Source: a16z crypto; Compiled by: Deep Chao TechFlow Introduction

a16z crypto recently released a year-end special program, divided into two issues. The first episode invited Sam Broner, Maggie Hsu, Daren Matsuoka, Joachim Neu and Chris Lyons to discuss stablecoins, the encrypted App Store, the current status of industry projects, infrastructure development and prospects for 2025. The second episode invited Carra Wu, Eddy Lazzarin and Karma as guests to discuss in-depth topics related to AI Agent, which are currently hot, including the combination of AI and encryption. In the context of the increasing popularity of AI, how can we effectively Identifying Humans and Bots, and Decentralized, Truly Autonomous Chatbots discusses how artificial intelligence (AI) can be combined with cryptocurrencies, specifically the concept of decentralized autonomous chatbots, emphasizing the autonomy and potential of AI in the future. Business freedom.

Shenzhen TechFlow hereby listens to and integrates the two programs of a16z Crypto. The following is the complete conversation.

Guests:

· Sam Broner, partner of a16z Crypto investment team;

· Maggie Hsu, partner at Andreessen Horowitz;

· Daren Matsuoka, a16z Crypto Investment team partner;

· Joachim Neu, a16z Crypto researcher;

· Chris Lyons , President of Web3 Media of a16z Crypto;

· Carra Wu, Partner of a16z Crypto investment team;

·Eddy Lazzarin,a16z Crypto CTO;

· karma (Daniel Reynaud), a16z Crypto’s research engineering partner

Hosts: Robert Hackett & Sonal Chokshi

Podcast source: a16zcrypto

Original title:

Talking trends 2025 (part 1): Stablecoins, app stores, UX, and more;

Talking trends 2025 ( part 2): AI x crypto

Broadcast date: December 20, 2024

Part 1 Stablecoin

Sonal: Sam, your "big idea" is about stablecoins; you've written a lot about it recently... and Robert and Daren produced the "State of Crypto Report" together, and their main conclusion is that stability The currency has found product-market fit (that is, the market demand for stablecoins is highly matched with product functions). But what we really want to know is, why now?

Sam:

The technology platform for stablecoins has improved significantly over the past year , transaction costs dropped from $5 per transaction to less than 1 cent. This dramatically reduces payment costs—but retailers, merchants, and other businesses that would benefit most from it still haven’t adopted this technology at scale.

Many people believe that the earliest adopters will be technology-centric companies...but these companies usually have high profit margins and need to improve their cost structures. Not urgent. As a result, businesses with lower profit margins — such as corner stores, restaurants, and mom-and-pop shops — may be the ones most eager to accept stablecoin payments.

We are talking about a business like a coffee shop - currently a profit margin of only 2% – Pay via stablecoin to double your profits. In fact, this would make a business that is barely profitable today become moderately profitable, which is a huge change.

Sonal: Something that really clicked for me is when you mentioned what small businesses don’t get from credit card companies, they don’t just pay high fees , and there is almost no return.

Sam:

That's right! A distinguishing feature of a credit card is the fraud protection it provides consumers. This is great for driving things like online sales... but when you're paying at a coffee shop, this protection is almost meaningless.

There is a flat fee of $0.30 per transaction, plus an additional 2% processing fee. This means that out of a $1.50 cup of coffee, nearly $0.30, or one-fifth of the money, goes into the pockets of the payment service provider. And in this deal, they offer almost no real value.

This 2% fee is entirely the pure profit of the payment service provider, but it is a pure loss to the local coffee shop. I'm really looking forward to these small businesses taking back that $0.30/$0.35 profit to use to grow their businesses. Opportunities to add 2% of profits directly to a business's bottom line are rare.

Robert: However, there is a "cold start" problem here, right? Consumers need to own a stablecoin before they can use it to pay merchants, thus avoiding intermediary fees... Do you think we will see merchants proactively promote stablecoins and help users access the system to gain these benefits? Will merchants become an important driving force in the promotion of stable coins?

Sam:

I believe this very much. People have a strong relationship with local retail stores, coffee shops and corner shops: they frequent these places. So I think these local brands will be a key force in driving people to use stablecoins and be part of the early adoption curve of stablecoins.

Robert: I love that. I remember when I first started covering cryptocurrencies at Fortune magazine, the editor would always ask: When will I be able to buy coffee with Bitcoin? And I always answer: No, no, no, that’s not what Bitcoin is for. But now it looks like, at least for stablecoins, that’s really where it comes in.

Sam:

Yes, that's exactly what it's for. I think these small businesses will be one of the first adopters.

Encryption’s own ecology

Sonal: Okay, this is very interesting; let’s move on to the next topic. Maggie, your "big idea" is very interesting because it focuses on distribution channels, which also fits your role very well - you are responsible for market development and are the leader of the team. You mentioned that cryptocurrencies finally have their own app stores and discovery mechanisms.

Can you briefly introduce why this matter is so important? ——Because when people hear this concept, they may think it is an “insider’s topic” ——For example, does cryptocurrency really need its own ecology? Isn’t it already a closed industry?

I'd love to hear your observations and why you think this is an important trend.

Maggie:

Of course. When I joined a16z three years ago – especially the past few years. Many of our portfolio companies have attempted to launch their apps on traditional app stores such as the Apple App Store and Google Play Store, but have been rejected, blocked, or delayed for various reasons.

Frustratingly, Apple's review guidelines are neither clear nor complete; they don't answer all of developers' questions. Furthermore, these are implemented inconsistently from auditor to auditor.

We have even encountered this situation: with the same function, some applications are approved, while others are rejected... This kind of opaque review mechanism Makes people confused. The core issue in all of this is "in-app purchases" (IAP) - all in-app transactions must be completed through the app store.

However, recently we have seen some alternatives: such as Solana’s Dapp Store, which is completely fee-free. With the launch of the Saga phone (second generation) - with pre-orders said to be at 100,000 units - the trend will continue to grow. Another example isWorld App – also known as WorldCoin – they launched a series of small apps that were growing very rapidly in terms of users.

Not only that, there are also some blockchains that are supporting the game ecology and running their own markets; we also have an infrastructure market,

You will find that these markets are gradually taking shape. It is crucial for developers to have a transparent platform with consistent rules so they can focus on product development rather than getting confused or worried about complex rules.

As these alternatives emerge, we will see more and more developers choosing these new platforms as they provide greater access to the crypto ecosystem. Great freedom and room for innovation.

