Source: Wanxiang Blockchain
From April 20th to 23rd, the 2026 Hong Kong Web3 Carnival, co-hosted by Wanxiang Blockchain Labs and HashKey Group, was successfully held at the Hong Kong Convention and Exhibition Centre. On the stage of the venue, Jerrina Yu, Marketing Director of FutureMoney Group, moderated a roundtable discussion on the theme “The ‘Golden Closed Loop’ of DePIN: When Machines Create Assets, Stablecoins Complete Payments.” The following is a summary of the discussion, with some omissions that do not affect the original meaning.
Host Jerrina Yu: Good afternoon, everyone. I’m Jerrina Yu from FutureMoney Group, and I’m the moderator for this roundtable discussion. First, please let each of our guests introduce themselves, starting with Shukyee.
Shukyee Ma: Hello everyone, I’m Shukyee Ma, Chief Strategy Officer at Plume. Plume is dedicated to providing a compliance layer for RWA, supporting various assets, including tokenized assets we currently collaborate with Apollo, WisdomTree, Pimco, Invesco, and others. We have numerous distribution channels in the Asia-Pacific region, and in the US, we hold various licenses to serve our asset issuers. That’s all for now.
Peter Chen: Hello everyone, my name is Peter, and I’m primarily responsible for Tether’s institutional business development in the Asia-Pacific region. My clients mainly come from traditional industries, including banks, commodity trading companies, and supply chain companies. We hope to leverage increasingly transparent regulation to move stablecoins from their original on-chain state to real-world applications and make them a reality. Thank you.
Steven Wang: Hello everyone, I’m Steven Wang from Goldwarts Limited. What we’re doing now is building a middleware financial application protocol layer based on smart hardware devices and settlement layers like wallets. We hope to seize the opportunities presented by the future development of the machine economy and embrace this economic transformation. Thank you.
Will One: Hello everyone, I’m Will from the Lao National Digital Technology Group. We are a digital technology company with an exclusive partnership with the Lao government. We are exclusively authorized to handle all related stablecoin licenses, national supply chains for digital technology and blockchain, RWA, digital identity, etc., and also hold a controlling stake in a third-party payment license, etc. Thank you.
Host Jerrina Yu: Thank you, everyone. Next, we will ask three general questions, and we hope each guest will answer them individually. After that, we will have individual questions in a one-on-one format. Let’s start with the general questions.
The first question concerns ecosystem synergy. Stablecoins serve as a crucial bridge connecting the digital and real economies. Looking ahead, what kind of synergistic effects or operating models do you anticipate between the stablecoin ecosystem and real-world machine networks?
Shukyee Ma: I think firstly, from the perspective of the stablecoin ecosystem, we can see that different stablecoin issuers are exploring different distribution methods. For example, Circle obtained a license in Bermuda. We can see that stablecoins themselves are shifting from being merely a means of payment to becoming a type of interest-bearing asset. Regarding machine-generated rewards, there are already many DePIN scenarios, such as Helium, where machines generate stablecoin rewards. I think the more important question is how we can find a more suitable way for the development of compliant stablecoins within a compliant framework, and also how to pass on the rewards to holders through machine-generated rewards.
Peter Chen: From the issuer’s perspective, we have a vision for stablecoins to become the de facto settlement layer for machine payments, smart agent payments, and robo-payments. This is not only because stablecoins are relatively stable units of account, but also because of the programmability of blockchain, 24/7/365 transactions, and borderless nature. These characteristics are perfectly suited for environments like those involving machines. For example, when a smart agent needs computing power for reasoning, it can pay a very small amount, i.e., pay-as-you-go. This environment and application scenario will gradually expand as the number of machines and smart agents increases, eventually covering the entire industry. This gives us a lot of room to promote stablecoins.
If you look at Elon Musk’s statement from an interview earlier this year, he explicitly said that the number of humanoid robots in the future should exceed the total number of humans. Therefore, we can foresee that in the relatively distant future (I’m not sure if it’s 5 or 10 years), when this scenario unfolds, we can imagine that every robot might need a crypto wallet, and the settlement layer for this wallet would likely be stablecoins. This is also the purpose of our goal in promoting this technology across various application scenarios and communities. Thank you.
