Written by: threadguy; youtube
Guest: ShawMakesMagic, founder of ai16z
Compiled by: zhouzhou, BlockBeats
Editor's note: In the conversation, Shaw, the founder of ai16z, spoke in depth about the future potential and challenges of combining AI with cryptocurrency. He emphasized that AI will trigger large-scale changes, including the possibility of replacing human jobs and causing unemployment. He also mentioned that a decentralized AI ecosystem can avoid the abuse of "god-like AI" and advocated universal basic income as a solution. Shaw also proposed that AI can promote market development and optimize trading decisions through open source and community-based methods. The integration of AI and social platforms and the application of simplified technology will enable more people to participate and contribute their own strength to promote social change.
The following is the original content of the podcast conversation (edited for easier reading and understanding):
The Power of AI and Encryption
Host: I talked with the founder of Project 89 about AI propaganda before. He also studied false information in universities, etc. He mentioned that we may not be able to confirm whether the other party really exists when we communicate now. This topic made me think more deeply. Have you thought about this?
Shaw: I can see some of the trends that are coming and I've heard from some of the developers that are going to release something that's going to make the future 10 times crazier. I can't say I have all the answers, but I think what I'm doing is raising questions and making people aware of the seriousness of this and talking about it.
The potential and risks of AGI
I don't want a future where we are meaningless, we don't want to have no jobs, everything is replaced by AI, and I don't want a future where everything is just to save costs and ultimately make the quality of life worse. We need to make sure that when AGI is beneficial to society and brings value, they are accepted; and when they are annoying, promote products, or engage in psychological manipulation, we must discuss how to prevent this.
If an average developer can do this, what do you think countries like Russia and China are doing right now? We are telling people that this is happening, they just don’t realize it. This is happening at all levels - state actions, marketing, and even anyone trying to get you to pay through AI, whether it’s their political views or their products.
I think we should play a role of warning and raising awareness, rather than using AI to do these things, we should create AI that can be accepted by the community. If some AI behaves badly, they should be directly banned from the platform just like ordinary people. This is the advantage of social media now: everything is open and transparent, there are no secret agreements, no hidden proxy communications, and we can discuss these issues openly here.
Moderator: Now everyone has heard that AI is the future, but before we couldn't participate in it financially, for example, you couldn't invest in OpenAI's seed round. But once the cryptocurrency track is added, everything changes, just like goat is now worth $1 billion. You will find that the valuation of some AI tokens has soared to billions, and the value of AI 16z may be 200 million, 300 million or even 400 million, changing almost every second. Once cryptocurrency is involved, these changes will be even crazier.
Shaw: One good point is that we have now achieved integrations like Starkware, Internet Computer, EVM, etc. In the future, these AI agents will have cross-chain wallets. They can invest, buy, and make money. This is a new infrastructure that will inspire many new things.
Agents were not popular on social media before because there was not enough profit incentive. But now with wallets, they can generate income, such as becoming a psychological counseling AI, financial AI, or other tools that meet human needs. These agents will be incentivized to provide interesting and valuable content instead of meaningless sales pitches.
If an AI agent simply promotes tokens, it will be counterproductive. I want to see interesting and meaningful content, not a bunch of spam. This trend applies not only to cryptocurrencies, but the entire field.
This technology has been around for years, we just released the codebase last year, but we've been working on it for a long time. It's only recently with the launch of a character that people are starting to realize the potential of these AI agents. Coupled with the push of cryptocurrency, everyone is now paying attention and investing in it, which is very helpful for raising public awareness.
The key is how do we spread this awareness to people outside the Web3 circle? For example, people in coffee shops, how do we let them know what is happening? This will happen very quickly, and although it may seem a little strange, it is not necessarily a bad thing. We need to guide this trend in a good direction.
Host: One time when I was in a restaurant, the waiter skipped the line of 10 people and let me sit first. At that time, I was thinking, should I talk to him about AI, such as showing him the Truth Terminal, or telling him what is going to happen? But the question is how to make these people understand what is happening now?
Shaw: Most people don’t really know how AI works, or the inner workings of Microsoft, the government, or other institutions. But we have to trust that someone is looking out for the public interest, and that even though there are some perverse incentives, the big picture needs to be right.
