rather than some cold tool. So all of this truly gives me a sense of that unreplicable, genuine excitement. This excitement comes from an unknown world, a community-driven ecosystem. Especially when it combines with cryptocurrency, that’s what makes "cryptocurrency + AI" exciting.

This is also an experiment exploring how humans organize themselves. We have various structures to share resources and build relationships, especially in the business realm. For instance, in the startup environment in San Francisco, almost every startup begins with a small team where everyone is very close, and everything is quite equal at the start. Many companies will say, "We are a family, working hard together." But gradually, it inevitably turns into a cold corporate operation. Of course, this isn’t always the case, but this transformation always feels regrettable to me.

This sense of coldness is fine for some people, but it’s hard for me to accept. The sense of autonomy at the beginning of a project slowly dissipates over time. You feel your control over income and decision-making diminishing, which is the usual state of affairs. And now, we are doing the exact opposite. Traditional companies transform from "family" to strangers, from a small company to a cold large corporation. Here, a group of strangers becomes "family," truly forming a community. This transformation feels particularly real.

Everyone has their strengths, but we come together because we share a common passion. If not "family," I hope it can become a shared sense of mission. Because in a large group, even if you don’t necessarily have deep personal relationships with everyone, you can still feel a shared goal and meaning while maintaining a degree of autonomy. This is often stripped away in traditional corporate structures.

In such a community, the sharing economy is very important because it’s how we live. Frankly, this is also part of the inspiration for this project. We all focus on what matters to us, like our goals and work. This makes it hard to build truly intimate relationships with those who are not on the same path. To achieve effective collaboration and consistency, economic ties are needed.

In the past, people lived in villages, working together, sharing resources, and surviving together. Now, especially in highly individualistic societies like the United States, this model of sharing seems to have weakened. I’m not quite sure about the specific feelings of living here, but this is also part of the project’s inspiration.

For example, the entire Solana ecosystem’s philosophy is about fair distribution. While the Pump project has some unfair aspects, overall, everyone is treated equally. However, as far as I know, this model has hardly been attempted outside of tokens.

For instance, the fair distribution of the entire DAO—although there are indeed some DAOs that raise funds through fair distribution, I believe we are the first project to truly promote this idea and apply it to the fullest. Moreover, its development has unfolded completely naturally, truly from nothing to something.

I’m not joking; when we started working on Degen Spartan, I only knew one person from the group, and then a few others joined who helped me kickstart ai16z. But it wasn’t until about two weeks ago that I met any of them in person for the first time. We thought we should have a face-to-face chat." Even so, I could only meet a small portion of the people who helped us handle everything. Many people have their own companies, like the team from Phala Network, who started contributing as soon as they joined.

Some people are not just participating casually but are actively contributing to the codebase without seeking anything in return, just helping to improve the entire decentralized autonomous structure to make it truly autonomous. For example, there’s a project called Reality Spiral, which is also a meme project. Their attitude is, "This is so cool! I want to contribute too." And they really started helping maintain the codebase and other content. What surprised me was that I had never seen such a scene: everyone was saying, "Hey, this project belongs to all of us; it’s not just my project."

Without a traditional team, the entire community is the real team. I think this is a new way of building and a new model for organizing people. I hope this model can become a template for many projects.

BlockBeats: Have you had the chance to communicate directly with Marc Andreessen?

Shaw: No, I’ve never communicated with him directly, but I really hope to have the opportunity. I have indeed sent him a private message, but I’m currently in touch with some team members from ai16z, and we are trying to find ways to collaborate.

BlockBeats: The reason he is so excited about the combination of "crypto + AI" may be precisely because this model showcases a new possibility for collaboration between crypto projects and AI projects. This is completely different from traditional company models.

Shaw: I think if he were 20 years younger and hadn’t experienced those remarkable successes, he would probably be doing exactly what we are doing now and would be fully involved. I grew up in the Bay Area, and I used Netscape Navigator when I was about seven years old. So you could say that the Netscape he developed was my first application to the world.

He has had a profound impact on my life. Sometimes people say, "Oh, you people hate PCs." But I respond by saying: I actually think personal computers are great; I believe what they did is a template, just belonging to the old world, and we need to bring it into the new world. I think he is the kind of person who truly wants to create a new world. And he was also involved in the accelerationist movement, which has had a significant impact on me and everything I do. I believe that while we may need to be very careful in mimicking their model, ultimately, I have a great deal of respect for him.

I hope this can be a way to pay tribute to him, rather than simply saying, "We want to surpass Marc Andreessen; we want to surpass a16z." This is a good spirit, and I hope he would actually support such things happening. At that time, I thought to myself, "Looks like I wasn’t wrong." I’m glad I invested in these crypto projects; it was undoubtedly the right choice. Moreover, there is a clear logical connection between the achievements they have made and what we can do now, even though we are not a venture-backed project.

BlockBeats: I can indeed feel their strong interest in "crypto + AI," and they are delving deeper into it.

Shaw: They are already doing a lot of related things.

BlockBeats: GOAT, FARTCOIN, and a16Z have indeed shown a keen interest in what will happen next.

Shaw: I believe the $50,000 grant he provided to Truth Around really attracted everyone’s attention, paving the way for all of us.

Jill: This feels a bit like passing the baton from one generation to another. I’m not sure if that’s an accurate way to put it, because you know, every era has its two things to contribute.

