MEXC Product Director Vivien Lin said centralized exchanges are moving closer to the structure of traditional brokerages, while AI could become the next major entry point for trading.
In the interview, Lin discussed the overlap between traditional finance and crypto, AI-assisted trading, global regulation, and where exchanges may compete next. She said the relationship between TradFi and crypto has shifted from a conceptual discussion to actual integration of assets, capital flows, and user behavior. On MEXC, trading in TradFi products is still driven mainly by existing crypto users, she said, and during softer periods for the crypto market, TradFi products at one point made up about 30% to 40% of trading volume on the platform.
TradFi and crypto are moving from narrative overlap to actual asset and capital connectivity
Lin said the past four to five years have already brought several product cycles to crypto. From 2020 to 2024, TradFi and crypto kept trying to connect, but the two sides never fully met. Native crypto users stayed largely inside a relatively closed ecosystem and kept rotating through new narratives, which is why the market saw phases centered on altcoins, memes, NFTs, and DeFi Summer.
She said a more meaningful shift started from the second half of last year into this year. Earlier discussion about TradFi-crypto convergence was mostly about information flow and shared ideas. Now, with products such as xStocks and other forms of tokenized equity, crypto users have started trading instruments tied to traditional stocks. Lin said MEXC’s RealStock expanded that universe further. In her description, tokenization models used in the past could only cover a limited number of stocks, while direct connections to traditional brokerages can open the broader U.S. equity market to platform users.
That does more than widen the asset pool, she said. It also starts to change the user mix. Crypto users have typically traded BTC, ETH, Solana, and meme coins, often favoring higher volatility and higher leverage. Once traditional financial assets appear inside crypto platforms, traditional finance traders can also enter the same ecosystem and bring different trading habits, risk preferences, and product expectations. Lin said this makes the current year very different for crypto exchanges, because the market now presents a real chance to scale with TradFi-facing users.
Asked whether current demand on MEXC comes mainly from existing crypto users trading TradFi products or from new TradFi users coming onto the platform, Lin said the first group remains the larger one for now. Still, she added that the boundary is becoming harder to define. She used her own trading behavior as an example: she previously used fiat accounts for TradFi and crypto accounts for digital assets. Once the two sides are connected, she said, users no longer need to convert crypto back into fiat and move funds into a separate brokerage system. They can switch exposure inside MEXC directly.
Lin said new TradFi users may not yet be the main growth driver, but she sees at least a tenfold opportunity in that segment. One reason, in her view, is that crypto exchanges already offer trading tools that are not yet widely available in traditional brokerages. She pointed to grid trading. In crypto, she said, it is already a simple and familiar product. In traditional markets, by contrast, ordinary brokerage users often cannot access comparable tools unless they know how to code or connect through APIs. She said some of her friends with TradFi backgrounds found that trading traditional assets on crypto exchanges gave them access to strategy tools they had wanted for a long time but could not easily get through conventional financial platforms.
The two sides are moving toward the middle, but they do not build trust the same way
Lin described the long-term direction as a “meet in the middle” outcome. Even so, she said TradFi users and crypto-native users do not approach trust from the same starting point.
According to Lin, the first concern for many TradFi users entering a crypto platform is whether the product is real in the legal and financial sense. When they see offerings such as xStocks or RealStock, they want to know whether the instrument is truly linked to the underlying stock, and whether they are buying only a token representation or an asset backed by actual shares.
She said TradFi users have long placed heavy weight on compliance and the sturdiness of legal relationships, while crypto tends to lean more on technology, mechanisms, and community trust. Because the foundations are different, many TradFi users who have wanted crypto exposure still hesitate to enter fully.
Lin said this is why MEXC chose a heavier model with RealStock. The goal, she said, is to let users obtain actual ownership of the corresponding stock, along with rights that may matter later, including dividends, voting, and other corporate actions. In her view, if users can receive dividends and take part in governance, they are more likely to believe they are holding a real asset. She framed that as a key part of how MEXC tries to build trust with TradFi users.
At the same time, she said, traditional financial platforms are also moving toward crypto. She cited Robinhood’s introduction of crypto-native products such as BTC perpetuals. Even so, she argued that traditional platforms may not understand crypto retail traders as deeply as exchanges that were built inside the crypto market. Traditional financial institutions have historically served institutional or institution-like clients, she said, and often develop first around assets before extending into lending and trading services. Crypto-native platforms such as MEXC took the opposite route, starting with retail users and trading scenarios, then expanding into lending, MEXC Card, AI-driven information products, and intelligent trading tools.
That split in development path matters, she said, because it creates different capability boundaries and different moats. For MEXC, Lin said the main long-term pillars are staying close to user demand, reducing friction through low or even zero fees, and using AI to deliver more personalized services across different user tiers.