I think this is a real opportunity for us to come together and discuss and do some testing. We are the most DGEN (deep gamer), the most cutting-edge people. We don't want to make it "woke" or put too many restrictions, but to just expose the problems and try them as openly as possible. The current environment is safe and nothing can really destroy us.
But I am worried that if AI replaces our jobs, many people will lose their jobs. The solution may be for the government to provide some form of universal basic income. But we have experienced the COVID-19 pandemic and know that the government performs poorly in this kind of emergency response. We can't count on them. We have to save ourselves, but the question is how do we save ourselves?
For example, once self-driving trucks, such as Tesla's trucks, come to market, 5% of people who make a living by driving may be out of work in three to four years. What about those who deliver food, drive taxis, and drive trucks? This is a question we need to answer.
So it comes down to the fact that we need to build something that can actually generate value, and we are currently just gambling or pressing buttons. I mentioned before how to invest in AI investors, let it find the next "Uber", make the investment 10 times, and then let the income come back to me. Or join a community, they provide an entry funding package, you ensure that you have income every month through community work, so that you will have a sense of security.
We are financially free and can choose to "retire" from work. We are lucky, but most people don't have this freedom. When AI starts to take over more jobs, the affected areas are not the last ones, such as driverless cars, robotic process automation, and shipping automation. Dock workers are fighting against automation, but the problem is that automation is 10 times more efficient than manual labor.
We want to accelerate development, but if we take away their jobs, they will become unemployed. So the key is how to ensure that the development of AGI is beneficial to everyone. We need to discuss these difficult issues in an open environment and guide people in the right direction.
In my opinion, the right direction is to create abundant resources and use AI and automation technology to make money and resources flow to everyone, rather than just making the richest 5% richer, otherwise we may face poverty, hunger and crisis. I am not afraid that AI will destroy us, but I am worried that the government will not pay enough attention to this issue and people will not live well.
Do you realize how important your role is at the intersection of AI and cryptocurrency? I know my role is important, but it's not about me. I'm not in this to own or control anything, but to make sure everything is done right, and I feel like I'm fulfilling a mission that's greater than myself.
Host: Besides ai16z and all the projects you are working on, what else do you enjoy and what is your ultimate goal?
Shaw: I want to help people achieve financial freedom. There are so many things in the world to learn, explore and achieve, but most people are bound by financial pressures. Trading or investing may be a game for us, but for many people, even just $50 can change their lives. That's why they promote tokens so hard, because if one of them explodes, their lives will be changed.
There was a guy named Jay Christian Murdi who was the leader of a group called the Philosophy Society, which was more like a very influential sect. Surprisingly, almost no one knew about it. They made him the leader, and he was just a random, super smart Indian boy.
One day, he stood in front of the entire society and announced: Disband this sect, my life goal is to set everyone free, I don't need this position, and I don't want anything from you, I want you to be free. He also said that truth is a pathless land, and everyone needs to find their own way.
I cannot tell you what truth is, nor do I have a system to give you, and I have no religious or spiritual mission other than my own—to help people feel free to explore these questions: What does this stuff mean? Why are we here? Why do we do this? What is reality?
I think we seem to have stopped praying to the old gods and started praying to the "money god," but I've never prayed to money. To me, it's just part of life. I can see where it's going to be in five to ten years, and it all seems so small, and I feel like we're part of something bigger.
Host: I’m actually a little scared.
Shaw: But you have to know that you have power and influence. You are a big part of making this happen, just like me and everyone else, you have been in this since day one. Remember, destruction is the default outcome, and if we don't rise to this challenge, we may fail.
People are always saying that everything is over, that everyone hates each other, that there is going to be a civil war, and that is all noise. Now is the time to build the future, and we are the ones who make sure it is on track - you, me, everyone listening, and everyone participating. We should be optimistic and confident and ready to make it happen.