Steven Wang: Actually, this is how I see it. We’ve been talking about DePIN’s golden closed loop, which I believe should have three layers. The first layer is the asset layer, where any machine that generates valuable data—for example, a DePIN device—may need to tokenize the data it generates. This usually requires infrastructure like Plume. Another layer is the settlement layer, like the digital yuan, which allows valuable assets to flow freely on the blockchain. However, there’s still one missing layer: the financial protocol application layer. For example, how should my payment be authorized, who should authorize it, or are there other application rules involved? This layer is currently underdeveloped.
Overall, the current collaboration among these three layers is still quite rudimentary and basic. Why is that?
Because many devices generate valuable data, the process of converting this data into valuable information currently involves uploading it to centralized platforms for monetization. The digital yuan itself hasn’t yet been widely adopted in the overall business logic of machines. Therefore, we’ve identified this opportunity. Perhaps in one to two years, any machine, including the intelligent agent behind it, will possess a decentralized identity and a blockchain wallet, as Peter mentioned, to handle the digital yuan. Ultimately, this will allow it to bypass all centralized platforms and subscription models, enabling direct settlement and value flow between devices according to machine-readable rules.
I believe that Hong Kong, for example, will be a very important node. Why? Because Hong Kong already has the policy framework and reserves for RWA tokenization, the issuance of compliant digital currencies, and even cross-border data asset transfers. We believe that in the next two to three years, once the machine economy achieves scalable development, Hong Kong will definitely become such a settlement center. That’s my view.
Will One: The previous three speakers have already shared a wealth of ideas, covering everything from compliance and micro-payments to the interaction between technology and machines. I’d like to envision the future and add some insights into the investments and opportunities we’ve seen in the industry, which are largely in the form of “flow payments”—payments between intelligent agents and between machines. Regarding giving machines decentralized digital identities, we are also planning that. When assisting countries like Laos, given their limited populations, we want to open up these opportunities to future silicon-based life forms, or even further into the future, machines, addressing everything from closed-loop identity verification and compliance to value creation, capture, and transfer between machines, as well as data security and privacy. That’s my perspective, thank you.
Host Jerrina Yu: Thank you to all the guests for sharing their perspectives on the ecosystem and operating models. We know that for the machine economy to fully realize its potential, stablecoins and seamless payment connections between machines are needed. So, could you all discuss how stablecoins can bridge the gap between the complex mechanisms of blockchain and large-scale adoption in the real world?
Shukyee Ma: I think we first need to clearly define the role of stablecoins in mass adoption. Stablecoins offer greater programmability and reduce cross-border friction, among other benefits. However, from a mass adoption perspective, user experience is more important—how the stablecoins in a wallet are used and in what scenarios. Therefore, I believe stablecoins can be a channel to facilitate mass adoption, but having stablecoins as a tool doesn’t guarantee mass adoption. We need to build upon the stablecoin ecosystem. Then we can see that AI agents, smart agent payments, and even yield optimizers, using AI as a tool, can reduce transaction effort and time on the blockchain through stablecoins. In this way, they can facilitate faster communication between smart agents. I think stablecoins are ultimately a tool; how AI uses them is just one example.
In addition, as we mentioned before, wallets are a crucial platform for making payments, investments, and various settlements using stablecoins. Stablecoins have begun to open up this channel, but how to use subsequent tools is also a very important topic.
Furthermore, I think an even more important point is that as the types of stablecoins increase and their use cases expand—including allocating user funds and helping users find assets and returns on stablecoins—all settlements on these platforms should be subject to a very comprehensive compliance and regulatory framework; that is, they must occur on a relatively compliant platform. I think this is something Plume has been working on.
Peter Chen: Regarding this issue, I think there are several aspects worth discussing. Especially the issues Shukyee mentioned earlier concerning regulation, taxation, or accounting standards, these are gradually maturing. We can see that many jurisdictions globally are beginning to provide clearer directions for legislation. However, this isn’t my main focus, although it’s very important because it’s fundamental. Additionally, I’d like to echo Shukyee’s point about user experience. Many of you here are familiar with blockchain, stablecoins, or the crypto industry. But for the general public, achieving large-scale adoption requires starting with user experience, abstracting out the relatively difficult technical hurdles to enable rapid adoption by many users. For example, wallet user experience, gas fee integration, and key management are sensitive and security concerns that many non-crypto-native groups around me are very afraid to encounter. Therefore, if these technologies can be abstracted and made into a user-friendly form, it will actually greatly promote the industry and end-users.