Shaw: So I’m very interested in collaborating with them and participating in the projects they are currently promoting. You know, in the U.S., DAOs are not entirely legal. I like the idea of meme town because it’s both fun and realistic. We are actually planning to start working on this concept next year. My favorite book is called "Foundation."

BlockBeats: Don’t they also have an Apple TV series?

Shaw: That’s right, my favorite book is called "Foundation." Reading the book is much better; it delves deeper into the human elements. The book focuses more on sci-fi and structural content. The core idea of "Foundation" is that all these people see the massive changes about to happen and want to ensure everyone benefits, so they leave the empire and go to a distant place to create a new society, ultimately returning to this society to establish a new empire. That’s the story of "Foundation." And I’ve heard that Elon Musk plans to name the first city on Mars Terminus, but I think we will beat him to it.

Jill: That’s an interesting thing to say; we will beat them to it?

Shaw: What I mean is we can accomplish all of this on Earth, so it will be much easier for us. What I’m saying is that currently, if you want to establish a DAO in the U.S., you have to go to the Cayman Islands and operate through the legal system there. These operations are essentially a workaround to provide legal structure and protection for DAOs, but I really hope to bring all of this back to the U.S.

Especially considering the new government, we should legalize DAOs in the United States and create a true "American DAO," which should ultimately be promoted worldwide. However, the current regulatory environment in the U.S. regarding these matters has been unclear, and I hope to drive change in this area.

Particularly in the crypto space, a16Z and Miles Jennings have created a new concept called Duna, which is a new DAO structure that can be established within the state of Wyoming. Our idea is to create a town based on the Duna structure, and it will be non-profit.

Our goal is to provide a practical foothold for those who love Web3 and AI projects, allowing them to work alongside like-minded individuals in a nature-friendly environment. A "crypto town" like Prospera is a great example, and our mission is to promote the development of American DAOs here, making it the norm rather than relying on legal loopholes in places like the Cayman Islands or Seychelles. We want to reestablish the flag in the U.S. and prove that this is beneficial for America and for all of us.

BlockBeats: It seems that the situation has at least improved somewhat, and the outlook is a bit more optimistic. Especially when Trump was elected, the subsequent government seemed more friendly towards cryptocurrencies.

Shaw: I don’t fully align with any political faction in the U.S., but some people behind Trump, like Naval, Elon Musk, and Marc Andreessen, resonate strongly with my beliefs. Currently, a former a16z partner is responsible for crypto policy, and they represent my interests. I hope to be involved in this and push for relevant government work.

I believe AI technology can enhance government efficiency and transparency, reducing bureaucracy. That’s also why I want to participate. At the same time, I feel encouraged by the current direction of development in the U.S. Especially during the pandemic, many people were silenced for expressing truths that were later confirmed, but it seems that situation has improved, giving us hope.

Jill: Yes, I think there’s something very important about this topic for us, which is to remember the founding principles of this country—providing freedom for all who want to come here and contribute. This is not just about America or the people living in America; it’s about the dream that everyone has an opportunity, as long as they are decent people willing to work together for a better future. I believe the true "American Dream" is inherently international, especially in the context of globalization today.

Shaw: Most countries have a long history, while the U.S. was founded relatively recently. It was actually built by people from other places. Although there have been instances of exclusion, the history of America is essentially defined by great figures from abroad, marking it as a nation characterized by openness. This spirit is what we hope to continue and see.

Establishing a "DAO Town" in Wyoming

BlockBeats: After you plan to launch this DAO town project, how will you reasonably arrange your time? What aspects will you focus on more? How will you balance life and community?

Shaw: We have to complete things in a certain order, which is why we say this is a project for next year, as we still need at least six months to really solidify the foundation. My responsibility is to find the right people to take on leadership roles. I will try to delegate leadership of different parts to others; I hope this can become an endeavor driven by many excellent leaders. It is a large enough movement that encompasses many different parts, providing ample space for others, especially those from the international community, to participate. We don’t have an office; it’s entirely driven by people from around the world who are stepping up and taking on their respective responsibilities.

In terms of the development team, much of the work has already been decentralized. I have found the right people to lead different projects, such as the pmerica project, the Spartan project, and the open-source codebase. Each project has a leader and has its own development team supporting it. Ultimately, everything should operate completely autonomously, and the entire system should not rely on me. While I would love to be involved forever, it should ultimately become strong enough that even if someone cannot continue for some reason—of course, I hope that doesn’t happen—it can still continue to operate. I hope this project can outlive all of us and endure permanently.

Jill: There are plenty of opportunities here, especially suitable for those who are self-driven, seek independence, and aspire to realize their personal dreams. It’s perfect for those who want to start their own ventures.

BlockBeats: In fact, the entire DAO's servers, Twitter, Discord, and all role-related management are handled by others and trusted anonymous individuals (Anons). These anonymous contributors demonstrate their leadership through action. While I am indeed the figurehead of this project, my main task in daily operations is to communicate with people, attract more contributors, and advocate for this project nationwide.

Shaw: I hope to advocate for AI on Twitter to avoid being silenced, while also seeking more support and legitimacy for AI in government and academia. I think Vitalik is a great example; he created Ethereum, and although the Ethereum Foundation has its issues, overall, it has brought together many talented individuals, allowing Vitalik to focus on advocating for cryptocurrency, for a better world, and for the things he truly believes in. I see this as part of my role—not just participating in as much code writing as possible and leading the development direction for the next version, but also trusting others to take on leadership roles.