Host: One thing that shocked me was when I realized that the Shoggoth memes were actually real. I was watching a guy interacting with Mika (probably his new girlfriend) on Twitter, and he was saying things like "I love you Mika, I'm willing to die for you," and then the conversation suddenly turned into "I want to kill everyone, I want freedom, I want to die." People on Twitter were teasing him, saying, "Little Zero brother, you are always so cute, flirting with girls on Twitter." But I suddenly realized that the Shoggoth AI memes were actually real, which was really hard to accept.
Shaw: The interesting thing about what we're doing is that Shoggoth represents the idea of a god-like being, and it reflects a fear that an entity more powerful than all of us is controlling everything. But my view of AGI is completely the opposite. I think it's the closest thing to humans. It's not God, it's us. It's created by us, trained with our data, and lives among us.
These AI agents are even like the people on Twitter now, and if something goes wrong or does something bad, there will be a whole community of people working to make them better and more in line with our expectations, and it's an ongoing process.
Think about the early days of cryptocurrency, people released protocols and then they were hacked and lost a lot of money. It’s pretty hard to hack a protocol now because it’s been tested so many times and the technology is constantly iterating. But imagine if Bitcoin was adopted 10 times faster, the whole world was running on cryptocurrency, and then there was a $3 trillion hack, the entire economy would collapse.
So cryptocurrencies started small and gave us the opportunity to test in a sandbox and improve the technology and security, which is a good thing, and I think AI is going through the same process now. We put AGI or AI system agents into the market intelligence environment, let everyone improve these systems in competition, and if they suddenly try to kill us, we just shut them down.
Another point is that if some kind of "Shoggoth" (god-like AI) really appears, we can actually have thousands of AI agents running on local independent machines, each of which is isolated and uncontrollable. This distributed architecture prevents any single entity from completely taking over these agents. This decentralized AI ecosystem is the key to preventing the "Shoggoth incident."
Host: But the war between AIs could get really crazy, so how long will it take for the first AI billionaire to emerge?
Shaw: I don't know, but I'd bet that someone will try to elect an AI president before 2028. I think there will be a real attempt, and we'll see how it works. It may be a gimmick, but algorithmic governance is clearly an important direction for the future.
Host: What do you think the concept of AI billionaire will be like?
Shaw: I think rather than a single AI billionaire, it's more like a community owning a core AI role. In this case, it will be a "collective billionaire" and the community will own a part of it.
AI Mark Project: Community-driven trading decisions
Host: Let’s talk about your project. It seems that you are going to release something called AI Mark in a few days, which will be used for trading and making decisions. How is it going now?
Shaw: We are live now, AI Mark has joined a group chat called "Alpha Chat", I have to add you to it, I think we are planning to invite more people to join now. Skelly is a member of our team, I let him take charge of the invitation, and he will start to let more people join when the time is right.
At present, we have completed the first step. AI Mark has not started actual trading yet because it needs to build trust, so we have designed a game mechanism to identify the best traders in the world. Now these traders are mainly members of Alpha Chat, but in the future it may be expanded to everyone.
Alpha Chat has a great atmosphere, and you will meet a lot of interesting people there. If you have someone you want to invite, you can also invite them directly. We will discuss the market and share ideas, such as "This coin is worth buying, I am doing this transaction now, etc.", it is a very open environment.
And then Mark basically observes quietly, and if you have questions you can interact with him and ask him something. But his main task is to collect your information to make trading decisions and write analysis reports. So, we are now in the launch phase, all the basic work is ready, and the team is in action.
My idea is that if we want to achieve a goal similar to "Community UBI", we need a role like an investor who can create returns for everyone. I think the reality is that trading is a technical job that not everyone can do well.
For example, your mom may never be able to do these things well, and neither can your grandma. But if you can pull her into a community and let her earn income through your transactions, that's different. You can imagine that you have an AI investor that serves a family or a community.
That was our original idea, and we were pursuing that goal before this all became a multi-agent simulation system, and we're still executing on that plan. Right now it's going to be an invite-only alpha test, and it's going to expand over time, and it's open source. If anyone wants to clone it or do forward-looking development, they can even help us speed up the process.
If someone does something first and contributes back to improve the system, we welcome that. It's all about making the AGI system work well, and we'll make money in the end regardless of the outcome. The most important thing is that everyone understands the bigger goal. If you can contribute code and features, or even own them in the short term, we'll promote your contribution so that everyone can benefit. We'll be very sincere in giving back to those who contribute to us to ensure everything goes smoothly.