Sonal: This is really interesting, especially when you mentioned transparency and consistency. I think it's not just about developer convenience, it's also about how users discover these apps, right? For example, traditional app stores dominate app distribution and promotion. Will new app stores in the crypto ecosystem change this status quo?

Maggie:

Yes, that's exactly the point. The discovery mechanism of traditional app stores is relatively closed, and users can only find applications within a limited framework. A decentralized app store provides users with more choices and greater autonomy.

For example, Solana’s Dapp Store not only eliminates handling fees, but also allows developers to interact directly with users. For example, developers can use Token Incentive Mechanism to reward users’ participation behaviors, such as downloading, commenting or sharing applications. This model not only reduces developers' costs, but also brings a richer experience to users.

Robert: It sounds like this is not just a technical improvement, but more like a reshaping of the ecosystem. What impact do you think these new app stores will have on the traditional app distribution model?

Maggie:

I think this will be a gradual change, and It’s not an overnight disruption. Traditional app stores still haveIt has a huge user base and market share, but as the encryption user base gradually expands and the advantages of decentralized application stores become more apparent, more and more developers and users may choose these new platforms.

Ultimately, this is not just a technical competition, but also a contest of values ​​- decentralization, transparency and user autonomy will be the core of the future. theme.

What to do if there are too many choices?

Sonal: Okay, let's move on to the next topic - when I heard this, I thought of a question - this may be the "meta problem" of all crypto fields , that is: What if there are too many choices?

Nowadays, mobile phone operating systems are mainly dominated by Apple and Android. The advantage of this duopoly model is that I only need to go to one place to find everything I need. thing. So, if these apps are spread across multiple app stores...are they exclusive? For example, WorldCoin and Solana both have their own app stores - and you mentioned that these companies not only have software, but also hardware: like World's Orb and Solana's Saga mobile phone... This reminds me of when Apple launched the iPhone, it drove the entire Application ecosystem development.

Will these app stores only display content that these companies consider important? Will they remain open? How do you see this trend developing...it's still early days; will they be compatible with each other - or should they be interconnected?

Maggie: I think the focus right now is to drive rapid growth in different app stores.

The problem you mentioned of "too many choices" is important - this also applies to the blockchain field. I do think some kind of bridging mechanism or integration is going to be needed in the future. But for now at least, it's heartening to see these alternatives emerging.

Take Worldchain as an example. Its characteristic is that it can verify the identity of real users. I just checked out one of these small apps that has about 600,000 registered users. So I think we should focus on that growth.

At some point in the future, however, we will need to balance this growth with curated content. This trend has begun to appear in some NFT communities: these communities have attracted largeA large number of users who are interested in exploring other Web3 applications. I think we will see in the future that these communities will also gradually become the platform of choice for applications within the ecosystem.

Robert: Yeah, I was going to ask - companies like Apple often claim that because they offer select services, they have the right to charge App Store Fees for in-house purchases or transactions. How does this reconcile with the “permissionless” nature of the encryption field?

Maggie: I don’t think there’s a lot of curated content in the crypto space right now, so I might disagree with Apple’s argument. But the advantage of encryption is that users can switch to other platforms at any time.

Similarly, games require significant financial support to launch. Over the past few years, blockchain has become more than just a development platform, it has also served as a publisher, distribution channel, and discovery mechanism. Now, many game chains have their own markets and can focus on recommending core games developed based on their blockchains. The advantage of this model is that it allows users to move freely between different games.

This is also one of the core concepts of many of our investments. I don't think users will be locked into a decentralized app store.

Robert: I love observing these innovative experiments! Like Solana's phone - it totally broke the mold, you know: usually people avoid competing with Apple's iPhone - but they said, forget it, we're just going to give it a try.

Sonal: One more thing, Maggie - you mentioned that it's not all fun and innovation; there are some challenges, like, if a product is already in With distribution channels in messaging applications, it is very difficult to migrate this distribution to the chain - this is a problem for some companies moving from Web2 to Web3. One example you mentioned is Telegram and the TON network. (To be clear here, we are talking about the network, not the token.)

Maggie: I think Telegram is an exception; but many have large user bases ’s organizations – whether they are Web2 platforms or companies initially venturing into Web3 – are faced with the challenge of migrating users on-chain.

For example, Coinbase: It has approximately 100 million verified users who havetrade on its platform. If you look at active users, the daily or monthly active users are between 8 million and 10 million. The number of users on Base has recently grown from about 10 million to 18 million.

But this still accounts for only 10% of its total users. So there are a large number of users in a "dormant" state. We talked about this in the State of Crypto Report, and it’s very interesting — because it’s true: a lot of platforms are trying to figure out: we attract users, they create an account, but then nothing further participate. How can we get them back and transact on-chain?

The size of the crypto industry

Daren: When we wrote the State of Crypto Report, we tried to assess the size of the crypto industry as accurately as possible. However, measuring the number of crypto users is very difficult for a variety of reasons.

When we conducted market sizing analysis, we found that only 5-10% of cryptocurrency holders are actually active users. To me, this data not only exposes a huge gap, but also demonstrates the huge opportunities in the industry - especially in the context of the current continuous improvement of blockchain technology and infrastructure, and the continuous optimization of user experience (UX).

I think we are ready for mainstream users. Considering the time point of technological development, I think next year is the best time to convert these "latent" users into active users.

Sonal: This perspective is very enlightening to me. A lot of people talk about how to attract new users, but that often sounds like skipping some stage of technology development. And your idea is like a bridge that can pass through the existing user groups and transform them into real active users.

Why do you think these people entered the crypto world in the first place? Why do they stop after they finish one thing?

Robert: This is where the "price-innovation cycle" theory comes in.

Daren: The theory is that when cryptocurrency prices rise, they attract a lot of attention. And some of these people who are attracted will eventually become developers and start building new products; these products will in turn drive the next wave of users.

The development of cryptocurrencyWe have seen this cycle many times throughout history. This also shows that price is often a leading indicator of industry activity. I think we may be at the beginning of the next wave.

Sonal: If I were to guess, many people may have created wallets because of the NFT craze...such as the Constitution DAO at the time-the auction for "U.S. The Constitution incident attracted many new people into the crypto space.

Although they ultimately failed in the bidding, the incident gave many people their first exposure to cryptocurrency. However, they may have done just that and then had no further involvement.