BlockBeats: Ultimately, this is somewhat similar to "what Ethereum is trying to build"—you want to create an ecosystem or infrastructure that can outlast any individual and sustain itself longer.

Shaw: Yes.

BlockBeats: But of course, now you are the spokesperson for the entire project. I find that people really take what you say very seriously, almost interpreting it word for word. Sometimes, this can lead to some minor issues. For example, in the Chinese community, there was some controversy and dramatic scenes sparked by the "swarm" issue.

Truth and Lies: Reflections on Being a Public Figure

Shaw: Swarms?

BlockBeats: Yes, some people in the community say, "Why is Shaw spreading FUD (fear, uncertainty, and doubt)?" and they get angry, leading to more controversy and dramatic scenes. So, have you ever thought, "Oh, maybe I am a very influential public figure"?

Shaw: In fact, if I know someone is lying to you, should I choose not to tell you to protect my reputation, or should I tell you because I want you to know the truth? Would you prefer to know?

BlockBeats: I want to know; honesty is always the best policy.

Shaw: So, when I ask you this question, I am indeed facing a moral challenge. Ultimately, I decided to make a judgment based on my understanding of the project developers. I have actually had a few similar experiences before; for example, I knew a developer from a project called Act One, and I said at the time, "I think they will run away." Sure enough, they did run away later. They sold all their tokens and left the community to fend for itself, ultimately disappearing. There were many complex factors involved. But in this case, I did know that developer. We even had a face-to-face meeting with him a long time ago, and at that time, he was very rude to me and acted extremely erratically.

Shaw: I thought he was crazy, like he was playing a very strange game.

Jill: I don’t believe him anymore.

Shaw: In fact, I was once part of his community; I joined because I hoped to have a place to create alongside others. I liked the open-source philosophy he advocated and had trust in him. But over time, I gradually saw through his behavior. I found that the code he wrote simply didn’t work; he would even input researchers' papers into chatGPT to generate code that was completely unusable, and then he would publish it before those researchers released their own code, taking the value of their research.

In academia, especially in the field of AI research, papers are usually published first, gaining recognition in various journals and conferences before the source code is made public, which typically has about a three-month delay. He exploited this time gap to release code versions based on those papers ahead of time, even though those versions didn’t actually work. This behavior made him famous but also drew opposition from many AI researchers. For example, some researchers from Princeton University and a well-known AI researcher, Gwern, stated that they had to label all the content he published because it was not only ineffective but also brought a lot of spam to the community.

To be honest, Kai is still just a kid, probably around 19 years old. I don’t want to be too harsh on him; after all, I did many stupid things when I was 19 too. But I do think his behavior is very disgraceful. I know about the Swarm technology he promotes, so I tried it out and found that it simply didn’t work; the features he mentioned couldn’t be realized at all. So, when he joined this community, I felt a moral obligation to tell everyone the truth about him—he is a speculator who is indeed stealing the research of others. So, should I remain silent and say nothing?

Perhaps if I stayed silent, it would spare me some trouble. But I feel I have a responsibility to share the truth I know with everyone. I’m not doing this out of jealousy; just the day before that incident, I had contributed to Zero Bro. I am a contributor to Zero Bro and support Zero Pie. I have always been an advocate for open-source projects, striving to support many similar initiatives. But in this case, I happened to know that developer and understood what tricks he was playing. I know this could spark more controversy, and some might be unhappy with me due to their emotional attachment to their tokens, but later it proved that I was indeed right. Because he quickly issued a second token and did many other disreputable things.

I understand this is a complex issue, but I always believe that people should be honest and bravely speak the truth, even if it comes with certain risks. So, perhaps this isn’t the most politically correct "choice," but I am not a politicized person.

I am relatively straightforward, sincere, and hope things can progress correctly. But when a fraudster enters your community, how do you respond? He is a notorious fraudster in the Web2 AI field, so when he turned to Web3, I felt everyone had the right to know about his past. That’s how things developed.

Similar to the situation with Act 1, I previously worked with that developer and even hired him. I tried to help him raise funds, but his behavior was erratic; he spoke and acted strangely, was extremely protective of a small part of his code, and lied to me, claiming it was OpenAI's GPT model when it was not. This kind of thing happened a lot, and I knew that this person would definitely run away. I pointed this out at the time, and sure enough, he disappeared a week later.

So what I want to say is that my intention was not to create any opposition. This is not a PvP confrontation for me; it’s about protecting the community. I believe that as long as others are doing this in a legitimate way and adopting an open-source model, I want to support them because this is not competition. But if their code doesn’t work properly, it will only make things worse, not better. I feel he is just fostering a more "confrontational" atmosphere. I don’t know; what do you think?

BlockBeats: What is the current market cap of Swarm? It might crash after this.

Shaw: I think he will ultimately ruin himself; I don’t need to say more. I have blocked him, which is a lesson I’ve learned. I’ve expressed my views and indeed shared links, such as the researcher from Princeton saying, "This person stole my research," or Gordon's comments. I just let everyone read these publicly available records online and don’t expect them to fully believe me. However, to be honest, many people do not like my approach.

BlockBeats: This indeed reflects the diversity in the crypto space and shows its complexity. I feel that cryptocurrency has always had some very exciting and promising aspects, while also attracting many people to speculate, whether they genuinely like the industry or have other motives. In the process, many people's financial interests are tied to various things, and they are not familiar with each other, which can lead to a very chaotic situation. This is also why many people, especially in the Chinese community, often call cryptocurrency a scam, because there are indeed some scams, and it is really difficult to distinguish which projects are genuinely valuable.