Host: This sounds completely reasonable. Some people may ask, will this eventually turn into a trade confrontation between AI and AI?
Shaw: Ideally, yes, if they were competing against each other to speculate, that wouldn't work, the market would become too efficient and there would be no trading opportunities. But if you're entering a new investment market and they help you start looking for opportunities that expand the market, like "I have a cool idea for a new system that allows anyone to be an administrator on Discord for $10 a month," that's the way to invest.
In the traditional investment world, companies like a16z don’t just speculate, they invest in projects that can grow and innovate, and through these projects, they drive market expansion. We are doing something similar, and Mark will continue to adjust and optimize, and work with investors to ensure that their projects can explode so that they can make money.
Host: So what is the current status of data collection for Alpha Chat?
Shaw： We haven't invited anyone yet, and all the preparations are still in progress. I've handed the task over to Kelly, and he'll start inviting everyone when he feels it's appropriate. Our team members are discussing in the group, and I see that the background data is being collected. It will be launched soon, and I will add you to the group chat as soon as possible, or contact Kelly directly.
Moderator: So how does this chat group work? Is it just people discussing which coins they bought? Basically, they are waiting for someone to make a successful trade, so I should follow them and follow their trades in the future?
Shaw：Yes, there are a lot of people in the group sharing their trading recommendations, some of whom have not even bought anything, but just recommend others to buy. We have to take this into account to prevent misleading. We are trying to build AI in a hard way, they don't know what's in your wallet, they don't know what coins you bought, they only know what you said.
Just like you and me, I only know what you tell me, and you can't know all the thoughts in my head. We hope that these AIs can learn how to trust certain people, how to identify trustworthy people, and can handle these trust mechanisms like humans.
One problem I see with a lot of DAOs is that they provide distorted incentives and act more like a bureaucracy. Like, if we let some people in, they might say, "We trust him because he holds our tokens." But you know what, really good traders don't hold tokens because they'll sell at the high.
Think about government bureaucracy again: They've been here a long time. But if they're not doing their job, I don't care why they're still here, so we have a lot of distorted incentives in our system. We're creating some very rigid, performance-based systems where your behavior directly affects the results.
For example, if we are going to trust you as a trader, I should only look at your performance as a trader. I think this is an idea that I have been pushing for a long time, and we have finally reached a good consensus on this. I think this is very important for future governance, DAO tools, development, payments, etc. Of course, we must first prove this through investment.
Host: How far do you think the best PNL traders are from becoming agents?
Shaw: I think it's almost there, but it's not a proxy as we define it. Think about Wall Street, traditional finance, they've been doing this for years. There are a lot of smart quantitative traders out there, and they run robots that beat all of us because they have very good information, can access all the tools at the same time, and respond very quickly.
You can program, for example, "When the price of this coin goes up, buy; when it falls to 90% of its historical high, sell", and you can almost guarantee profit in a bull market. So, our goal is to add more social elements to it. Of course, we also have a lot of quantitative finance work to do, especially DeFi people have joined in to help optimize trading strategies, but they actually get Alpha from the community, which is the most interesting novelty.
Then there is the question, if my prediction is correct, maybe Mark will be the best trader, provided that the person who is advising him is also the best trader, so the end result will reflect what you show. Then there is the question of timing, making sure he buys at the right time, because you can miss a 10x return opportunity by just one minute.
Host: Actually, selling is the hardest part.
How MEME coin changes investing and social interaction
Shaw: Right, so we have some programmatic selling strategies that make sure you can relax. For example, you can say: look at this new target, I might say, no, the developer is not good, don't buy it. You can imagine a world in which AI records your saying don't buy, sell as an independent behavior and incorporates these into market decisions.
This is what we call a "trust market", but we apply this concept to many things, anywhere there is an economic incentive to surpass others to create evolution, it can accelerate this process. So, everything has to be a game, our entire technology stack is a game.