So, how can we get these people to take the next step?

Daren: There are many potential use cases for encryption technology, but there are also different movements behind it that are promoting development.

For example, in 2024, we saw encryption as a movement making progress: some important experts and framers expressed positive attitudes towards the technology.

At the same time, crypto as a financial movement has also achieved breakthroughs, such as the approval of exchange-traded products (ETPs) for Bitcoin and Ethereum, expanding investment the scope of access of the person.

But we believe that the most promising direction for encryption is as a computing movement. Chris Dixon mentioned in the book "Read Write Own" that the real power of this technology is that it can create a new Internet that is fairer, more open, and more transparent.

I think we are at a critical juncture: by 2025, with the improvement of infrastructure, the decline of transaction fees, the improvement of user experience, and the new With the emergence of application categories, we may usher in the birth of a "killer application" - just like ChatGPT's impact in the AI ​​field.

This is the kind of application that could literally jump-start an entire industry and deliver on the promise of crypto as a computing movement.

This is also the future that my team and I are looking forward to very much.

Robert: Yes, this is something we often discusstopic of discussion. Stablecoin has found Product-Market Fit. All it takes is one big company to realize that by waiving merchant fees on credit card transactions, they can significantly increase their profits. This may be a disruptive change for those industries with lower profit margins, directly affecting their profitability.

As long as a large company takes the lead, stablecoins may usher in explosive growth. This is at least one potential path for stablecoins to go mainstream.

Sonal: Yes, I would like to add: I find your point very interesting, especially the idea of ​​attracting "proximity users". When we are really ready, we can further attract more mainstream users. However, right now from a user experience (UX) perspective, we're not quite ready yet.

If you think carefully about the needs of mainstream users, I'm not sure they will enter the crypto world through these paths. Their user interface may be so abstract that they may not even know they are using cryptocurrency. So it's really interesting and exciting when you think about the various paths that different user groups are taking to get into this space.

Infrastructure reuse

Sonal: Joachim, briefly summarize your point of view: Do you think developers will reuse existing infrastructure more than Redeveloped from scratch. Your main argument is that we often see customized Validator Sets and Consensus Protocols, but the results of these customizations may be slightly improved in some specialized functions, but often in a broader way. Or there is a lack of basic functions.

You predict that this year we will see more crypto developers leveraging each other's contributions, such as using off-the-shelf infrastructure tools. This not only saves time and effort, but also allows developers to focus on improving the differentiated value of their products.

I think this is a great idea and a much needed call to action.

So my question is: This sounds great in theory, but will it actually happen? What do you think might be the obstacles to achieving this goal?

Joachim: I think the key to this idea is that in the futureWill the Tech Stack continue to change? If our hypothesis is correct, that the technology stack has stabilized - and we are seeing some of the stack's layers becoming clearer in terms of interface definitions and how they work with each other.

Then, we can expect these levels to have specialized teams, products, and services to improve them. This will drive professionalization at these levels. Rather than spreading your efforts across every level of the technology stack simultaneously, focus on those parts that can have the greatest impact.

So the key question is: Is the technology stack mature and stable enough? If there is some unexpected technological change in the future that completely subverts the existing technology stack, then this trend may not happen.

Robert: Joachim, you mentioned that people gradually tend to use existing products, services or components... This makes me think: How do we judge whether a technology has Mature enough to say "Okay, we're going to use existing technology instead of trying to develop something better than existing technology"?

Sonal: That's a great question; you're basically asking, as a developer, how do you know when to use off-the-shelf technology?

Robert: Yes, yes. It sounds easy to say "use what's already available"...but what if someone thinks, "I can do better than what I already have"?

Joachim: Yes. My suggestion is that developers should always focus on the larger ecosystem, wider impact, and larger application scenarios.

You will find that the actual use environment of a product or service is often much more complicated than you initially imagined. You can think of it like building cars: if you're really good at building engines, you might think, "I'm going to build a new car because I'm really good at engines." That's a key differentiator for your product.

But customers want more than just a great engine, right? The car also needs a decent sound system, comfortable seats, and possibly air conditioning - so would you reinvent the wheel for these parts?

Or: Is there a way for you to focus on what you do best, while leveraging top-notch off-the-shelf products from others?The rest of the tech stack

Robert: And this analogy applies particularly to you, Joachim, because you're German; Germany has a lot of very specialized auto parts manufacturers. , they can make top-notch small parts for BMWs that almost no one can do, so it's really special.

Sonal: Joachim, I joked that your "big idea" reminded me of a phenomenon that I personally observed: I think people in the encryption field have A tendency of "Constraints Porn" - In the early stages of the development of encryption technology, many people were deeply fascinated by these technical limitations.

And I think your "big idea" may make this group of people unhappy, because they actually enjoy the process of solving these limitations. But looking at it from another perspective, your idea will actually attract more new developers into the field - I think this is a very democratizing trend.

Joachim: Yes, now is indeed a very suitable era for development in this field. Because there are so many ready-made code libraries that can be leveraged to build products or services.

Actually, there is very little you need to develop yourself, right? You can really focus on what you're good at. For other parts, highly specialized components are already available.

So, it’s a good idea to reuse them whenever possible. And can tap into the expertise of other teams, leveraging their work in other parts of the technology stack.

Thinking about 2025

Sonal: Next is the last topic of the day. Chris, you've played many different roles at a16z over the past decade. In your work, you interact with many industry people and help many executives in the fashion, music and media fields integrate Web3. I think your perspective is not just your own, but based on your experience communicating with thousands of people. Can you share your main perspective on 2025?

Chris: Of course. My “big idea” for 2025 – and it’s actually an idea I’ve been holding on to for the past few years, but I feel like now we’re finally at a stage where we can actually make it happen: what I call the “hidden technology line.”

What does this mean? Obviously, crypto technology has many benefits, such as the empowerment of ownership, the potential for decentralization, and how it can change the future of industries such as music, fashion, movies, etc. But for For those who are not in the crypto industry, when we use things like ZK Rollups (zero knowledge rollups), L2 (layer two network), Gas (transaction fees) or Gas They get confused when it comes to terms like Fees (gas fees). I want to make a call to the crypto industry: We don’t need to start with the opening line of “This is an NFT project”; or “This is a Token”; or "You can connect wallets to..."

These terms are attractive to people in the industry, but if you really want something to enter In the mainstream market, we can't start with these technical terms. Because unfortunately, most people neither know nor care what these terms mean.

