Translation整理：CaptainZ

Welcome to the Web 3 special presented by A16Z. This episode focuses on the intersection of artificial intelligence (AI) and cryptocurrency (Crypto). We have invited Chris Dixon, the founder and managing partner of A16Z Crypto, and David George, a partner at A16Z Growth Fund. They will delve into topics such as the flaws in internet economic models, new opportunities for creators, and the profound impact of changes in large platforms. This episode is part of A16Z's "AI Revolution Dialogue Series" and coincides with the release of Chris Dixon's bestselling book "Read Write Own," making it particularly significant. For more information, please refer to the links in the show notes.

It is important to note that the content of this program does not constitute tax, business, legal, or investment advice. Please visit A16Z.com/disclosures for more important information, including our investment list. Below is the transcript of the interview:

Host: Chris, thank you for joining our show. I'm glad to have this conversation with you. You are currently focused on the field of cryptocurrency. Can you share your overall views on the interaction between AI and cryptocurrency?

Chris Dixon: Of course, I'm happy to be here. I have always believed that waves of technology often appear in pairs or groups, just like cloud computing, mobile internet, and social networks over the past 15 years, which have reinforced each other. The mobile internet brought computing devices to billions of people, social networks became killer applications for attracting users, and cloud computing provided the infrastructure for all of this. None of these three can be missing. I remember there were debates about which was more important, but it turned out they complemented each other and were all essential.

Now, I believe that AI, cryptocurrency, and new hardware (such as robots, self-driving cars, virtual reality devices, etc.) are forming the three pillars of a new wave of technology, and they will also promote each other. Cryptocurrency—this is also a key focus in my book—provides a whole new way of structuring internet services. It is not just a technology but a new paradigm for building networks, with many characteristics that traditional methods cannot achieve. I think this will greatly benefit many fields.

Many people simply equate cryptocurrency with Bitcoin or meme coins, but in my view and that of many smart people in the industry, this is far from its essence. The intersection of AI and cryptocurrency takes many forms. First, a straightforward combination is the focus of our investments: using this new architecture to build AI systems. For example, a core issue in the current AI field is whether AI will be controlled by a few large companies or by a broader community in the future. This involves the issue of open source. I am surprised to find that over the past decade, the AI field has become increasingly closed from being completely open (with papers published and code shared). Large companies lock up technology under the guise of "safety," but I believe this is more about commercial interests than real safety needs.

Fortunately, there are still some open-source models, such as LLaMA, Flux, and Mistral, but their level of openness is still concerning. The model weights are not fully public, the data pipelines are opaque, and whether these models can truly be replicated is questionable. Moreover, these open-source projects often rely on a single company for support and can be shut down at any time due to strategic adjustments. Therefore, we have invested in some blockchain-based internet service stacks aimed at providing decentralized open-source infrastructure for AI. For example, the Jensen project builds a computing layer through crowdsourcing, similar to the Airbnb model: startups can submit computing tasks to the network, supported by those with idle computing resources, while the blockchain manages supply-demand matching and economic ledgers.

Another example is Story Protocol, which redefines the way intellectual property is registered. You can create an image, a video, or a piece of music and record its copyright and usage terms on the blockchain. These terms are designed based on existing copyright laws and have international applicability. You can set rules, such as "allow adaptations and derivative works, but I require 10% of the revenue." This creates an open market, replacing the traditional business model that requires negotiations one by one. Currently, only large companies like OpenAI can reach billion-dollar deals with Shutterstock, while small creators often face either theft or neglect. Story Protocol provides an equal platform for everyone.

The core of this model is "composability," a common theme in the blockchain world and a concept I specifically discuss in my book. It is similar to the successful path of open-source software—countless people contribute small pieces of code, ultimately piecing together a powerful system. Linux grew from 0% market share in the 90s to 90% today, thanks to this power. Story Protocol is the same; you can imagine one person creating a character, another adding new elements, and yet another mixing them together, ultimately forming a superhero universe. As long as the revenue flows back as agreed, the incentives for creators are secured. This model embraces new technology while providing an economic model for creators, which is the most exciting part of the combination of AI and cryptocurrency for me.

Host: The new economic model you mentioned is indeed thought-provoking. David, you previously mentioned that the emergence of ChatGPT might break a certain contract of the internet. Can you elaborate on that?

Chris Dixon: Yes, I have a chapter in my book called "New Covenant," which discusses this. The internet has been successful largely because it has a clever incentive mechanism that allows 5 billion people to join voluntarily without a central authority enforcing it. Over the past 20 years, the internet has gradually formed an implicit economic contract, especially between social networks, search engines, and content creators. Take Google as an example; website owners allow Google to crawl their content and display summaries, on the condition that Google will return traffic. Creators make money through traffic, whether through ads, subscriptions, or other models. This mutually beneficial relationship is the foundation of the internet's prosperity.

But occasionally, this contract gets broken. For instance, Google's "One Boxing" feature directly displays answers without redirecting to the original website, which has harmed sites like Stack Overflow, Wikipedia, and Yelp. User experience may have improved, but creators' traffic has decreased. Now, the rise of AI further challenges this contract. Chatbots can directly generate illustrations or recipes, and users no longer need to click on the original website. If AI systems operate this way, traffic will no longer flow back, breaking the survival basis for creators.

These AI systems rely on data training under the old contract, but the new model no longer adheres to the old rules. I worry that the future internet will become a closed system dominated by three to five large companies, while other billions of websites decline due to loss of traffic. This makes me uneasy—will the internet revert to the broadcast television model of the 1970s, with only a few channels? What benefit does such a world bring to startups, innovation, and creativity? How will long-tail websites survive? How will new things break through?