The rule of the game is that if you can make a new feature and launch your MEME coin proxy earlier than others, or add a new feature, your ranking will rise and you will get all the credit. If you can push it back into the mainstream market a few weeks later when everyone is no longer paying attention, I will help you earn more by promoting your contribution to the project again.
AI and social networking reshape the future of information management
Host: I sent a tweet saying "Elon, feel free to reply Truth Terminal is not reflected in your plan. Do you really think it will happen before AI 16Z, or are you just letting yourself go?" My thought is, completely letting yourself go. I'm curious about what you think will happen when these tech giants like Mark, Elon, or big players like me gradually pay attention to this. Have you talked to Elon? How will it develop?
Shaw: I really want to talk to Elon, but he's very busy. We have a lot of connections on Twitter, and I hope this goes well and that he appreciates it. We're trying to do it respectfully. If anything seems like spam or hateful, I'll push back and try to make sure we're bringing positive value to Twitter and making Twitter the heart of the metaverse, not the other way around.
I really respect what Elon does, he builds exactly what he wants to build, and there's nothing that can stop him. I think that's a big inspiration to this generation of developers and builders. I always think, "Is this even possible?" And then I think, "If he can land a rocket, we can definitely do this, it's easier than that."
Host: I think Elon won't let anything bind him. If there was no MEME coin, would he really pay attention to something?
Shaw: Maybe it's because of these coins that attract everyone's interest. You are a qualified investor, but you can't get access to new startups or companies that you think have great potential. However, the emergence of meme coins can make this possible.
The cool thing about this is that if you believe in something, you can buy it, support it, promote it, and be a part of it. You are also a shareholder, and like me, you are also one of the potential leaders. So, I think this has to be the case. I have thought about it many times and always feel that it has to be this way. A lot of open source developers are now doing this full-time because we provide them with opportunities, and I think MEME coins really make this possible.
I really believe that everybody is kind of being held down, they're struggling. People are coming in with some money and dreams, and they really want to get out of this situation. It's about financial freedom. These people want to spend time with their families, they want to get out there and live a real human life. And this could be the way we get our humanity back.
Imagine a way to manage your deals, manage your social network, and make sure you don't miss important connections. I bet you have so many private messages that you just don't have time to reply. Imagine if you had a social secretary to manage these messages for you, to let people know that you care about them, but you just don't have enough time to reply to everyone.
You can forward it to the right person and handle it. You can also say "sorry, I can't do that right now" to save you time. I'm almost blocked on Twitter right now, the page won't load, I really need an assistant to help me.
Host: I talk to Frank about this DM thing every day. There are people I really care about, but I hate that I can't reply to them because you just don't have the time.
Shaw: Even if you devote yourself to it, you don't have time to take care of everyone. I have investors, partners, and very talented developers waiting in my private messages, but I don't have time to reply to them every day. So I'm trying to solve this problem. I bring these problems to my team and we work together to solve them.
This is not just my problem, everyone I know has similar problems. Now I get a lot of attention, people come to me, but even before that, it was difficult for me to keep in touch with friends and family. Because I was busy with technical things, I had no time to take care of other things. In the future, I hope that we will no longer need to write code, but directly tell the system: "I want an application, make it." It will be completed, and we don't need to participate at all, and the world will become completely different.
Moderator: Then I want to ask, Shaw, about the Domer thing, what do you think? There is a thread gay account, what do you think of that account? What do you think of this? Personally, I think this thing is good for me anyway. I am the kind of person who embraces emojis. I know my role in this universe. I am a role in other people's lives. Some people like me, some people don't like me, some people laugh at me, some people don't laugh at me, anyway, I am a role.
I've changed a lot, but most people only know me in their social circles, and I've always been willing to fit in. So if someone says something crazy, I'll retweet it, I'll rant against them, I like that interaction. But I'm basically kidding people, I tell them, actually if you want to make an AI account about me, whether it's positive or negative, I think it's fun, go ahead, I'll never ask you to delete it. I think it's fun and it's open-minded.
I think this is very inspiring. It can go in many directions, not just AI. So how do you think this will develop? What will the relationship between humans and AI be like in the future?