Secondly, as a settlement tool, stablecoins need to be fast on suitable networks and capable of settling small amounts. With current technology, it’s possible to settle transactions as small as $0.0001. This enables fast, high-frequency settlements in many scenarios. Achieving these capabilities would significantly shorten the gap between the general public and widespread adoption. I’d also like to mention that Tether launched Tether Wallet last week, essentially based on this idea—to abstract away these complexities and make them more readily accepted by the industry. Our CEO has also clearly stated that this is a product designed for billions of people and trillions of intelligent agents and machines. Its importance will grow exponentially in the future machine economy or robotics economy. We will also provide development kits to allow partners to easily integrate and plug-and-play these capabilities directly into their applications. We hope to leverage these partners to continuously advance towards an ecosystem of large-scale machine growth. Thank you.
Steven Wang: Regarding mass adoption, I’d like to share my own perspective. A few years ago, when we were working on the mass adoption of compliant digital currencies, I believed our approach was somewhat flawed. Why? Because we were educating customers about blockchain, digital wallets, and gas fees—this was extremely costly. So when will true mass adoption occur? It won’t happen through educating customers, but rather because it offers them practical value. For example, why did compliant digital currencies first see mass adoption in many remote, underdeveloped countries? It’s because the fiat currencies in these countries might not be strong enough, and their financial infrastructure might not be adequate, so local people would prefer to use compliant digital currencies as a store of value and a payment tool. That’s why we’re seeing this kind of mass adoption.
Returning to DePin, I believe mass adoption will occur after the machine economy truly reaches scalability. Why? Traditionally, when you buy a smart device, it’s a consumer product. While it generates valuable data (such as usage habits, health data, and family data), this data remains controlled by the platform. Users who generate data are constantly providing it without receiving any value in return. However, with the advent of the new era, users can earn economic rewards from each use of the device. The narrative shifts from “I’m just using a device” to “This device is making money for me,” generating daily rewards. The entire experience will be different. Furthermore, device manufacturers will shift from a one-time purchase model to a long-term revenue-sharing model, as the entry point moves from the platform to each individual device. Therefore, platform providers and hardware manufacturers will be more willing to manufacture more and better devices, encouraging longer-term user engagement and providing greater long-term revenue sharing—this is the change in the business model. I believe this is the process that will lead to mass adoption.
So why might this happen in Hong Kong or the Greater China region? Because Hong Kong, as a financial center, is backed by Shenzhen and the entire Greater China region’s strong industrial chain. Therefore, as the machine economy develops, all these entry points and payment gateways will shift to hardware devices, and large-scale adoption will naturally occur. That’s my point of view.
Will One: I think the previous speakers all spoke very well, and now it’s my turn to summarize. I’d like to take this opportunity to share my thoughts. Based on what I’ve heard and combined with my previous points, I have roughly three points. First is compliance and KYC, which is the most pressing issue. Second is user-friendliness, although this user group may be divided into carbon-based and silicon-based users. From a carbon-based perspective, we’ve extracted three abstractions: account abstraction, chain abstraction (don’t worry about which chain), and, from a sovereign nation’s perspective, currency abstraction (don’t worry about which stablecoin you’re using). For most carbon-based users, making good use of these is sufficient. As for silicon-based users, large-scale adoption still has a long way to go, and the underlying protocols need further upgrades. Regarding data, I believe trust is even more crucial because data belongs to everyone, not just to some other entity. Therefore, this may require deeper privacy computing technologies, compliance regulations, and some economic distribution systems. This leads to the third point, which is definitely economically driven, and I hope everyone will discuss this aspect together.
Host Jerrina Yu: Everyone gave excellent answers to this question. Actually, Peter and Steven have already raised my next question. The machine economy currently has these characteristics: high frequency, microtransactions, and low cost. So, my question is: is the current stablecoin infrastructure truly ready to support the machine economy globally?
Shukyee Ma: First, from a proof-of-concept (PoC) perspective, if I issue a stablecoin on a specific blockchain and only use it for settlement in a single scenario, regardless of whether it’s machine-to-machine or machine-economic, it can do it well. Because on any mainstream blockchain, such as Polygon, Solana, or Ethereum L2, you can find a good way to settle it regarding gas fees and finality. However, when it comes to real-world applications, starting with stablecoins themselves, there’s a fragmentation problem—we have many different types of mainstream stablecoins; there are several different types right here.