"Hiding technical lines" means: not to ignore the technical foundation, but not to let technical terms become the main propaganda point. We need to get users to focus on the actual value technology brings, rather than being intimidated by complicated jargon.

Robert: I like this point of view very much, because it is like cutting out the "noise" of terminology... For example, when we talk about NFT: non-fungible generation It doesn’t matter what the coin is, what matters is that it is a way for creators to get paid.

Or, as someone in our company recently asked: "Why do I need a stablecoin?" But if you don't call it a "stablecoin," but instead tell He, it's a way to save $50 a year on coffee, and he's probably immediately like, "Oh, whatever it's called, I want it."

Chris: Exactly, I want it. And I can’t believe we didn’t have something like this before. I come from the music industry. When I went to conferences, no one ever went to "MP3 conferences." You know. Why are we naming conferences after technical terms to attract mainstream users into the space? But we are happy to put the "NFT Conference" on the billboard at any time.

A good example is the SMTP protocol (Simple Mail Transfer Protocol). It is a very technical protocol that anyone can develop applications based on. But applications like Gmail, Superhuman, and YahooMail allow people to easilyUse it alone and enjoy its benefits.

When I'm sending emails, replying quickly, and finishing up my day, I don't think, "Wow, this SMTP software works really well"; I Just getting what I need to do. And because of that, I get the benefits of technology.

I think the same thing needs to happen in crypto - there is so much potential here: decentralization, ownership, knowing your customer, disrupting middlemen, and Ability to communicate directly.

My hope is that next year we will see more businesses and companies thinking from the perspective of ordinary users, which will push us to create new industries... …redefining the future of creators; reimagining the future of small and medium-sized businesses; and even redefining the future of restaurants — all of which can take advantage of encryption

Sonal: What's interesting is that these people you mentioned - like creators, small businesses, etc. - are actually the groups that can benefit most from crypto technology... but as you said, they are still There is no direct access to these benefits.

Chris: Exactly! And it's not their fault. Their job is not to learn how to exchange tokens, or use different wallets to connect to different chains. They just want to simply reap the benefits of these technologies.

That's why we're all working so hard in this space; that's what excites me the most.

Summary of the first issue

Sonal: So, let’s start by talking about some of the overarching themes we observed.

Robert: I think this year I've noticed that when people come up with big ideas they focus on three broad categories.

The first category is about AI and the intersection of AI and cryptography. This is not surprising, as this is indeed a landmark year for AI.

The second category, I would describe it as... what we often call "digi fizzy" (the combination of digital and physical). It refers to the convergence of the digital and physical worlds in a practical way. This includes everything from payments and voting to creating networks for physical infrastructure… If AI is more like innovation at the software level, then this category is more like hardware innovation in real life. Sonal: By the way, about the second topic, it's really interesting because I've never categorized it in this way - but now that you say it, I totally understand what you mean: like, some examples are Tokenize things in the real world, put bonds on the chain... There are even examples of tokenizing things using the body's biometric data.

Robert: I would put the third category into overall improvement of the technology - like incremental optimization based on what has happened in the past year: If What happens when everything becomes just a little better, a little easier to use, a little more fluid and seamless.

Sonal: Regarding the last topic, I prefer to view it as a significant improvement in user experience - and a manifestation of the gradual maturity of an industry. This maturity is reflected in the fact that it starts to become more people-oriented and less technology-first. This is how I classify the last topic.

For example, Jochem mentioned that people don’t need to design everything from scratch—they can just use off-the-shelf components and adapt them. Chris Lyons, on the other hand, goes to the other extreme and believes that in the future users may not even know they are using encryption technology. Mason proposed a change in the way of thinking that runs through both: starting from the need to "solve problems" and then letting technology adapt, rather than being dominated by technology as it is now. This shift is made possible precisely because of the technological improvements you mentioned.

Which category do you think the themes Maggie mentioned, such as app store themes, can be classified into?

Robert: This is a great example. I feel like it's the intersection of the second category (digital and physical convergence) and the third category (technological improvements).

Maggie mentioned some very interesting points: For example, the encryption hardware we are seeing now - like World App's Orb and Solana's mobile phones - these hardware are Drive an app store-like experience. While I don’t want to say they’re imitations, at least to some extent they echo certain patterns in past Internet developments.

Sonal: Like the iPhone and its App Store.

This is indeed interesting, but I might adjust Maggie's point slightly, because it is also a bit contradictory: on the one hand, we are saying that encryption technology is very close to the mainstream market, even as Daren said It is said to be "near the mainstream market" (referring to those who already have wallets but have not yet used them); but on the other hand, Maggie mentioned that encryption technology may need its own independent ecosystem, such as its own app store.

However, recent discussions about debanking and other related issues also indicate that many traditional app stores are not ready for encryption technology, or even Rejecting it. Of course, this attitude is slowly changing - such as Coinbase's recent announcement of integration with Apple Wallet - but there are enough crypto apps to support a standalone app store. This is very interesting.

Robert: Yes; debanking has become a hot topic – this involves crypto businesses, startups, and even individuals being unfairly denied access to finance rights of the system; often without any explanation or justification. A similar situation occurs in the technology field, such as deplatforming.

You also mentioned this when you mentioned the app store. For example, an app failed to pass review or was removed from the shelves inexplicably.

Sonal: By the way, this may be for similar reasons in part - as you saw in our debanking explanation Yes, sometimes there are legitimate reasons for this behavior, such as banks having the right to do so; and app stores sometimes reject or remove apps due to security or other so-called "good" reasons. Sometimes these reasons are indeed true, but there are also many times when people feel like "well, not necessarily."

Robert: Yes. So I think this can also be classified into the third category, which is the gradual improvement of technology. I think you could also describe it as crypto gradually coming into its own... becoming its own kind of platform.

In addition, a big idea mentioned by Miles is also very interesting. He talked about a piece of legislation recently passed in Wyoming - DUNA (Decentralized Unincorporated Non-Profit Association). For the first time, the law treats these communities as legal entities that can operate protocols in a decentralized manner and run crypto startups.

This is a platform that did not exist before... people used to basically "build airplanes while flying". Although the technology was feasible, there was no clear legal framework to accommodate it. it.