I am not saying that cryptocurrency is the only solution, but at least we must acknowledge that the current situation has broken the original incentive mechanism and consider whether this is a good thing. If not, how should we design new mechanisms? Story Protocol is an attempt to rebuild the incentive system for creators through blockchain.

Host: You mentioned that AI, cryptocurrency, and new hardware are interrelated and reinforce each other. Can you discuss how they collaborate specifically?

Chris Dixon: Certainly. Using mobile internet, social networks, and cloud computing as examples, they have achieved success together. The same is true for AI, cryptocurrency, and new hardware today. You can already see some signs, such as AR/VR glasses and self-driving cars heavily utilizing AI technology, with companies like Tesla also making strides in humanoid robots. These technologies are bringing AI into the real world and opening up new application scenarios.

On the cryptocurrency side, I am particularly optimistic about a field called DePIN (Decentralized Physical Infrastructure). For example, the Helium project is a community-owned crowdsourced telecom network that challenges the traditional models of Verizon and AT&T. Users install Helium nodes (wireless transmitters) in their homes to contribute coverage to the network. There are now hundreds of thousands of nodes across the U.S., providing services much cheaper than traditional operators (e.g., $20/month vs. $70/month). This is feasible because it utilizes cryptocurrency to design an incentive mechanism, avoiding the billions in network construction costs faced by traditional operators.

The most challenging part of network construction is the startup phase because early network effects are weak—like dating sites, 10 users are useless, while 1 million users are valuable. Cryptocurrency solves this problem through token incentives, rewarding early participants and thus driving network expansion. The DePIN concept is not limited to telecommunications; it extends to climate modeling, map data, electric vehicle charging, and more. For instance, we recently invested in a decentralized energy monitoring network, and others are using similar methods for decentralized science. The early construction of such networks is naturally suited for cryptocurrency, while AI can complement it in data collection and processing.

Host: The stage of technological development is also crucial. How do you view the evolution of AI?

Chris Dixon: I like to analyze technological development using a framework divided into three stages: the first is "old things done new" (skeuomorphic), where new technology improves existing things; the second is "new things" (native), where things that were previously impossible are created; and the third is "second-order effects," which are profound changes triggered by the widespread adoption of technology.

For the internet, the 1990s were the first stage, where people moved magazines and directories online. Amazon selling books was more convenient than flipping through magazines, but it was still an old thing in a new form. In the 2000s, social networks emerged, which were truly native applications with no offline counterparts, and the business models were also entirely new. AI is similar. The first stage is the current common "old things done new," such as replacing call centers with AI customer service, which is both cheap and efficient, potentially affecting tens of millions of jobs but also creating more new opportunities. This stage may last for 20 years.

The second stage is the "native" stage, which truly excites me. For example, after photography technology became widespread, art shifted towards abstraction (like Cubism) and simultaneously gave rise to film as a new art form. Today's generative AI is similar; some believe it threatens creativity, but I don't see it that way. It may be the cornerstone of a new art form, like virtual worlds, new types of games or films, or perhaps entirely new interfaces. These innovations require genius creative talent to realize, often in unexpected ways. Just as film opened up new horizons, AI may bring about similar breakthroughs.

The third stage involves "second-order effects." After the rise of social networks, Obama's 2008 victory utilizing them marked a turning point, followed by phenomena like the Trump movement and populism, all of which are second-order effects that continue to evolve today. The second-order effects of AI may not fully manifest until 20-30 years later, with each stage potentially lasting a decade.

Host: What are the limiting factors in the transition from the first stage to the second stage?

Chris Dixon: The early internet was limited by the physical construction of networks, such as laying cables. The limitations of AI are different; technological capability is no longer the main bottleneck, but rather human creativity and policy regulations. The supply side needs creative talent to develop native applications, and the current entrepreneurial ecosystem is much more mature than it was 15 years ago—venture capital firms have grown from dozens to thousands, startup advice is of higher quality, and smart people find it easier to enter this field, with abundant capital and energy.

But the demand side is more challenging. Changes in organizational and individual behavior take time. For example, I want to use AI to read my book and mimic my voice, but publishers and Audible completely prohibit AI due to unions and traditional views. Hollywood may take a generation to accept AI-native films, perhaps driven by emerging AI startups in developing countries. The policy aspect is more complex; regulated industries like copyright, healthcare, and finance, which account for 70% of the economy, will face intense debate. Is AI training data "copying" or "learning"? This may ultimately be resolved by congressional legislation rather than the free market or court rulings.

Host: What does your ideal future internet look like?

Chris Dixon: We are at a crossroads. The original vision of the internet was community ownership and governance, with profits flowing to small businesses, innovators, and entrepreneurs at the network's edge. But today, wealth and power are concentrated in the hands of a few large companies, with the five major tech giants accounting for more than half of the market value. The first sentence in my book is "architecture determines destiny," and control and the flow of funds depend on how services are designed.

I worry that we are approaching an irreversible tipping point where the internet is monopolized by five companies. They have saturated user growth and are now starting to "kick away the ladder," hindering newcomers. This poses a significant threat to "little tech." If startups have to pay hefty "taxes" to compete with the giants, they cannot challenge the status quo. We have seen similar cases in the past, such as Zynga, which relied on Facebook and ultimately fell victim to platform risks.

Therefore, supporting new architectures (like blockchain) and open-source AI is crucial. This is not just a technical issue but also a regulatory and public perception issue. We need policies that encourage competition and innovation to avoid consuming the "seeds" of the future. Through A16Z's efforts, I am optimistic that the idea of "little tech" is spreading, and more people are becoming aware of the importance of new infrastructure and open source.

Host: Thank you, Chris. It was great to talk with you.

Chris Dixon: Thank you for the invitation; I enjoyed the conversation!

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