Furthermore, regarding cross-chain functionality, whether stablecoins can truly achieve chain independence is questionable, and cross-chain interactions will inevitably encounter friction. More importantly, different countries have different regulatory frameworks. Hong Kong has its own regulations, while the US, Bermuda, the UAE, and even Laos have their own distinct regulatory frameworks. I believe the first priority should be enabling stablecoin issuers to establish on-chain liquidity, making it smoother. Only then can we return to discussing application scenarios, which would be more reasonable. However, from a technical perspective, including the stability, gas fees, and finality across various chains, I believe the infrastructure is ready.
However, this is the pattern in every industry: technology comes first, followed by various proof-of-concept experiments, and we are also conducting various experiments. Later, when communicating with regulators or policymakers, we will have different use cases, and then we will formally enter the real growth phase within the compliance framework. Therefore, I believe this has enormous potential.
Peter Chen: For the first time, I’m at a loss for words because, frankly, this is exactly what I wanted to say. But we all know that the individual technical modules are, in fact, ready. Whether it’s sub-second settlement, fast block generation technology, or tens of thousands of transactions per second, the individual technologies are ready. And you need to be able to settle very small amounts, including the $0.0001 or even less I just mentioned. We know that in traditional payment channels, anything less than $0.10 is not economically viable for settlement because the cost of traditional payment channels may exceed that. Blockchain can handle these small payments. However, there’s a significant gap between the readiness of individual technical modules and truly coordinated, unified, large-scale adoption, and seamless mobility. We know that many tech giants, including Google, Coinbase, Stripe, etc., have invested a lot of effort in this area, and Tether and its portfolio companies have also spent a lot of time. This is the current reality.
Therefore, the most important part is not technological readiness, but rather how to create a large-scale ecosystem negotiation process to get everyone to agree on the same standard and adopt it. This is the only way to achieve large-scale adoption within a reasonable timeframe. That’s my opinion, thank you.
Steven Wang: Actually, the two previous speakers have already addressed this question quite extensively. I agree with their point that after so many years of development, blockchain technology is completely different from the previous impression of being slow in block generation and taking a long time. Modern high-performance public chains can even rival the most advanced financial transaction matching systems. In this process, it is more than sufficient for payment settlement needs that require sub-second or near-real-time processing.
However, returning to what the guest mentioned earlier, the real bottleneck lies at the application protocol layer. This includes two parts: First, if we’re referring to machines or dePIN devices, they lack identity. This is actually the first and biggest problem. Just like people in underdeveloped regions who lack KYC or identification documents and cannot be accepted by the traditional financial system, machines are the same; they themselves lack identity, and therefore require some legal entity mapping behind them. Second, there are other application obstacles. M2M Pay creates such a protocol layer, integrating various financial application protocols, including compliance mapping, payment authorization, pricing, and valuation rules, making them machine-readable, thus lowering the barrier at this middle layer. This way, you don’t need to worry about whether you’re paying in Hong Kong, the Middle East, or another country’s system. Ultimately, our protocol layer can achieve cross-regional and cross-protocol abstraction. I believe this is a crucial part of stablecoin infrastructure meeting the needs of the machine economy’s expansion.
Will One: Okay, I’ve already exhausted all my “words,” so all I can do is laugh. I think those of you who have been involved in both Web2 and Web3 for a long time should understand this. In Web3 native development, 90% or more is proof-of-concept. Even the few high-quality projects that have been painstakingly selected, including Tether and Plume, are currently in the proof-of-concept stage in terms of overall business returns. This actually highlights our position. We hope to invite partners from various fields to quickly build these promising and potentially valuable platforms together. By creating a large alliance and consolidation, I believe this momentum and approach will, in turn, accelerate the development of other major countries. This could be a form of indirect acceleration, and I extend a sincere invitation!
Host Jerrina Yu: Thank you all for your insightful presentations. However, due to time constraints, we will now ask each guest a question. Let’s start with Shukyee. Shukyee, we know you are the Chief Strategy Officer of Plume, so I would like to ask you to share where Plume stands in our discussion today, and in which areas it serves the most.