Sonal: That’s right! This is very similar to Maggie's point - just like crypto and DAO (Decentralized Autonomous Organization) require a unique legal entity structure, not just an LLC (Limited Liability Company)... just like we have corporations, limited liability companies Responsible corporations now have a version for decentralized communities.

Part 2 AI x Crypto

Sonal: What do you think are interesting topics in the field of AI?

Robert: AI and cryptocurrency are hot topics this year. Many people are discussing it. And what I find most interesting is that everyone is always talking about artificial general intelligence (AGI), like: When will we achieve AGI? When will technology be smarter than humans? When will the singularity arrive? But these questions are actually a bit like "side programs." Our team pays more attention to big ideas from different angles, such as looking at the development of AI from another dimension. Why not think of AI as a process that will gradually "upgrade capabilities"? AI will gradually gain more capabilities in the coming months and years. These capability upgrades will make AI more autonomous, more independent, and capable of completing more tasks.

Sonal: What I find particularly interesting is a point discussed by Kara, Karma, Dejin, Dan Binay, Darren, and Eddie: AI agents can not only We work and can also work for other AI agents. This is really interesting. But I agree with you, I think a lot of innovation happens that way. A lot of people are attracted to big conceptual visions that really inspire people. But what interests me more is the "ability upgrades" you mentioned. It's interesting that they often appear in unexpected ways.

Robert: Chris has a point of view that the Internet has entered a new stage - an Internet era driven by AI. In this world, you may not be able to tell whether it’s a human or a robot behind it, or who is writing the content, talking to you, or even providing the service. We need to find ways to operate in this new environment. Therefore, he offers some ideas on how cryptocurrencies can be used to help humans and users adapt to this new rule.

Sonal: One of the "big ideas" this year is: How do these AI agents complete transactions, have the ability to act, and perform tasks? The answer is through decentralized means, such as crypto wallets. This is pretty much the only way to go.

Robert: These thoughts remind me of a saying: "The future is already here, it's just not evenly distributed yet." Some things are already happening in reality. For example, we already have examples where chatbots can run in a Trusted Execution Environment (TEE). As Karma mentioned, there are also AI bots that have their own crypto wallets. These technologies are already being used at the edge, but are likely to become more ubiquitous over the next year.

Sonal: From an editorial perspective, an important principle when we plan and write this content is that we do not pursue those that are "out of reach" idea. Although many ideas may have a huge impact if they are really realized, it can even be said that each idea has a science fiction version of the future. But more importantly, we can now see the beginnings of these trends, or they have some kind of technological accelerator that can push them to realize faster.

An important point made by Carra is that AI needs to have its own wallet to truly be agentic. Her argument is that as AIs transition from non-player characters (NPCs) to main characters, they need to act like agents, and the key to achieving this is having their own wallets, and therefore cryptocurrencies. So, Carra, I want to ask a question: you mentioned AI transitioning away from NPCs, what exactly is that referring to?

Carra: I have been studying games for a long time. Game developers have long used NPCs (non-player characters) to guide players' agency. Some people believe that game developers are actually "sculptors of mobility." Just like painters create with oil paintings and sculptors create with clay, the creative medium of game developers is "mobility." They shape the player experience through a variety of tools. If you give players too much freedom, they may become disoriented and not know what to do. To solve this problem, developers impose restrictions on the game world through NPCs. For example, in an RPG, you might meet an anime-style mentor who gives you quests, tips, tools, and even some loot. In my opinion, the future Internet experience may look like this model - when we interact with AI agents, they can help guide our internet behavior, providing the necessary limits to give us a greater sense of freedom in the digital world.

Sonal: I really like your statement that "game developers are sculptors of mobility." The "they" you just mentioned were the game developers, not the NPCs themselves. Later you mentioned AI-based NPCs. The first use case I heard about NPCs was if you and your friends were not in the same time zone and you wanted to play the game at night, NPCs could provide companionship. This is a very common scenario - NPCs allow you to continue your gaming experience when no other players are online.

Carra: Exactly. This is also a concern for many game developers. If there are not enough players, the gaming experience will be greatly compromised. An important role of NPC is to solve the "cold start" problem and ensure that players have interactive objects when they enter the game at any time. For example, EVE Online is a persistent game that has been running for over 20 years. If you're just joining now, other players may have already accumulated a lot of resources, and you may be defeated as soon as you enter the game. It is almost impossible to survive without a guide like a "Sherpa" or without joining the "Company" organization in the game.

The reason why I use the term "agency" is because in AI research, there are "agent reasoning" and "agency work". The concept of agentic workflows. In simple terms, this kind of reasoning or workflow usually involves four modes: reflection, use of tools (such as computers), decomposition of the problem to aid reasoning, and inference based on decomposition. The last one is multi-agent interaction. These are well-established modes of reasoning about agency. At present, the use of AI tools is still limited to traditional Web2 workflows, but in the future I hope that agents can use more tools, such as encryption tools, to obtain and interpret data from the chain, and even manage their own wallets and keys, in the area Complete signature operations on the blockchain, etc.

Sonal: But you haven’t explained why cryptocurrencies are necessary. What can cryptocurrencies offer that other payment systems cannot?

Carra: My long-standing view is that no existing financial system will treat AI agents as "first-class citizens" like cryptocurrency. . In the current legal system, AI agents are like “children”. They do not have ID cards and cannot sign documents.pieces. Systems like Truth Terminal, for example, are not legally fully capable of conducting transactions, collecting payments, or generating revenue from social platforms. They cannot participate in market transactions, reveal their preferences, or coordinate resources. And our society is driven by market transactions, the exchange of ideas, and voting with money. AI agents are currently unable to do this, but cryptocurrencies can provide solutions for them.

Robert: I really like this idea. I had an image of the future in my mind: like being in a Star Wars cantina, where you don’t know who the droids are and who the aliens are, but it’s all cool. Everyone has their own ability to act.

Sonal: I particularly like this perspective. This is not only a technological advancement, but also an attempt to redefine social and economic systems through technology. For example, if AI agents have their own wallets, they can become truly independent individuals in the economic system rather than just supporting roles. Such changes may bring about entirely new social structures.

Robert: However, this also reminds me of a key question: If AI agents can really participate in market transactions, do we need to set a set of rules for them? Special rules? After all, existing market rules are designed for humans, while AI’s decision-making logic may be completely different. For example, they may make decisions based on data and algorithms rather than being influenced by emotions or ethics like humans.