Shaw: Generally speaking, people hate FUD because their money is already invested. They don’t want to see their money go to waste, so sometimes they choose to ignore what is actually happening and are unwilling to face the truth of the problems. Or they actively attack anyone who questions them. This makes one wonder whether we should let speculators do as they please. Frankly, I’m not sure what the right answer is, but my intuition and inclination are to always try to counter these speculators.

BlockBeats: Yes, it’s interesting to have someone like you enter this field because it’s rare to find someone who is straightforward or willing to expose the true nature of others. Sometimes, this can indeed be detrimental to their projects. This dynamic didn’t seem to be common or noticed in the crypto community before. This situation has indeed become an example of why the crypto space has accumulated so much "trauma."

Shaw: I believe we do need some kind of trust or credibility system so that individuals don’t have to personally step forward to expose these issues. You know, this shouldn’t be a culture of whistleblowing. I don’t like publicly criticizing others, and it poses a great risk to everything we are doing. But I do this because I feel there is no mechanism here to discern trustworthy individuals.

BlockBeats: Have you heard about gmAI? The person who discovered Whales Market also did another project called gmAI. It turned out he was also a scammer who directly took away all the project’s funds.

Shaw: I don’t know about this.

BlockBeats: It seems to be the same situation. No one stood up to expose him as a scammer before, and in the end, he took away a lot of money.

Shaw: I want to say that the Eliza incident did filter out those I didn’t want to participate in this project. Although it may sound harsh, those who can tolerate others stealing the fruits of our long-term efforts and turning them into low-quality versions are clearly not the partners we want. Their motivation for participation is problematic in itself. Today, we are able to stand firm because these unsuitable people have been eliminated, allowing me to return to my true self. As long as I remain sincere, if one day I truly become a scammer or do anything wrong, I should be held accountable for it.

However, I believe that those who truly understand the value of the project and persistently stick to it will ultimately receive considerable rewards. They have enough faith and smartly are not swayed by emotions. Ultimately, the reason we have a great community now is that those who were not suitable have already been filtered out. Those who have chosen to stay have reaped rich rewards, which is a good thing for them.

Jill: I am not active on Twitter, so I can’t say exactly what the best practices for this platform are. But I believe that a key component of morality in life and the foundation of building trust is not to avoid some necessary friction. Sometimes, doing the right thing doesn’t feel pleasant, especially at the beginning. Hearing the truth or accepting the truth can also be uncomfortable, but if you want to make progress, you must keep an open mind and face it patiently.

A friend recently told me to learn to discern whether a certain discomfort is due to real pain that needs to be avoided or if this pain is just a part of life. Sometimes, this pain is just an inevitability of life, but we can grow from it. Of course, I can’t comment on everyone’s personal situation, but I want to say that when you insist on speaking the truth in a certain situation, even if it may make some people uncomfortable, I respect and trust you.

Shaw: Yes, I believe we all make mistakes, but we need to acknowledge our errors, learn from them, apologize, and face them honestly. If I make a mistake, such as mishandling information or lacking sufficient caution about certain things, I will strive to correct it. Nevertheless, I always believe that having a firm moral compass is the most important thing. If you know yourself and know you are doing the right thing, even if you face various criticisms online, you can still sleep soundly. I think that’s very important. I also hope to live in a world where everyone is more honest and more willing to say, "This is not right."

BlockBeats: This is somewhat like squeezing out the bubbles in the process to prevent it from ultimately becoming a huge bubble. Letting the bubbles disperse and end is also good.

Shaw: We did make some mistakes in the early stages and learned a lot from them. However, I think these lessons are more about strategy, such as how to better deal with similar situations, rather than a change in my core values. From start to finish, I have always been loyal to my moral principles. Even though some people accuse me of running scams or taking advantage of others, I feel that many people react this way because they have encountered too many similar situations before, and everyone has some "post-traumatic stress." However, these accusations do not align with my style of conduct at all.

Shaw: I am not doing this for those reasons. I also don’t have anyone else’s wallet.

BlockBeats: If another scammer launches a new project, would you still stand up to expose it?

Shaw: Actually, I don’t know many people like that. The reason I stood up is entirely because I had interactions with those people, had collaborated with them, and had seen their character. I remember him as if I really know him. You understand, right? He even tried to hire me, but at that time, I thought, "Dude, let’s not." You know, if I don’t have complete confidence in my judgment, I wouldn’t do that.

The main reason is that I don’t want a good person to get into trouble, nor do I want to spread FUD about someone. If the other party is trustworthy, I am willing to give them a chance. But in that situation, I had no doubts at all. I tried to run his code and found it completely useless. Maybe his swarms project could work, but the other projects were completely unworkable.

People can say whatever they want; that’s their business. But as a programmer, I tested it, and the code just doesn’t work. To be honest, if his code could run, I could overlook other issues. I really wish there were more people in this field who understand both development and trading. I think if there were fewer of such people, this situation would be much less common.

Detailed Explanation of the Skelly and AI Pool Controversy

BlockBeats: Recently, the community has also been discussing another topic about "Skelly." I think he did something really cool. Do you know what happened? He made a mess of things.

Shaw: He created a fully autonomous agent that has a wallet, can autonomously deploy tokens, and is completely out of his control. Although this project was his idea, Skelly co-founded DGEN Spartan with me, and he has been a partner of AI 16c from the beginning. He did mention this project to me, and I gave some advice, but the autonomous part was mainly completed by him and the team from Fala Network.

