AI × Web3: Who Will Lead the Era of Blockchain Transformation?

When the technological paradigm truly shifts, we often see the hype first, rather than the system. The wave of AI we are currently experiencing is no different.

As a primary investor, I have always believed that betting on the transformative forces deep within the industry is far more valuable than chasing superficial narratives.

In the past year, I have seen a large number of projects such as RWA, Consumer, and infoFi - they are undoubtedly exploring the intersection of the real world and on-chain systems.

But the increasingly obvious trend is that regardless of the path a project takes, it must ultimately enter the collaborative logic of AI, using AI to enhance competitiveness and efficiency.

For example, RWA, thinking about how to optimize risk control with AI, off-chain data verification, and dynamic pricing is the future;

Alternatively, Consumers or DeFi that urgently need excellent user experiences also require AI to complete user behavior prediction, strategy generation, and incentive distribution. I won't elaborate further on similar applications in other sectors.

Therefore, whether it is asset digitization or experience optimization, these seemingly independent narratives will ultimately converge to the same technological logic: if the infrastructure does not possess the integration and carrying capacity of AI, it will not be able to support the complex collaboration of the next generation of applications.

In my opinion, the future of AI is not just about becoming "stronger and more widely used"; the real paradigm shift lies in the reconstruction of collaborative logic.

Just like the early transformation of the internet, it wasn't because we invented DNS or browsers, but because it allowed everyone to participate in content creation for the first time, turning ideas into products and giving rise to an entire open ecosystem.

AI is also on this path: Agents will become everyone's intelligent co-creation entity, helping you turn expertise, creativity, and tasks into automated productivity tools, and even monetization.

This is a question that is very difficult to answer in today's Web2 world, and it is also some of the underlying logic I see in the AI + Web3 track: making AI collaborative, transferable, and profit-sharing is the system that is truly worth building.

What I want to talk about today is the only project so far that attempts to systematically build the underlying operations of AI from a chain-level structure: Sahara.

The essence of investment is the worldview, recognizing the value system of choices.

My investment logic is not just the narrative of public chains plus AI, then seeing which team seems to have a better background and placing a bet.

Investment, in essence, is a choice of worldview, and I have always been asking a core question: Can the future of AI be owned by more people together?

Can it leverage blockchain to reconstruct the value attribution and distribution logic of AI, allowing ordinary users, developers, and other roles to participate, contribute, and continuously benefit? It's simple, I only believe that such projects can become disruptors if this logic emerges, rather than being just "abandoned public chain +1".

In order to find the answer, I basically scanned all the AI projects I could access until I encountered Sahara. The answer given to me by Sahara's co-founder Tyler was: to build an open, participatory ecosystem that everyone can own and benefit from.

This sentence is simple, but it precisely hits the soft spot of traditional public chains: they often serve developers in a one-sided manner, and the design of token economics is mostly limited to Gas Fees or governance, rarely able to truly support a positive cycle of the ecosystem, and even more difficult to sustain the development of an emerging track.

I am well aware that this road is full of challenges, but precisely because of this, it is an irresistible revolution - which is also the reason for my firm investment.

As I emphasized in my previous article discussing the "Evolution from Web2 to Web3": the true paradigm shift lies not in creating a single product, but in building a supportive system.

And Sahara was one of the most anticipated cases in my assessment at that time.

From Investment to 8x Valuation Follow-up Heavy Investment

If I initially invested in Sahara, it was because it was doing exactly what I believe to be the true mission of AI – building the AI economy and infrastructure system. What drove me to chase after investing at an eightfold pre-round valuation in just six months is the rare strength I felt in this team.

Two co-founders, one is the youngest tenured professor at USC, specialized in AI. I believe the value of a tenured professor in their 90s is not only in the academic field, but also in the fact that at this age, they still have dreams, energy, and the determination to realize those dreams. Knowing Professor Ren for over a year has shown me what it means to work for more than ten hours a day, to be emotionally stable, and to be a humble genius.

Tyler, an investment director at a certain platform, is responsible for North American investments and incubators, and his understanding of web3 goes without saying. He is astonishingly disciplined: he only sleeps in multiples of 1.5 hours, insists on working out regardless of how busy he is to maintain his state, and doesn't touch a single piece of candy for a clear mind, working over 13 hours a day. I once jokingly said he was a robot, and he simply replied, "I am lucky to have this busyness today." His source of dopamine comes from making progress on projects every day; dreaming is his passion, and he doesn't need any other fuel.

I am very grateful to have met them, which changed me. I have finally started to sleep regularly as much as possible, my emotions are gradually stabilizing, and I'm working out...

So when someone says that Sahara gained the favor of capital because of luck, I always candidly add, "The pursuit of capital is an inevitable result." I vividly remember how difficult primary financing was in this market cycle, yet Sahara was being pursued by investors in the primary market.

What everyone remembers is that a certain investment institution invested in Sahara. Sahara has opened the investment era for Samsung's entry into the Web3 AI field, and its winning of the Samsung AI Award is an important reason for Samsung's investment. In addition, some AI-heavy funds, national banks, and so on are also guests of Sahara. What you can see is a group of institutions that are more focused on traditional technology and industrial resources beginning to quietly bet on AI × Web3 because of Sahara.

Capital will only pay for certainty in direction and execution - this is the positive feedback regarding the depth of Sahara technology, team background, system design, and execution capability.