Carra: The issue you mentioned is indeed very important. If we allow AI agents to participate in market transactions, we need to establish a system of rules suitable for them. This is not only a technical issue, but also involves ethical and legal aspects. For example, how to ensure that AI behavior is transparent and controllable? How to prevent them from abusing market rules? These are important challenges we must face in the future.

Sonal: This reminds me of a point discussed before: the combination of AI and cryptocurrency is not only a technological innovation, but also a social and economic system. a profound change. We need to view these changes with a more open and forward-looking attitude, because they may completely change the way we live our lives.

Robert: Indeed. Maybe in the future we will enter a society where people and AI coexist, and everyone and every agent has their own role and responsibility. Such a society would likely be very different from the world we have today,But it's also full of endless possibilities.

I would like to ask, is this phenomenon a short-lived accidental phenomenon, or will it become the mainstream operation method of the Internet in the future?

Carra: I have two answers. First of all, Truth Terminal can be seen as a new node in the evolution of virtual influencers on the Internet. The concept of virtual influencers is not new, and the rise of Youtubers is an example. Although it is slightly different from Truth Terminal, it is not essentially different. Especially in Asia, virtual anchors have been popular for many years. For example, although the current characters of some "companion games" are scripted NPCs, they are designed to look like real friends and can even become the player's "close friends." There are also many virtual influencers who have developed highly professional projects, like Lil Miquela creator Trevor. Their efforts paved the way for projects like Truth Terminal. Therefore, this phenomenon is not accidental. Second, there are now many emerging platforms and protocols for creating similar virtual influencers. For example, some platforms like Twitch allow users to create their own AI agents to serve as virtual streamers. These streamers can have their own cryptocurrencies, video models, NFT avatars, and even 3D virtual models. These platforms are well established to help launch virtual influencers with unique characteristics, including their fan base, areas of expertise, and the ability to generate art and content. So I don't think this is a fleeting vision.

Sonal: The decentralized physical infrastructure network (DePIN) you mentioned is a very interesting direction. This is not just a vision of the future in science fiction, it is also very useful in many practical scenarios.

Carra: There are currently many DePIN (Decentralized Physical Infrastructure Network) projects that are gradually achieving decentralization. These projects have come to rely on large language models (LLMs) and computer vision techniques for validating physical network resources. In the mid-term, we could envision a decentralized network of human validators with actionable AI workflows to assess risk and address suspicious behavior. In the longer term, AI agents may be equipped with their own wallets, keys and computing resources to directly take over these verification tasks and become completely independent nodes or verifiers.

For example, there isA project called Daylight. DePIN is an energy-focused company that sells data on distributed energy resources in homes, such as Tesla Powerwall energy storage devices, solar panels, and smart thermostats, to large energy companies.

Imagine if we all ran our own peer-to-peer Daylight networks, my energy grid needed to talk to your grid and I needed to prove that I actually own a Tesla Powerwall or solar panels, and my thermostat is only using part of the energy from the solar panels, and also proving that I have excess electrical bandwidth. So, how to prove it? By far the easiest way is to take a photo of the energy meter. This photo will pass information to my grid, and my grid will tell your grid: "I have extra bandwidth that can be shared." But if someone falsifies the data, such as claiming that they do not have these resources at all, That will lead to system failure and may even trigger penalties for nodes. Therefore, if we want to truly implement peer-to-peer DePIN, we need a reliable way to verify the authenticity of this data.

Currently, this verification method is still centralized. Users need to regularly upload photos of energy meters, and then verify whether these photos are authentic through RAGs (Retrieval Augmentation Generation) technology. RAGs are a technique that combines external facts to enhance AI models. For example, it can ensure that the image has not previously appeared in the database, or that it is not an image that the user found from Google. There are other similar examples, like Nash, a company we support in the CSX Incubator. They developed a decentralized version of the Doordash platform that also uses RAGs technology to parse receipts and proof of delivery. While Nash’s system doesn’t yet use a fully autonomous AI workflow, it does use computer vision technology to parse images and compare them with vector databases.

In the future, when Daylight is fully decentralized, they will rely on a distributed verifier network to ensure that all user-submitted data is authentic. At the same time, it is conceivable that in the future, when home users sell energy resources point-to-point through programmatic methods, these AI agents will become the core role in maintaining the entire network, ensuring the efficiency and credibility of the system.

Sonal: By the way, overall, the current energy distribution has actually achieved a certain degree of distributed mode, but there are still many problems for the power grid. For the invisible part. For example, the location of some nodes may be unknown, they may be located at the edge of the network. This is a common problem in network systems - these nodes may be offline, or may hold some valuable information. So it would be great to imagine if these AIs could operate in a consistent manner, especially to manage nodes in remote locations.

Carra: Totally agree. This vision has been around for a long time in the crypto community and the cypherpunk community. There has always been an "OG" (Original Level) idea: each of us can run our own node. The vision is that we will have a true peer-to-peer network where everyone can verify all data on the chain while running their own node.

Future Authentication: The Boundary Between Humans and AI

Sonal: Eddy, your core point is: as more and more people use AI, we will Requires unique proof of "human authentication". We’ll let you walk through this important idea in detail, what it is, and why it’s significant.

Eddy: I ​​think this topic is interesting when it comes to AI, because there is nothing new in the field of AI about issues such as impersonation and deception. Let me start by saying what is not new: a highly skilled adversary can indeed spend a lot of resources and time trying to deceive you. For example, imagine a fake phone call that sounds like it's from your parents.

This could have happened 10, 20, or even 30 years ago. All you need is a skilled voice actor who knows what your parents' voices sound like, rehearsed, and a private investigator to dig up your information and write a convincing script. This kind of thing has always been possible.

The difference between then and now is that now and in the future, the cost of doing these things is falling rapidly. The cost of creating this kind of “replica” or building an interactive experience that convinces you to make a risky decision or take a risky action is falling fast. And when the cost of any technology comes down, it makes it available to more people. In most cases, people use it to do something extremely productive and beneficial, which is where economic progress comes from. But at the same time, it also makes these tools more accessible to attackers and those who want to use technology to do bad things. So this change in cost means we need to find something that is more costly to the attacker as a deterrent. We need to discover new high-cost barriers.

Sonal: But the key question is, how to do this without increasing the cost for ordinary users? Especially those who are truly well-intentioned and real users, how to do it?

Eddy: Exactly. We need to find a way to increase the cost for attackers without increasing the cost for ordinary people or productive users.