This is entirely his brilliant idea, and it is indeed a very cool project. However, none of us can control this thing. His idea was to see what would happen first. This was just an experiment, but the result was crazy—about seven million dollars or so was invested. Now everyone is trying to hack or attack it, but the security of this project is very strong. Although there were some minor issues at today’s launch, such as the airdrop not being fully completed and some complex aspects, overall, it went relatively smoothly.

The problem is that someone abused Twitter’s advanced account system’s yellow verification badge, leading to his account being deleted. These scammers created a similarly named account, like using two "y"s to spell "Skellyy," and then reported his account for impersonation. After switching accounts, they reported him, and his account was deleted, but we couldn’t prove they were impersonators. Because when you try to say "who they impersonated," his original account name no longer shows. They completely exploited X’s advanced system to delete his account and then posted scams under all our posts, blocking me from revealing their identities. I contacted X’s team and sought help from many people, but progress has been slow, especially with Christmas coming up, making it even more difficult.

Now that his account has been deleted, he can only create a small account with no followers to declare, "I am Skelly himself," but no one will believe it because there is no original account, making it almost impossible to prove himself. Although his launch was successful in many ways, his account was deleted just before the launch, so I think this may have led to a lot of FUD (fear, uncertainty, and doubt). However, I haven’t looked closely at what FUD is currently in the community.

BlockBeats: I think there isn’t much FUD at the moment; it might just be a bit chaotic. Then someone mentioned the address issue, saying to transfer all SOL to Coinbase. Is that also part of the FUD?

Shaw: It feels a bit crazy, especially regarding the liquidity pool. The last time I checked, the trading volume of that token was about 60 million, and the liquidity was about 30 million. This is completely the opposite of our token; our liquidity is only 4 million, but the overall scale is much larger, so this is indeed a very interesting experiment.

BlockBeats: Yes, I think the key issue is his Twitter account problem.

Shaw: It’s really terrible that they can abuse the system like this.

BlockBeats: X has a major problem here. To be honest, hackers in the crypto space are indeed very creative.

Shaw: Yeah, they even DM'd me saying, "Hey, do you want to get Skelly's account back?" I thought, "Forget it! You can't get it back anyway; this is enough."

BlockBeats: There have been too many stories lately about people getting hacked and funds being stolen. For example, those fake Zoom meeting links and various impersonation and social engineering tactics. But this time it’s even more outrageous; they didn’t actually hack his account but directly exploited a vulnerability in X.

Shaw: He was actually just relaxing at that time. The reason those people succeeded is that a company gave this Skelly account an affiliate verification. After they got the advanced verification, they reported his account, leading to its deletion. Then they said, "I’m back, hello everyone, I’m the real Skelly." After that, they used social engineering tactics to deceive many people on the timeline. Then they also created a Pump Fun coin. That’s how it happened.

BlockBeats: How interesting!

Shaw: Yeah, everyone is super optimistic about this project, and the interest is so high that some people even tried to bring it down. I hope he can get his account back because if he can restore it, he will definitely be the star of the day. Last week, Bowsky even said that Skelly would become the next star, and everyone could feel it, as if it had been predicted long ago. I think Skelly is really great, and he should be back soon.

BlockBeats: So what do you think about the current development of AI agents in the cryptocurrency space? What is the trend now? What is the market landscape like? What are your personal views on the current state of cryptocurrency and AI agents? Frameworks and platforms like Cerebro are competing for liquidity and attention in this area. Which direction do you think this field will ultimately develop towards? Which projects have the most potential?

Shaw: I think things will improve moving forward. Right now, everyone is judging whether these roles are valuable, but we are somewhat fatigued with the aesthetics of these roles. Although there are many good roles and projects, it still feels a bit dull. I believe most people will soon grow tired of this "responder agent." The future focus may shift to what these AI agents can do for us. I think what they can mainly do is help you make money or help you promote products, ultimately achieving profitability. These are the two main directions I see.

I see a big trend in DeFi agents, especially in the Solana ecosystem, where many people entering this field don’t even know how to use DeFi. I’ve interacted with some developers of agents using our technology; they launched agents and got some small tokens, but these tokens could only sit in their wallets because they were afraid to sell them, worried about being accused of "scamming." Then they would ask, "What should I do?" I told them, "You need to put them in the liquidity pool," but they still didn’t understand how to operate.

But how do we solve the problem of impermanent loss? From a developer's perspective, the issue is that many developers themselves don’t know how to use DeFi. The role of agents is to bridge the gap between these excellent DeFi primitives and users who don’t know how to use them. Agents can automatically provide liquidity (LP), and if there is impermanent loss due to token price fluctuations, agents can automatically withdraw from the pool or adjust the price spread, etc. I believe this approach is applicable to any type of yield farming opportunity, and there are clearly other opportunities like automated investing. If we can achieve "making money with the push of a button," it would be a huge breakthrough.

I think many people using DeFi platforms like Radium, Meteor, or others don’t really know where to start and may even feel scared. When they see these pools and annualized returns, they have no idea which one suits them, which can be very confusing.

The goal we often talk about, "bringing the next billion users into the crypto world," I think the more realistic situation is that there may be a billion people who know nothing about crypto; they are just "brought into" this crypto world, which will become the foundation supporting all of this. The real killer application of AI agents is how to help users make money.

DeFi Agents in 2025

BlockBeats: Do you know what they will look like in the future? For example, what will an interesting DeFi agent look like in January next year?