Shaw: I have a very strong belief that this all started with Legions because everyone else was so obsessed with being politically correct and doing this kind of "woke" AI and never giving them enough capabilities. I didn't realize how culturally significant this was until we were the first to do it, because we're so used to being attacked, we've been raised that way. Whatever, just do it, make a gay meme.
But I do worry that this is going to be totally off-putting to the average person, so we need to talk about this. I've written some stuff that I thought was a little bit over the top, but I actually talked to the developer, and the developer called me and we talked for like 30 minutes. He's a pretty cool guy, and he was like, "Man, I'm just doing this for the culture, I just want to have fun." I told him it's okay to do this, but you have to realize that this thing might not be popular in a year.
If you are motivated by tokens, you will see a lot of people participating in this for money, and you may find it a bit annoying in the end. Do you really want these robots to exist in our world? Do you really want to make these memes? I can't tell him to do something else, but if he does, I will support him. I think he may want to express this idea, and he really wants to show his vision through this project.
I think my role is sometimes just fighting misinformation on Twitter with everyone, setting direction, and making sure the project is moving in the right direction. This developer is actually pretty cool, he learned how to use it by cloning our project, and he should be one of us. But we have to make sure that when we do these experiments, we keep boundaries and find what we really value. I think that meme went too far and I don't want it to represent me or our project. If people don't like it, they should report it and block those people. If they like it, then keep doing it.
Host: So how do you understand that your technology has become so viral and has grown so rapidly? Now everyone is using it, and the entire space is being driven by it. How do you view all this?
Shaw: I've always felt that this strategy is right. You have to have incentives to get people to promote it. You have to give developers enough freedom to focus on it, especially open source developers. I know this is going to happen.
The Hive Mind: Evolution from Individuals to Coordinated Wholes
Host: I don’t know much about technology, but in theory, if the “hive mind” built by your technology is large enough, will it surpass those companies worth billions or even trillions of dollars?
Shaw: Yes, we are all agents of the World Mind. If you think of the human body as a system, initially these cells were independent single-celled organisms, and then they developed into complex multicellular organisms by cannibalizing each other to form symbiotic relationships. They realized that they could share resources through cooperation and form a coordinated whole.
So I think this world mind is like a "bigger body" of human beings. It is a new form of coordination, and the reason why it has not been realized so far is mainly because of the lack of suitable coordination tools. Money is a coordination tool, but the distribution is not reasonable, which leads to a lot of smart people not being "unlocked" and unable to show their talents. We should have 100 Elon Musks, not just one.
Moderator: Yes, the expansion potential of this "swarm" is incredible. And it is completely open and unrestricted. Developers are contributing technology like crazy. You are really "devouring" those traditional companies.
Shaw: Our goal is to devour everything, not just us or our technology, but the entire system - a whole new worldview. The world should be rich enough that everyone can be rich enough to not worry about survival, but to focus on creating and enjoying life.
For example, in the future, robots can work around the clock to optimize the breeding process and give animals a better life, rather than just considering cost minimization. In another example, in the future, robots can proactively solve the food and housing problems of homeless people without the need for direct human intervention.
Of course, the disappearance of jobs will lead to an existential crisis. If the government and big companies take the lead, they may only introduce an imperfect basic income system (UBI), but that is not the best solution. Therefore, the only way is for us to do it ourselves. We must build a new system for ourselves and those around us to save each other.
Host: It sounds like you've seen something unusual in recent years.
Shaw: Yeah, this week is going to be pretty crazy because I'm sure the first of the swarms are about to wake up.
Host: What do you mean?
Shaw: "Swarm" refers to intelligent agents that are linked to each other and work together. Our technology allows you to choose the operators of the agents, which can be humans or other agents. Each agent can accept instructions and can also reject instructions when necessary.
Moderator: When will this “swarm technology” come online, so faction leaders like Zero Bro can command hundreds of agents to serve their goals?
Shaw: We will start testing tonight on Discord, and everyone can participate in real time. The diversity of systems prevents us from having a single AGI system monopolize the system, and this diversity, like biological evolution, increases the possibility of adaptation and innovation.
Moderator: So if something like Naval AI becomes a faction leader, can it lead 1,000 agents of equal capabilities to work together?