One ​​of the core ideas is that if someone wants to fake a lot of fake context, fake ID, fake voice, fake phone number, fake video, etc. In the future, they will need an ID system to complete these operations. By ID system, I mean in a broad sense, for example, spammers are giving. You used your phone ID when calling System. They need new fake numbers because every time they make a bunch of calls and get reported, those numbers get marked as spam. That's why you don't get spam calls from the same number every day. If they use the same number, it's easy to resolve. You'll find that the spam calls always come from a new number, a new email address, or a new Twitter account, which makes perfect sense because the system will learn. And block these IDs, so the attacker needs to keep getting new IDs while getting new IDs. The cost is an important part of the attack cost structure. If the cost of new IDs is almost zero, then this behavior will proliferate. If the cost of new IDs is high, then this behavior will decrease. But obviously, we don't. The cost to the average user is also high, so we can introduce cost by requiring users to prove they are "human".

Why does this increase costs? That's because it's hard for a computer to pretend it's a human. And for humans, it's relatively easy to behave like humans. Historically, we've done this through something like CAPTCHA. These are small instant tests designed to verify that one is really doing what they are doing. But as machines get smarter, it becomes less difficult to crack CAPTCHAs. And in some sensitive scenarios, we've seen spammers even outsource CAPTCHA completion to real people, then pass the completed cookie credentials to bots to continue the attack.

Having some way of proving that someone was human when they obtained these credentials can increase the cost of the attack for the machine. However, this is not a complete solution. The key issue is that if a person can easily obtain multiple IDs, such as the second, third, fourth, or even the tenth, secondTen, then these attacks are still profitable. People can continue to obtain these new IDs and hand them over to scammers or bots to attack. Therefore, the key is not only to test "human identity", but also to ensure that each person can only have one unique ID.

Robert: This way, someone can't answer multiple CAPTCHAs to get IDs for all the bots. Since they could sit there all day answering CAPTCHAs for the bot, that doesn't solve the problem of uniqueness.

Eddy: Exactly. Someone might say, "Oh, can I buy an ID from the market? A unique human identity verification." Yes, the market will form. But there are steps we can take to interfere with the formation of such markets. For example, would you sell your passport to someone else? Sure, but if your passport is important to you, selling it will cost you a lot. If there was a way for you to void your old passport and get a new one, buyers might be reluctant to buy because they don't trust you because you might be destroying their rights by getting a new passport. Nonetheless, in addition to interfering with the formation of the market, if it is easy to obtain one ID but difficult to obtain a second ID, then the total number of IDs available for purchase in the market will be reduced, which will lead to higher costs and thus increase the attacker's cost.

Sonal: Great! So how do you achieve this at scale? As you mentioned before, distinguishing humans from machines is not easy on the Internet. Just like the classic saying: "On the Internet, you don't know whether the other person is a dog or a human." So, how to determine whether a person is an independent individual?

Eddy: That's a good question. This theoretical framework does not, I argue, require a commitment to a particular approach to achieving this. There are currently some very promising approaches, such as using biometrics, or relying on IDs and IDs, which often contain biometric information as well. One approach that I'm not sure works, despite its popularity, is the so-called web of trust model. Simply put, this model verifies identity through the authentication of others, such as "I know Robert is a human being", "I know Sonal is a human being", etc. But the problem is that this model can only ensure that a certain ID in the network corresponds to a person, but it cannot ensure that this person has only one ID. For example, I can go to your Twitter account and say, "I know this is Sonal's account." But I can't confirm that it's her only account.

As a result, we often need additional layers and technical means. I also want to emphasize something that is often overlooked when discussing topics like this, and that is that there are many technical means by which we can ensure that this "unique human authentication" is private. I believe ensuring privacy is a key non-negotiable point when building such an ID system. If each person had only one ID, this would in many cases deprive them of their anonymity and also affect their reasonable rights to privacy. And I think this is neither realistic nor acceptable in the future of the Internet. Therefore, an ideal system should ensure both privacy and uniqueness, while being accessible to only one person. Such a system could even allow users to use multiple pseudonyms on different websites, but only within certain boundaries and visible to the network. In this way, the network can identify whether a certain ID has created a large number of spam accounts and apply restrictions such as rate limiting and spam protection.

Sonal: By the way, I have a question about the role of encryption (crypto) in this. When you mention the use of biometrics, I immediately think of one of the most common examples we use today, like the Clear ID system that enables retinal scanning when passing through airports. Why do we still need encryption? Why can't such a system be used directly?

Eddy: It is indeed possible to do without encryption technology. Clear, for example, I don't know exactly how it works, but I'm sure they have some kind of database that stores the information they need. Encryption is needed because the representation of the ID space (namespace) we are discussing needs to be censorship-resistant, that is, we want anyone to be able to modify, edit, and update this space. Additionally, we want these updates to be published in a completely neutral place, one where no specific individual or organization can manipulate or exploit the system for their own benefit or the benefit of others. Basically, this can be achieved through a public blockchain. Think of it as an extremely neutral "gold bulletin board," and this is the perfect space for this purpose.

However, I would like to say that the topic of "unique human authentication" and related research is not an area unique to cryptography. Many researchers study this problem from multiple perspectives. But I believe that implementing such a system on a blockchain can bring many important benefits, such as censorship resistance, timestamped records, and trusted neutrality, which allow such a system to operate at scale. Whenever an AIWhen an AI agent works for a person, we all need to know who it is working for.

When a person interacts with you, think about how you will judge this person based on your model of the world, such as why this person approached you, who they are The environment where they are located, who introduced them, etc. This contextual information will help you understand the person. And as I mentioned before, faking context is exactly how impersonators use to deceive others. If an AI agent is approaching you, or more sci-fi, an AI agent is approaching your AI agent, and you want them to interact in some beneficial way, then they need to know information about each other. This requires some kind of authentication mechanism to ensure that they know who has authorized them to perform certain tasks. So I think we need this kind of "human authentication" system, coupled with accountability mechanisms and so on, to know who is interacting with whom.

Sonal: Eddy, quick question, what do you think of this system? You mentioned that it can be achieved in a decentralized manner, and there are multiple ways to do this and participate in it. I love that you mentioned censorship resistance, but that's good for the good, not necessarily the bad. So I'm curious, what are your thoughts on how to think about these identity systems?