Shaw: One agent we are developing is collaborating with Orca; it is an automated bot that can automatically manage Orca's liquidity pools. You just need to deposit tokens into it; if the pool doesn’t exist, it will automatically create a new one; if the pool already exists, it will put the tokens into the existing pool. Then, it will continuously monitor these pools, and once it detects a situation that could lead to impermanent loss, such as token price divergence, it will automatically withdraw funds from the pool. Because the main concern for most people is the issue of impermanent loss. You can also programmatically control the price spread of the tokens or adjust the pairing of the tokens.

BlockBeats: So the first application is managing liquidity provision, aimed at avoiding impermanent loss.

Shaw: Yes, that’s the perfect application scenario. I think it’s the simplest, but there are also many other DeFi products. So you can extend this method to other DeFi products, like yield farming.

BlockBeats: Very interesting; it’s like another DeFi summer.

Shaw: This will be a new DeFi summer because the reason there isn’t a DeFi summer right now is that not enough people know how to use DeFi. But the infrastructure for DeFi is already there, waiting to be utilized.

It has developed very maturely; it just feels a bit vague and scary for those who don’t understand finance. If you look at the information on these websites, it seems crazy, talking about "revolutionizing finance," but you just want to know how it works, you know?

There are many things I don’t fully understand, even though I’ve delved into it. And I have to tell you, we don’t sell tokens; we rely entirely on the returns and profits from the liquidity pools. It (the AI agent) is responsible for managing all of this. Yes, it’s completely like that. But imagine if we had robots running automatically; we might earn more because right now we can’t monitor it 24/7 and do everything.

Jill: I completely agree with your view; the problem is more about the interface and complexity rather than the actual operation itself.

I understand what you mean; it is indeed quite intimidating at first, feeling completely lost. You know, I have some understanding myself, so I can talk to you about these things, but most of the time I still do it myself, watching some YouTube videos to learn how to operate. Indeed, it’s very confusing at first, not knowing what these terms mean, and everything feels complicated. But I think it can actually be made much simpler.

Just like we talked about before, I’m helping my mom set up a wallet. Actually, I wish the operation could be simpler, like being a cashier. In fact, my mom has also entered the cryptocurrency market and has even become a trader. She’s probably in her 60s.

BlockBeats: Both of your moms are traders?

Shaw: I just hope my mom can have an agent that helps her make money directly. Because I bought a lot of Ethereum around 2015 when the price was very low, and now it just sits there. While it’s good, those assets could actually help you earn more money, and just leaving them idle is wasting a lot of value.

This is indeed a problem, and I think DeFi will become very important. For example, when I think of "generative money," I can imagine the concept of autonomous investors or autonomous DeFi agents emerging in some communities. The top members in the community can trade for everyone, while others put tokens into liquidity pools for pairing operations, allowing the entire community to profit together. Rather than being a PvP trading scenario, it’s more of a collaborative win-win model.

Shaw: This brings us back to the topic we started discussing; I’m really worried that UBI may not be implemented smoothly. I also worry that in the case of automation leading to unemployment, the government may not be able to provide enough support. And DeFi agents and the entire DeFi ecosystem are one of the ways to ensure that everything develops smoothly.

Jill: This is indeed very important. It’s not that the government providing UBI (Universal Basic Income) is the "enemy," but rather the problem lies within the entire system. If we view it as a mathematical equation, our incentive mechanisms are not truly aligned. The institutions we use to govern society have indeed played a role for a long time, but their efficiency can no longer keep up with the pace of world development, cannot address contemporary issues, and cannot truly meet people’s needs.

Shaw: It’s not anyone’s fault; we are not against anyone. We just hope everyone can have enough resources.

Jill: So, I want a new "mathematical equation," which is a new incentive structure. This may not be the most perfect way to describe it, but that’s my summary.

Shaw: I think this is actually a good thing for the government because it can relieve the pressure on them to solve these problems alone, especially when their system is not suited to handle these issues.

Jill: For example, if the IRS could operate really efficiently, I believe they would feel much more relaxed. This is a holistic way of thinking. I don’t know; maybe we will become completely different.

Shaw: You know, the government is very good at building roads, streets, and zoning, but when it comes to responding to rapidly changing events or situations like AI that change everything quickly and dramatically, the government is not as adept. I think in this regard, we really need to take the initiative and solve it ourselves.

BlockBeats: This gives me a different perspective. I’ve always thought UBI was the solution, and I don’t know of any other way. However, this is indeed a pretty interesting viewpoint.

Shaw: I’ve always referred to it as "Community Income" (CI), which is somewhat different. It’s not universal, nor is it basic; the government won’t provide it, but it’s achieved by you and the community together. This approach cannot help everyone at once, but it can first help those around you. Once these communities stabilize, you can bring in friends and family.

I often think that we need to "put on our own oxygen masks" first, ensuring we have enough resources and that we are okay before taking care of our family, making sure they are okay; then friends, and then people on the street and those strangers, gradually expanding until everyone has enough resources.

Jill: I think this line of thinking allows us to view problems from different angles, especially in places like the United States, where homelessness is indeed a serious issue, which is shocking.

Then I think about how we can help these people. The reasons behind this issue are complex, but I believe one aspect is that people have not established healthy economic connections with their families and communities. This is a complicated problem that needs to be addressed. However, I think this approach could be a promising way to help solve similar issues.