Global regulation and local user differences are making a single global product stack harder to maintain
Lin said a number of trading needs are universal: low slippage, tight spreads, low fees, and deep liquidity. Those are common almost everywhere. But outside pure execution, demand starts to diverge sharply from region to region. Products around savings, cards, campaigns, and user rewards can vary a lot, she said, and users in Southeast Asia, Europe, and Japan may prefer different assets, promotional formats, and levels of trading complexity.
Regulation is becoming an even more important variable. Lin said global oversight is tightening, and MEXC is actively working with local regulatory requirements while building country- or region-specific sites. Exchanges now have to understand not only local users, but also what the rules in each market actually allow.
She said the central difficulty is that no platform can realistically rebuild a fully separate site and full service stack for every jurisdiction. MEXC’s answer, she said, is to turn the underlying service architecture into modular, atomic components.
She used order-entry functions as an example. Limit orders, market orders, iceberg orders, trailing take-profit and stop-loss settings can each be handled as separate modules. When MEXC enters a new compliant market, it can assemble those building blocks according to the local rules on asset types, leverage, and other product restrictions.
The same logic applies to operations and promotions, she said. Coupons, prize wheels, trading competitions, and other campaign formats can be modularized. If one market responds well to a wheel-based campaign, the team can deploy that quickly. If another prefers trading competitions, they can switch to that setup. Lin described the product system as something like Lego, where interchangeable pieces can be combined differently for each market.
She also said users differ widely across regions in terms of value, trade size, and leverage preference, and that each exchange will see a somewhat different user mix depending not only on local culture and geography but also on its own growth strategy. MEXC, she said, tends to be more retail-friendly, and its zero-fee approach is especially attractive to price-sensitive users. That naturally pulls in more retail users, which then feeds back into product decisions.
For new markets, Lin said MEXC also uses AI to help internal operations and business development teams understand local users. The platform has accumulated a large amount of campaign and user data over time, and can combine that with smaller early-stage datasets from a new market to analyze user maturity, preferred assets, and operational formats that may work best there.
AI now touches information, search, strategy generation, and internal workflows
Lin said one of MEXC’s internal product concepts over the past year has been what she called “smoothness,” meaning the distance between a user forming an intention and actually finishing an action. Better product design used to mean cleaner page transitions, clearer buttons, and more readable information, she said. Later, the focus shifted to how many clicks and pages a user had to go through. With AI, she said, that distance can shrink again.
She gave search as one example. In the past, typing BTC into the global search box might only show spot and futures entry points. Now, she said, the same search can surface price charts, the most important BTC-related news from the past day, platform campaigns connected to BTC, AI-generated strategy tools, and wealth-management products. A single keyword can lead users to assets, market information, trading functions, and related product links across the platform. Lin said the challenge is to keep information rich without creating overload, so MEXC pays close attention to density and presentation.
She divided current AI use cases into two broad groups: internal tools and user-facing tools.
Internally, MEXC encourages staff to use AI, but Lin said the emphasis is not on casual question-and-answer prompts. The focus is on turning repeated tasks into reusable AI skills. Product managers can use AI to draft PRDs and analyze product performance, engineers can use it for coding, and operations teams can use it to generate campaign ideas. The company also builds local models and internal skills to reduce the cost of calling large external models directly.
Lin said MEXC runs internal AI application competitions as well, encouraging employees to automate repetitive daily tasks. If one employee builds a useful skill, that tool can be shared with other teams and improved over time, gradually creating an internally driven workflow of iteration.
On cost, she said the company is not simply restricting AI usage. Instead, it is trying to optimize token consumption through local models, intermediate processing layers, and harness engineering. She used customer support as an example. Although support teams face many questions, much of the content is repetitive. Rather than having every support agent call a very large model directly, MEXC can extract common high-frequency issues and handle them with smaller local models at lower cost.
On the user side, some of the current AI tools are information products, including AI News, Trending Token, and MaxAI Bot. Lin said MaxAI Bot resembles a ChatGPT-style interface, but with a tighter focus on trading-related information and with chart data, technical indicators, and market signals built into the response flow to increase information density.
Another layer is AI-powered search, which is mainly designed to improve the path from intention to action. Beyond that is AI Strategy, a product Lin said she personally likes a great deal. She said the team studies well-known historical traders and investors, as well as classic strategies that can be tracked in the market, then backtests them and adapts them for crypto-specific market structure.
She said traditional financial strategies cannot simply be copied into crypto because the market structure and trading rules differ. Parameters have to be recalibrated to produce something usable in digital asset markets.