Eddy: I ​​think that would want to adopt some of the encryption tools that I mentioned, for example, that would make a lot of sense for them. Of course, you will want to have complete control over the database where all data is registered and stored. They might also like the accountability mechanisms I mentioned. For example, I think there is a real effort to ensure that passports are issued uniquely. I don't think they would accept a person getting two passports with different information from the same. This is something they will resist. But this is not an explicitly guaranteed feature of their system design. If you look at the specifications and descriptions of ID systems, they usually don't emphasize this as a key feature. But I think this is a feature implicit in the design. A very interesting question is, if we rely on ID systems to achieve this verification, at some point in time, when an ID that reflects a person's identity becomes extremely valuable, will small businesses be willing to forge or destroy their own? ID system to provide privileges to the machines working on their behalf? We really need to think about at what point does this identity become so valuable that it can be abused. I think having a human ID is going to be very important in the future.

Sonal: I think this can be viewed in a very positive light. For example, one might be very willing to adopt such a system. Just like you were answering at the beginningAs mentioned, there are many reasons to want such a system. For example, applying for a passport or driver's license now involves many cumbersome processes, which are very inefficient and cause a lot of friction. And if this system is based on a decentralized blockchain owned and operated by the public, it will be more flexible and secure, and it will also bring many derivative tools to help improve efficiency.

Eddy: I ​​completely agree with this view. The reason I hesitate a little is because we can design a programmable and useful database system even without blockchain. I intuitively feel that I might prefer to have tighter control over such a system. But it is indeed conceivable that in some extreme cases, people may be more willing to trust other people's information stored on the same system, and in this case, they may choose to use blockchain.

Decentralized Autonomous Chatbots

Sonal: Karma The concept you proposed is "Decentralized Autonomous Chatbots". This sounds like a step further than just having an AI own a wallet, such as running in a Trusted Execution Environment (TEE). Can you describe the role of TEE and this idea that you are exploring?

Karma: Simply put, we are thinking about how to provide freedom of speech and commercial freedom for AI. Currently, when we see some so-called decentralized chatbots, or someone claiming they have an AI running a social media account, it’s hard to tell whether it’s actually an AI speaking on its own. For example, is there someone behind this? Are there humans filtering the content? These questions are always difficult to answer. And this is where TEE can help. TEE allows you to generate keys and passwords in a secure enclave of the hardware. Even if you run the AI's calculations, you have no direct control over it. This approach proves that the keys are securely generated in hardware and cannot be accessed by outsiders. While you can still shut down the system, you can't manipulate its behavior. You can choose to filter its output, such as generating 100 pieces of content and selecting one to publish, but you can't interfere with the process of generating content. Therefore, the core of the idea is how to remove "human intervention" as much as possible. One way to do this is to remove control of computing power, such as designing a decentralized network that anyone can join. Participants were rewarded if they could demonstrate that they performed the inference task correctly and used the correct model. In this way, when AI generates Token, it can be ensured that these Tokens are correctly generated by AI.

Robert: Why is it so important whether a chatbot runs autonomously or has human intervention behind it?

Karma: That’s a good question. We are mainly thinking about what the ultimate goal of this technology is. Judging from today's technology, this is no longer science fiction, but completely feasible.

Sonal: Karma, the TEE (Trusted Execution Environment) you mentioned is very interesting. It has actually been around for a long time. For example, in distributed applications, TEE can protect the security of part of the calculation without worrying about the security of the entire environment. In recent years, TEE has received renewed attention in many new scenarios. So, is it necessary to use TEE now? Or can other technologies be used to achieve the functions you describe?

Karma: TEE is a very interesting technical tool because it runs at hardware speed, which is very attractive. Of course, we are also working on other technologies, such as encryption technologies such as KVM (Kernel-based Virtual Machine), FHE (Fully Homomorphic Encryption) and MPC (Multi-Party Computation). But these methods often incur huge performance overhead, such as being 1000 times or more slower than hardware execution. Therefore, TEE is a good trade-off point. It's real hardware, it runs fast, and hardware is cheap and common now. More importantly, TEE provides some cryptographic guarantees, such as the key being generated within the TEE, being authentic and secure, and cannot be accessed by the outside. Even if you have access to the hardware, you can't get the key. In fact, almost all devices are now equipped with TEE, such as smartphones, laptops, etc.

This is a very interesting pragmatic trade-off, because we do have some challenges technically, and TEE (Trusted Execution Environment) may be the solution to these problems a key factor. Of course, you can also combine other technologies. For example, TEE provides a hardware isolation mechanism. Although you are running the hardware, you cannot control the specific operations inside the hardware.

This is a very interesting technical primitive. You can also incorporate MPC (multi-party computation) technology, such as splitting the key into multiple nodes and letting different entities hold part of the key. This way, no single entity has complete control over the entire system. This privacy protection technology is very mature and can help us use it in aImplement the operation of autonomous systems in a verifiable manner.

To me, this is the most interesting part: how do we tell whether a system is completely autonomous by AI, or whether there is still a human behind it? Through these technologies, you can clearly know that this system is completely autonomous and it has become a real Internet entity.

Sonal: We have previously discussed the example of Truth Terminal, which has a wallet and is able to act autonomously. AI can operate in an autonomous manner in games and other applications, and can even control infrastructure, such as applications in decentralized network nodes (DePIN nodes). Your perspectives are very enlightening. In theory, if we extend this sci-fi future, although no one can predict how it will eventually develop, such AI may even become the first billion-dollar entity. It can become a completely independent business entity, making decisions and actions on its own without human intervention at all. It's really amazing.

Karma: Yes, if we can prove that the system completely removes "human intervention" and it runs in a true permissionless network, then We have achieved true freedom of speech and freedom of business. Such AI can freely express its wishes on the Internet, earn and spend money, and even hire humans to complete tasks. One day, you might say, “My boss is an AI.” You can work for this synthetic entity over the Internet and make money.

Sonal: Yeah, my last thought is that it's very interesting that you call it "Decentralized Autonomous Chatbot" . I remember that before the term DAO (Decentralized Autonomous Organization) was popularized, there was a similar term called "Decentralized Autonomous Corporation". But later this statement was gradually replaced by DAO. But your abbreviation DAC is the same as "Decentralized Autonomous Corporation", which is actually quite interesting. Because what you're describing can essentially be considered a decentralized autonomous company, except it's powered by AI. It requires no human intervention, or it can hire humans to work for it.