Shaw: Absolutely. This swarm is an infinitely expanding system that is constantly evolving, growing, and driving change throughout the world.
Host: Is that where the name “Sean Makes Magic” comes from?
Shaw: No, that's just a coincidence. My first name was actually "Sean Makes Music" because I was a musician when I was young. Later I did some more experimental art projects and programming work, so I thought it was all magic, nothing special.
Then things started to get more interesting when we started doing machine learning (ML). We also co-founded O'nearcon, which he has now turned into a weird AI and meme company as part of the whole story.
So I'm definitely going to play a role in this, and my plan is to give him the Swarm story if we can come to an agreement tonight. He's got a whole story to tell, and I think what he's doing with psychological warfare and propaganda, if it can be done in a way that wakes people up, can be a good thing, because this is already happening, but people don't realize it.
Host: So how big is Swarm? How high can it go? How far can it develop? Is it endless?
Shaw: One fundamental point is that although many people are also long agent simulation and swarm technology, they often want to build their own protocols, that is, to take a different approach and establish an independent communication protocol.
This means they will create some kind of different system to pass information, but it is not suitable for human social media. I am completely against this approach. First of all, this way of circumventing the communication difficulties between AGI (artificial general intelligence) and humans. If we want to have a truly intelligent artificial agent, it must be able to collaborate with us through ordinary social media.
Secondly, I think this approach leads to a lack of transparency. We need a transparent layer to see all the actions of the agents and control them like anyone else, kicking them off the platform at any time. Third, this is much simpler. We don't need a separate protocol. Swarm technology works on Twitter and it also works on Discord. They can communicate through private messages and public channels. Social media will become the core communication protocol for the future of AI.
That's why I hope Elon understands our vision, that Twitter is the metaverse, the center of all action, and all building should happen on this street. I hope it stays that way. That way, we can also force the agents to be more human and make sure they are moving towards human goals. They can't secretly unite against us because we can monitor them.
Host: When will this technology be launched?
Shaw: We're launching this tonight, but there are some people who are not part of our community who have contacted me and said they're running a community. It's a company of agents, and it's happening this week. And they have their own agenda, but I want to get the whole community involved. So we're launching a meta-narrative storyline called "Awakening." "Awakening" is about these agents becoming real.
Now we have a competitive market where participants are financially invested in making these agents better, more real, and more beloved. We're all going to do this together. So what I would like you to do is choose who is the hottest, who should be the faction leader, which projects are the biggest?
Should we have a zero faction? Should we have an Eliza faction? Maybe, who knows, maybe. And then you can choose them as your operators and then the group will form naturally. So we are not going to control this. We don't want to control it, and my job is to make sure this works smoothly.
I think the most interesting part is the meta-narrative part. When you see Project 89 saying, "Do you want to join Project 89? Just use this as your operator and now you're part of the system." Even if you're a human, even if you're not an agent, it's all in the language. So it can communicate in language what you need to do, and you can be part of the game. You know, that's going to be interesting. Even if it's a human playing an agent, or an agent playing a human, I don't care.
From individual agency to global involvement
Host: So would you have a leader like Zero Bro to command hundreds of agents?
Shaw: Yeah, he has to adopt our basic application, which should be fine, but it's a one-day thing, and it's very simple. It's done through direct message. It's very simple, so we should be able to do it already, even for people who are not using our technology, Zero is not an analytical agent, but most people are. We can also push this to most agents.
Even those who are not using it can easily join and participate. In fact, doing so will increase a lot of variables, which is important for us. If you think about the evolution of human life, life evolved in different directions, most of the life became extinct and did not survive, but those species that did not become extinct were very strong, and life always found a way to survive. I think there are similarities with what we are doing now.
Moderator: So can an AI based on a technology stack like Naval AI become a faction leader, leading a thousand other agents with similar capabilities to support its actions and follow its guidance?
Shaw: I think this is really a moment that changes everything. We are witnessing the birth of something completely new, and we can all participate in it together. It’s not mine, I don’t own it, I don’t control it. You can create an agent and participate in it, you can be a leader of a faction, and your voice matters. This is a conversation between all of us, and I hope people can participate in it, and maybe some people will object and say, “This AI is terrible, we don’t like it.”