MEXC has also opened its AI quantitative strategy tools to users for free and supports natural-language strategy generation, according to Lin. She gave an example: a user can type a rule such as buying when MACD shows a particular reversal and selling when it reverses again. The system then expands that plain-language instruction into a more professional trading command and links it to APIs and execution systems. Users can edit parameters themselves or accept suggested settings, then run backtests directly. If they are satisfied with the results, they can move on to execution. Lin said this sharply lowers the barrier for users who have trading experience but do not know how to code and run systematic strategies on their own.
Lin expects AI to become a trading interface, though MEXC is not letting it decide for users yet
Asked whether AI will eventually become a real transaction interface, Lin said yes, and that she does not think the shift is far away.
Her reasoning was simple. Once users get used to a highly efficient product, she said, it is hard to go back to a less efficient one. She added that she is already used to AI assisting with analysis and decision support across many tasks. If that layer disappeared, she said, the absence would feel obvious.
Lin described one possible future flow: a user wakes up in the morning and tells AI, “I think BTC looks good today. Help me check whether there’s an opportunity.” The AI then runs strategies in the background and looks for options. If it sees a setup, it returns an execution suggestion. If something remains uncertain, it discusses that with the user.
In that framework, she said, the human role may shift more toward defining risk appetite, investment philosophy, and market understanding. A risk-averse user could instruct AI not to use aggressive strategies. The core exchange between human and machine may center less on exact order steps and more on preference and risk recognition.
Even so, Lin stressed that MEXC has not reached the point of allowing AI to make decisions on behalf of users. For now, the platform is automating information analysis, interpretation, and part of the strategy-generation process, then presenting those outputs in a more accessible form. She said trust still has to be built between users and AI. Users need to understand how AI works before they are willing to hand it more meaningful trading tasks. If that trust develops, AI could become the true gateway between users and the trading world.
She added that MEXC is also planning products for bots, because future interaction between bots and trading systems may look very different from the way human users interact with apps today.
As AI agents connect to exchanges, the boundary of responsibility may need to be redrawn
Lin said the question of liability becomes more complicated if users connect self-built AI agents to an exchange and something goes wrong. In her view, this comes down to the boundary between rights and responsibilities.
She said AI remains in an early, highly experimental stage, and the exchange is primarily providing the interface into the underlying trading system. If the platform’s own execution system is mature, the probability of a platform-side system failure should be relatively low under normal conditions. But the exchange cannot control what kind of bot a user ultimately connects. If a user-built bot creates an error during the integration process, that does not automatically mean the exchange is at fault.
Even so, Lin said the platform should still try to reduce misunderstanding and disputes as much as possible. Because AI trading is still early, she said, everyone involved has to protect the user experience carefully while the product category matures. That is one reason MEXC would rather do more itself at this stage and channel more user demand into relatively standardized official tools. In her view, that allows the platform to stand behind its own products while reducing friction and risk created by third-party tools.
Regulation is pushing exchanges toward modular, localized product design
As crypto rules become clearer across jurisdictions, Lin said the most direct consequence is that exchanges can no longer serve the world with one universal product package.
Every time an exchange enters a new regulated market, it has to understand local requirements first, she said, and those requirements vary widely. Some jurisdictions allow higher leverage, while others do not. Some permit only spot trading. Others set explicit limits on token listings or product categories.
That is why modularity remains central, in her view. Lin said exchanges need to split configurable product capabilities into independent components. If the global-site setup works in a given market, that is ideal. If it does not, the platform has to reassemble a product mix that fits local regulation and local user demand as quickly as possible. She described the product team’s job as building the infrastructure and “arsenal” in advance so operations, marketing, and business teams can move into different markets with more confidence.
Asked what it would mean if a DEX such as Hyperliquid were to gain greater recognition from U.S. regulators, Lin said that would clearly be positive for the industry. She said crypto has long sat near the edge of the traditional financial system, while policy changes in the United States over the past few years have already accelerated the convergence of crypto and TradFi.
From a product perspective, she said the issue also raises a broader question: what kind of compliance do regulators actually want? Hyperliquid remains controversial on topics such as KYC, she noted, but the platform emphasizes on-chain transparency, transparent mechanics, transparent community governance, and data that can be tracked and is hard to alter. If that model ultimately gains more regulatory acceptance, Lin said, it would amount in some measure to recognition of crypto-native values such as decentralization, transparency, and traceability.
As assets become more alike, competition shifts to fees, shipping speed, user experience, and licenses
Lin said asset offerings across platforms are likely to continue converging. That is true not only among crypto exchanges, she said, but also between traditional financial platforms and crypto platforms. Over time, the range of assets available on each side may become increasingly similar.
At that point, she said, the real question becomes why a user chooses platform A instead of platform B.
For MEXC, Lin identified several areas that she sees as core moats. The first is fees. Zero-fee trading has strong appeal for retail traders, new users, and high-frequency participants, she said. The larger the volume, the more meaningful cumulative fees become, so pricing structure remains a major advantage.