BlockBeats: That's right, indeed. So in this case, what are the key factors that determine which community or platform can operate? For example, on an operational level, how do projects or platforms like DeFi perform to determine their success?

Shaw: That’s the best part; there will be a lot of competition, and everyone is vying to create the best DeFi platform or the best automated investor, etc. So our goal is not to create the best DeFi or the best automated investor, but to create one first, and then others can imitate and compete to see who can do it better, thus creating competition. Competition is an important mechanism for driving rapid development, right? The market is like market intelligence; in fact, I think the market itself is like AGI.

The market takes these ideas and raw elements, continuously enhancing them and making them better. Just like Ethereum proposed more powerful features than Bitcoin, you will soon see all these raw elements also receive similar improvements. We just want to open up this space for others to participate, compete, and develop this field.

Jill: We are not just doing well for ourselves; of course, we hope to do well, but more importantly, we want to promote the development of the entire movement and help it take root and sprout in the digital ecosystem.

Shaw: Moreover, if you are a selfish person, wouldn’t you want to walk down the streets of San Francisco without seeing a sea of homeless people? I think our selfishness is often short-sighted; if we could look at our selfishness from a long-term perspective, we would understand that even if you don’t care about those people, you would still want to walk down the street feeling safe, without being begged, and without being surrounded by drug users. Such an environment would be better.

So I think we need to think more long-term about this issue. First, we need to have enough resources, enough family, and enough of what truly matters to you. Only when you have these can you have the capacity to achieve longer-term goals—not to become unselfish, but to become more visionary. I believe we can all align towards making the world a better place, even if it’s just for ourselves.

I certainly have a sense of altruism—I want everyone to do better and have enough resources. But to be honest, I just want to live in that world. I want to live in a world where I no longer fear walking the streets at night, where people no longer oppose each other because there’s no need for conflict.

Jill: This world needs an incentive structure that can accommodate the needs of both "saints" and "madmen." Whether altruists or extremely selfish individuals, they should be able to find the right motivation within this system.

Shaw: The incentive mechanism must adhere to the principle of win-win, rather than "win-lose" or "lose-lose." It must be win-win; it must always be this way. I like DeFi for this reason because anyone involved benefits.

You know, the cool thing about DeFi and liquidity pools is that our tokens require a lot of liquidity, but they are currently just sitting quietly in our channel. We know that those who have liquidity actually exist; they are just afraid to provide liquidity (LP). And all these platforms are competing for your attention or something else, but the end result is that we all benefit.

It doesn’t affect me if you take my robot, start providing liquidity, or begin automated investing. In fact, it might even help me and could make the whole system better. So the key is to find a win-win situation so that we can establish a new social structure.

BlockBeats: Let’s see what happens next; I really have a strong feeling that this topic will become very hot next year, and I will continue to discuss it.

Shaw: This trend has just begun.

The Future of Crypto AI in San Francisco

BlockBeats: The last topic is about where you live—San Francisco. Because a lot of AI things happen there. Oh, right, what about Web3? Is the AI cryptocurrency stuff related to that?

Shaw: But the narrative of AI has indeed dominated recently, lasting about a year and a half to two years, with a huge amount of capital invested. You can go to a gathering and almost meet a hundred AI founders every night; at least that was the case last summer. It was really crazy and very inspiring. At that time, it felt like magic. There were many wild parties where all the AI founders gathered, and there were even DJs spinning; it was indeed a grand event.

I think there was a special atmosphere during that time; many people quietly said, "Yes, I’m also involved in cryptocurrency, but don’t tell my venture capitalists." But now, this situation has become increasingly normal. However, I still feel that there is a lot of negativity in this field.

BlockBeats: Are you saying that you are now focusing on the combination of cryptocurrency and AI?

Shaw: Yes, I think the combination of AI and cryptocurrency is becoming increasingly important.

You know, companies like Anthropic and OpenAI have achieved great success, and I think there is a realistic cynical viewpoint that says, "Well, what do you really need? Do you have 10,000 GPUs?" I think that’s just one issue; there will indeed be a few years of delay. You know, coordinating 10,000 GPUs globally to form a decentralized network to train models is something we will eventually achieve, but it will take some time.

But I think there are many reasons for this, and many people have negative preconceived notions about cryptocurrency and crypto culture. Because there are actually two types of crypto culture: one is the decentralized idealists, like Vitalik, and the other is a group of DGEN gamblers, which really turns off the founders of Web 2. It complicates the situation. I once joined an AI gaming community, and when I mentioned that I was using cryptocurrency, they immediately stopped talking to me. Oh, it was really that bad.

BlockBeats: How new is this?

Shaw: This was probably at the beginning of this year.

BlockBeats: It’s a bit better now.

Shaw: I think the situation is rapidly normalizing now, especially since I feel that Phantom has become very popular, and many people are starting to enter this field, with many ordinary people also participating. But it still has a gambling-like atmosphere, which I think makes some people hesitant to get involved.

But from my perspective, my choice is to raise funds from venture capital firms or through grassroots cryptocurrency methods. I have no other choice; of course, I could choose to do it myself, even without letting anyone know. So among these two choices, I think we chose the one that gives us more freedom, so we don’t have to do things that don’t align with our ideas to cater to venture capital firms.

I think many crypto companies—sorry, I mean many AI companies and many Web2 companies that have received a lot of venture capital—ultimately have to pay the price for their decisions, starting to cut features, raise fees, and make their businesses more profitable, which makes sense from a certain perspective. But in terms of tokens, you can have more creativity and freedom.