Shukyee Ma: Today we’re discussing two major players in the gold-chain closed loop. One is stablecoins, which some of you are already working on. So what’s missing? As a compliant settlement layer, what are we settling? First, how can we reduce the fragmentation and friction between different stablecoins on our chain or platform? I believe Plume, with its embrace of traditional finance, compliance, and the composability capabilities of decentralized finance, can achieve this. Second, how can we transform these assets, including stablecoins, into a tool for generating returns, that is, defining investable scenarios and assets? Although we started with yield-generating financial products, all layers and tools are ready, regardless of future developments in the machine economy, or more financial products and other asset classes. As a settlement layer, the most important thing is how to create more composability of different assets within a compliant framework, enriching everyone’s investment and settlement scenarios. I believe Plume is very suitable as the underlying platform.
Jerrina Yu: Thank you, Shukyee. Now I’d like to ask Steven a question. Steven, I’ve also heard you share some thoughts about your product. So, regarding machine-to-machine, which sounds like the next generation of the internet economy, our next evolution, many use cases aren’t quite clear yet. What do you think is the next inevitable use case that you’re very optimistic about, one that can’t be replaced by human intervention? In other words, what are the essential conditions for using machine-to-machine and stablecoins?
Steven Wang: This is an excellent question, and one I’d like to elaborate on. The logic behind machine-based or smart agent-based payments we’re discussing, currently, whether in traditional or encrypted payment systems, still addresses the issue from the user’s perspective. Essentially, it still requires user authorization. This is no different from Alipay’s small-amount, password-free payments on traditional platforms. So, what scenarios truly improve and enhance traditional payment methods, and in what situations is machine-to-machine payment absolutely necessary? This involves logic based on direct invocation by smart agents or smart hardware, and pay-as-you-go pricing.
Specifically, when an intelligent agent collects data from certain smart hardware devices and transforms it into valuable information, it connects this information with a merchant’s products or services through a protocol or network. This process involves numerous API calls, data searches, and even guiding users to complete tasks. First, this process cannot wait for direct user authorization; second, the value generated is very small, like receiving a coin for clicking an ad. Traditional payment methods such as Visa, Swift, or others are incompatible with this scenario. Only under these conditions, and after this technology is deployed on a large number of machines and achieves economies of scale, will machine-to-machine payments, including payments completed using blockchain, become an inevitable and irreplaceable method. Therefore, our M2M Pay is working on this layer. We are waiting for the later developments, as the usage of intelligent agents increases and smart hardware becomes more intelligent—not just with simple rule commands, but with more intelligent and discerning logic applied on a large scale—to inevitably involve using compliant digital currencies and blockchain for final settlement.
Host Jerrina Yu: Thank you, Steven. And I’d like to ask Will, do you think the scenarios Steven just described are feasible in developing countries like Laos? Or is it currently impossible due to regulatory reasons?
Will One: I think, especially in countries like Laos, they are very well prepared. Treating this as a rigid, overly strict process is tantamount to rejecting the opportunity. I believe the emergence of new technologies is also to help markets full of opportunities that haven’t been seen before. There are already many pioneering cases, including Plume, with whom we are reaching agreements; they are also offering compliance-as-a-service. We can learn a lot from their experience in compliance matters. We can refine and optimize these experiences to create locally tailored solutions. Of course, we also sincerely invite stablecoins from other countries, such as Tether, because we believe that the investment and development opportunities brought by introducing currencies backed by other countries’ sovereign credit represent a good window for collaborative development.
Regarding the local machine economy, we are already exploring it, including the large-scale deployment of solar panels. Laos has better-than-average sunshine conditions, with daylight hours twice as long as some parts of China, which has attracted many leading players. These machine-to-machine benefits will further attract partners around the world who have already conducted various technology verifications. This is our overall view. Of course, Steven has repeatedly mentioned the issue of data, which many jurisdictions consider. However, some developing sovereign countries may not have prioritized this issue, but we have considered it in advance when we are doing this. For example, we will adopt technologies that balance privacy and compliance, such as TEE or zero-knowledge proofs, each with its own focus. We will also combine other technologies and learn from other experiences to balance these aspects. Thank you.
Host Jerrina Yu: Due to time constraints, that’s all the questions we can ask today. Thank you to all the guests for sharing; it has been very inspiring. We look forward to seeing you next time!