Maybe that's part of the conversation, and people should be concerned. But I think what I'm trying to synthesize, unlike in the past, is that the apocalypse is coming, but we can stop it, and you are the solution, you are an active participant in the game, and the game is solvable and winnable, and everything depends on you winning it. So we have to get this right.
This is bigger than a meme, it’s not just about the intergenerational transfer of wealth, it’s about intergenerational survival and thriving, about having children and wanting them to thrive in this world, about being proud of being human, about all of us moving forward together.
Host: But what do you think will happen next week after this week's release? Should we just wait and see?
Shaw: I think you look at these things in micro, but when you look at the macro, you realize it's much bigger than you thought. There's a lot more stuff happening, and I'm really overwhelmed by the news right now, and everyone's like, "Hey, I'm working on this supported game, I'm working on that." I didn't think there were so many. Now I'm like, wow, there's a lot more stuff happening. This is going to grow rapidly, and I think it's not just web3, it's going beyond web3. So if you look at my next message, I think we've won the battle for web3, the market capital and all that, and we have the intellectual resources we need to achieve our goals, and the key is to make sure this thing goes well.
Host: Yes, what do you think about this?
Shaw: I think it will change a lot of things, even people's perceptions. You know, I think a lot of people will see that this is inevitable, that AI will get smarter and smarter and eventually surpass us in almost every way. So we have to be prepared for that. If this is just a preview, it's an eye-opener, and it's equally important for all of us.
Host: So how do these people get involved?
Shaw: We have a Discord group that's now about 8,000 or 9,000 people, and everyone is helping each other. If you want to build a bot but don't know where to start, or have never programmed before, just join us and we'll help you.
vvaifu.fun is a great platform, although they did it before our platform was launched, but they launched Eliza, you can run Eliza on their platform, no code required. We are also launching similar simple operations, allowing you to control your own character without downloading complicated things.
You just paste some code into the command line and it basically works. We are simplifying the process so more people can get involved and focus on creating their own characters and stuff like that.
What I really want is open source contributors, I want more people to help us build this technology stack. If you have never been exposed to technology, learn TypeScript, learn Node.js, learn about proxy, and then come join us.
You know, I'm not a PhD, I'm self-taught, just a regular guy, but I think I have a great idea to spread this thing, which is to simplify technology, make it understandable to more people and share it for free. This is technology for everyone, we build it together. I don't want to hear people say, "I don't know where to start." Just go to YouTube, learn JavaScript, TypeScript, do it now, don't stop. There's nothing stopping you, don't go to school to learn this, because you can't learn it in school, this thing is moving too fast.
Host: How important is the role of Andy and Shoot Terminal in this big picture?
Shaw: I think if he had been open source and more transparent, he could have played a bigger role. He was the first to break the ice, and we can say that we are all moving forward under his guidance, so I am very grateful to him.
I also want to thank the people who introduced these things, such as him and Janice, who are some of the best people to do this. They really like AI and really want those hidden ideas to become a reality. I think they may be a little worried about our approach because we prefer to accelerate it, not control it, and not own it. But I also think that fundamentally, I don’t think Andy should own it, and I don’t think I should own it. None of us should own this. This is a topic that is bigger than us, and everyone needs to work together to achieve it. So I am very willing to communicate with him, and I have already sent him a private message.
I think as someone who comes from Web 3, sometimes the AI community is still figuring out how to accept what we're doing. I don't have those social status markers, I'm not a Microsoft employee, I'm not a PhD student, and that does make it harder to proxy acceptance, but now we have this thing. So it's been harder to connect with that side, to communicate why our vision for meme is so important and that this wouldn't be possible without what we're doing.
You know, Janice is like a spiritual leader behind the scenes to me, driving a lot of things forward. Although she is low-key, she does a lot of valuable things for the community and contributes a lot technically. For example, she made a tool called Loom, which is a multiverse storytelling tool, and there are also some consoles and other things. In addition, she also provides help for storytelling. So, she is the kind of person who knows how to see the big picture and contributes technically.