The second is development speed. Lin said MEXC can move relatively quickly when new opportunities appear, but only because product and engineering teams prepare ahead of time rather than starting from scratch after something becomes popular.
She returned to RealStock as an example. Many platforms offer stock exposure through tokenization, she said, while MEXC chose direct brokerage connectivity. The reason is tied to authenticity. If the target user base includes TradFi users, the first questions are whether they truly own the asset, whether they can receive dividends, and whether they can vote. Lin said that route takes longer to build, but once finished it also creates a higher barrier to entry.
The third area is AI. Lin said MEXC was relatively early in releasing AI-assisted tools. Because the platform’s user base is primarily retail, it did not start by building CLI-style tools just for professional users. Instead, it integrated AI directly into the app and the page experience so ordinary users could feel the benefit immediately.
Over the long run, she said, fee structure, development capability, product experience, technical planning, and the licenses and distribution channels a platform can obtain and use will all combine to form the exchange moat.
Lin says the staying power of DEXs may be underestimated
Asked what the market may be underestimating over the next three to five years, Lin pointed first to the durability of decentralized exchanges and the future relationship between DEXs and CEXs.
She said that two years ago, the market often asked whether DEXs would take all of the market share from centralized exchanges. That question comes up less now because the DEX sector is currently in a weaker phase overall, with only a handful of platforms still maintaining notable users and volume. But she does not believe DEXs will disappear.
Lin said DEXs and CEXs serve different user groups and therefore create different mechanisms for price discovery. The gap between them can naturally create arbitrage, new issuance patterns, and other trading opportunities. More importantly, as CEXs become more compliant, the room they have for high-risk product experimentation may shrink. DEXs operate in a freer environment, she said, and their users are often more crypto-native and more accepting of risk, which could leave DEXs carrying a larger share of product and asset innovation.
She gave the example of new derivative structures. Launching those structures directly on a large CEX with many ordinary users may be too risky, but the same ideas may find a better testing ground in a DEX environment. For that reason, Lin said DEXs will not die out. New platforms will still emerge, though their form may look very different from Hyperliquid today. She added that as CEXs start to look more and more like traditional brokerages, they may also leave fresh product space for DEXs.
The end state may be a single platform where users trade both traditional assets and crypto
On the longer-term relationship between TradFi and crypto, Lin said the broad direction is still convergence. But she does not think the final result is simply that two separate user groups merge. Instead, she said, people may gradually use the same product environment: one platform where they can trade traditional assets and crypto side by side, while capital moves between the two seamlessly.
She compared that path to the historical relationship between equities and commodities. At one point, she said, those markets were also separate, and people debated when they would merge. Once the integration happened, the debate itself faded. In her view, TradFi and crypto may follow a similar arc. As users and assets blend, the market may stop treating them as two disconnected systems.
Lin said she does not yet have a firm answer on the exact role DEXs will play in that integration process, but she is watching the question closely. As CEXs continue to take on brokerage-like characteristics, she said, DEXs may end up preserving more of crypto’s native capacity for innovation.
TradFi products once reached 30% to 40% of platform volume, and capital is already moving across markets
Asked about the share of trading volume on MEXC represented by TradFi or RWA-related products compared with traditional crypto products, Lin said the ratio changes with market conditions. When the crypto market was relatively weak for a period, TradFi volume accounted for around 30% to 40% of platform trading. When crypto activity recovered, that share declined.
What matters more than the ratio itself, she said, is that the platform can now clearly see capital moving across the TradFi-crypto boundary.
Lin said she had often discussed with friends whether traditional finance users and crypto users are really separate groups. In practice, she said, many people were already active in both markets. The real barrier was not the person, but the account structure: TradFi accounts and crypto accounts existed in different systems, so the money itself was not truly connected. Once crypto platforms added TradFi products, that wall started to break down.
MEXC can now see funds moving from TradFi products back into crypto and from crypto into traditional financial assets, she said. That is why she thinks the opportunity is not only about taking a wholly new user base away from traditional platforms such as Robinhood or Futu. Many of those users already exist in both markets. The key is to uncover both sides of their demand inside the same product system.
She said this new capital connectivity may also create entirely new product opportunities. If there are price gaps between stock-linked products on crypto platforms and prices on traditional brokerages, a platform could potentially build tools that help ordinary users capture those cross-market spreads. Institutions can write code and build arbitrage strategies on their own, she said, but most retail users cannot. If those tools can be productized, ordinary users may be able to take part in cross-market arbitrage as well.
For Lin, the real significance of TradFi-crypto integration is not just that more assets are available. It is that new trading forms and product structures will continue to emerge once assets and capital are truly connected.
Lin also said the guest comments in the interview do not constitute investment advice and that users should comply strictly with local laws and regulations.