This is also why it can produce some real and solid structural indicators:

Over 3.2 million accounts have been activated on the test network, with more than 200,000 data platform annotators and millions queued (. Their clients include several leading enterprises, and they have already achieved revenue in the tens of millions of dollars.

On this infrastructure chain, at least from "who will do it" to "can it be done", Sahara has already gone deeper and steadier than 99% of the "AI Narrative projects."

![AI × Web3: Who will build the chain for this era?])https://img-cdn.gateio.im/webp-social/moments-cf477a069888f7e6e1a56aa0d7248930.webp(

The Ultimate Challenge of Public Chains: Ensuring Continuous Benefits for All Contributors and Driving Positive Economic Cycles

Returning to our initial judgment logic: in a system where AI and blockchain are combined, is there really a mechanism that allows every contributor to be seen, recorded, and continuously rewarded?

Model training and data optimization rely heavily on a large amount of labeled data and interaction support; conversely, if there is a lack of user contributions, the project itself has to invest more funds to purchase data and outsource labeling, which not only increases costs but also weakens the value-driven aspect of community co-construction.

Sahara is one of the few Web3 AI projects that allows ordinary users to "participate in data construction from day one." Its data labeling task system operates daily, with a large number of community users actively participating in labeling and prompt creation. This not only helps improve the system but also invests in the future with data.

Through the mechanism of Sahara, it not only improves the quality of the model, but also allows more people to understand and participate in this decentralized AI ecosystem, linking data contribution with benefits, and forming a true virtuous cycle.

A typical example is the Myshell project on a certain blockchain, which has rapidly built a high-quality dataset covering multiple languages and accents through Sahara's decentralized data collection and human-machine collaborative annotation, significantly improving the training efficiency of its TTS and voice cloning models. This has also propelled its open-source projects VoiceClone and MeloTTS to gain thousands of GitHub stars and over 2 million downloads on Hugging Face.

At the same time, users participating in data annotation also received token rewards distributed by Myshell, forming a two-way incentive loop between developers and data contributors.

Sahara's "permissionless copyright" mechanism ensures the open circulation and reuse of AI assets while protecting the rights of all participants—this is precisely the underlying logic driving the explosive growth of the entire ecosystem.

Why is it said that this is a scenario with long-term value support?

Imagine that if you want to build an AI application, you naturally hope your model is more accurate and closer to real users than others.

The key advantage of Sahara is that it connects you with a vast and active data network—hundreds of thousands, and in the future millions, of annotators. They can continuously provide you with customized, high-quality data services, allowing your models to iterate faster.

More importantly, this is by no means a one-time transaction. Through Sahara, you are connecting to a potential early user community; and these contributors are likely to be the true users of your product in the future.

This connection is not a one-time buyout; through Sahara's smart contract system and rights confirmation mechanism, it enables a long-term, traceable, and sustainable incentive system.

Regardless of how many times the data is called, contributors will receive continuous profit sharing, and the earnings are dynamically linked to usage behavior.

But this is not just a revenue model for data labeling and model training stages. Sahara builds an economic system that covers the entire lifecycle of AI models. In every aspect after the model goes live—such as calls, combinations, and cross-chain reuse—there is also a built-in profit-sharing mechanism, allowing value to be captured over a longer period.

Model developers, optimizers, validators, computing power contribution nodes, etc. can now continuously benefit at different stages, rather than just relying on a one-time transaction or buyout.

Such a system brings a compound effect for model combination calls and cross-chain reuse. A trained model, like building blocks, can be repeatedly called and combined by different applications, with each call generating new revenue for the original contributors.

For this reason, I agree with Sahara's underlying belief: a truly healthy AI economic system cannot just be the plundering of data and the buyout of models, nor can it merely allow a few people to reap the benefits. It must be open, collaborative, and win-win—where everyone can participate, and every valuable contribution can be recorded and continuously rewarded in the future.

![AI × Web3: Who will build the chain for this era?])https://img-cdn.gateio.im/webp-social/moments-f100e78f0ea1234e20c00aebaa95457a.webp(

But the closer we get to the real structure, the more challenges there are.

Although I am optimistic about Sahara, I will not obscure the challenges the project will face due to my investment stance.

One of the major advantages of the Sahara architecture is that it is not limited to any specific chain or single ecosystem.

Its system was designed from the beginning to be open, full-chain, and standardized: supporting deployment on any EVM-compatible chain, while also providing a standard API interface, allowing Web2 systems—whether e-commerce backends, enterprise SaaS, or mobile apps—to directly call Sahara's model services and complete on-chain settlements.

However, despite the extreme scarcity of this architectural design, it also poses a core risk: the value of the infrastructure lies not in "what it can do," but in "who is willing to do what based on it."

To become a trusted, adopted, and integrated AI protocol layer, the key for Sahara lies in how the ecological participants evaluate its technological maturity, stability, and future predictability. Although the system itself has been built, whether it can truly attract a large number of projects to land based on its standards remains uncertain.

It is undeniable that Sahara has achieved key validation: providing services to several leading enterprises, offering them relevant data services, and addressing some of the industry's most challenging data demand issues, which serves as an early signal of the feasibility of this system.

But what needs to be seen is that these collaborations mainly come from the Web2 world, which truly determines Sahara.