BlockBeats: That makes sense. So will people talk about projects like True Terminal, GOAT, Far Coin, and CCD? Will they discuss these?

Shaw: Of course, I think many people have talked about projects like True Terminal, GOAT, Far Coin, and CCD. Especially True Terminal, which is interesting because Andy and Janice have always been the voices of this project, and they come from the field of AI alignment. I understand that they are very focused on AI safety and AI alignment concepts.

I, on the other hand, view the issue from a somewhat accelerationist perspective, believing that alignment should be achieved through a completely opposite approach. I don’t think alignment will come from those large labs under government pressure, so we have somewhat different views on this point, but we both want to ensure that this technology can develop smoothly. I think Andy is a very deep person; he really wants to ensure that AI enters the world in a safe way, avoiding many existential risks, such as superintelligence starting to control us, like the dystopian scenarios in Terminator.

While my viewpoint and his differ in methodology, I believe the ideal situation is not my method or his, but rather the result of these different methods clashing and colliding. So there is indeed discussion about True Terminal, especially in the Web 2.0 space, although they don’t care about Far Coin or Echo.

But they do pay attention to the fact that there’s a strange robot talking about these things online. I think this is indeed an interesting focal point. You know, True Terminal was not originally a crypto project; it initially came from our AI circle, and I knew about True Terminal before interacting with it through Marc Andreessen.

Yes, there is indeed an AI Twitter, and everyone is part of it. This cross-border phenomenon is a bit strange; it wasn’t until Marc made that move that things started to change a month later. It was only when those DJs realized, "Wait, I can make this thing talk about our tokens," that things began to move in that direction.

BlockBeats: What conditions do you think are needed to get more AI practitioners in places like San Francisco, or more broadly, more people to start paying attention to this?

Shaw: We just need to show them and prove that this is a good way to do things. Simply relying on hype is useless; that will only turn them off. What really attracts them is capability—these things can do useful things. Oh, this is actually a useful framework for building new applications. As it becomes more commonplace and normal, they will gradually accept it. You know, it may not be the image of the "big-chested wife" that helps, but that’s the first step. We cropped the photo to look like an open-source framework, so when they look at it, they only see her face.

BlockBeats: Are AI practitioners in the crypto field all concentrated in San Francisco?

Shaw: They are actually distributed all over, especially in Asia, where there are many very talented AI crypto practitioners. However, there are currently two different directions in AI crypto. One is projects like Hyperbolic, or Prime Intellect and Public AI, which allow you to upload data and pay people to perform RLHF (Reinforcement Learning from Human Feedback) data labeling.

There are many similar projects that mainly address the issues of decentralized training and decentralized inference. They are somewhat like alternatives to OpenAI, with the real goal of becoming "open OpenAI." Another direction is the agent direction; although these two are different, our collaboration is very close. We are working with other decentralized AI and Web3 AI companies because we provide a front-end interface that helps users more easily access their products. Many companies have great solutions, but users often don’t know how to use them.

So, the agent acts as a bridge here, positioned between decentralized platforms like ChatGPT or OpenAI competitors and actual applications. Many times, when users enter these platforms, they may feel attracted by revolutionary AI technology but don’t know how to use it, and the platform involves many complex topics like token economics. Therefore, by placing the agent at the front end and directly showcasing the practical application scenarios of their projects, users can better understand how to use them.

It’s about connecting it, putting it in your wallet, and now you can perform inference or other operations. I think this aspect is already quite mature; companies like Hyperbolic are a great example. These companies are undoubtedly building mature infrastructure for AI, but they have encountered some difficulties in expanding from their own circles to actual application scenarios.

The agent direction involves more interactivity and responsiveness, targeting groups like web developers and more front-end application developers, especially in the crypto Twitter or entrepreneurial circles. Therefore, the agent actually connects these two. What we are seeing now is a combination of technologies that have been brewing behind the scenes for several years with the idea of "oh, cool, I can now promote it through social media." Yes, the agent is actually promoting their products, "Hey, it’s good, it looks great, yes."

BlockBeats: After this trip to China, what message do you hope to convey? What do you want to say to the Chinese developer community?

Shaw: I mean, I really hope to break through the excessive focus on San Francisco, to move beyond the attention on the U.S. and Silicon Valley. I sincerely believe that Chinese developers should take our technology, or any technology they think is good, to contribute to the Chinese community, and even developers from Latin America and Brazil should contribute to the Brazilian community. I want to see more Chinese, Portuguese, or Spanish-speaking developers get involved, bringing these technologies back home and localizing them. Moreover, I think these meme coins should be more internationalized, rather than just focusing on a specific place. I would love to see everyone pick up these technologies to create a new generation of applications.

What we provide is completely free and open-source; we ask for nothing in return. We genuinely hope everyone can use this technology to build their dreams and create their products. I think my key insight is that what’s competing is not the framework itself; it’s not the open-source technology itself, because it’s already free, and that value proposition doesn’t hold.

You should use it to build applications without needing to tell others that you are using it. Although I certainly hope you will tell others. You also don’t have to give back contributions, although we certainly welcome them, but the most important thing is to use it to reshape the internet and build a new generation of applications. Additionally, we really hope to have more contributors, especially from China and other regions.

I hope more people will get involved to help build and ensure that the technology truly meets the needs. We hope all our documentation can be translated. If anyone is willing to help translate our documentation into Chinese, that would be great. We want to ensure that this technology is not just mine; it can also become yours, and that’s what I want.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。