"GateLive Roundtable" Episode 9: From Chat to Agent: How Will the Crypto World Welcome the "New Subject" in the Age of AI Surge?

“Gate Live Roundtable Discussion” is a Chinese-language crypto roundtable series created by Gate Live, airing promptly every Wednesday at 8:00 PM, focusing on the most discussed industry topics of the moment. The show periodically invites core practitioners and frontline observers from blockchain, Web3, DeFi, Ethereum ecosystem, stablecoins, compliance, and policy fields for in-depth exchanges in a live setting.

The roundtable emphasizes a relaxed, open, and authentic dialogue atmosphere, exploring market trends, industry disagreements, and key variables from multiple perspectives to help viewers form clearer, more rational judgments amid complex market narratives.

This episode’s theme: From Chat to Agent: How the Crypto World Will Embrace the “New Subject” Under the AI Surge?

Guests: Well-known Chinese crypto community KOLs — Crypto.0824, Mark Ruo Mcrowe, Mr. Misi

This program is for informational exchange and opinion discussion only and does not constitute any investment advice.

(This content is compiled from the live replay, with text assisted and appropriately edited by AI. For the full content, please copy the link: https://www.gate.com/zh/live/video/e4cac1c21003342c2f28f968b8244385)

Host Jesse:

Hello everyone, good evening. Welcome to tonight’s GateLive roundtable. I am your host, Jesse!

Recently, whether on Twitter or in communities, I believe everyone has felt that—the rapid rise of AI seems to have entered another dimension.

In the past, when we talked about AI, it was mainly ChatGPT—“chat” and “generating images.” But with the explosive popularity of OpenClaw, we see the wind has completely shifted.

The first time we saw these so-called “AI agents,” they were no longer passive tools that respond only when asked a question. Instead, they became autonomous actors capable of accessing emails, managing calendars, touching permissions, and even directly interacting with accounts and funds.

From “tools” to “subjects.” AI not only thinks but also “acts,” even “trades” on our behalf. When this happens in the crypto world—an industry centered on “decentralization” and “digital assets”—are we ready?

Actually, this integration has already begun. Just a few days ago, we at Gate took a significant step—launching products like Gate for AI and Gate Claw. I strongly recommend everyone try them out. I used the “Little Lobster” (Xiaolongxia) deployment via Gate Claw—very smooth.

Back to the core question: in the face of this aggressive “new subject” era, how should we understand, respond to, and participate in it?

Today, we are very fortunate to invite three experts with deep understanding and practical experience in this field:

Crypto.0824, Mark Ruo Mcrowe, Mr. Misi

Welcome, everyone! Please briefly introduce yourselves.

Crypto.0824:

Alright, glad to join tonight’s roundtable to discuss the AI boom triggered by OpenClaw. I’m 0824, usually involved in contract trading on Gate Exchange. If interested, we can exchange ideas.

Mark Ruo Mcrowe:

Good evening, I’m Mark Ruo Mcrowe. Happy to be here at Gate’s roundtable. I’ve been trading for 5 years and enjoy using Gate Exchange for trading. Recently, OpenClaw has become very popular with many new features. I’ve tried some myself and look forward to sharing tonight. Thanks.

Mr. Misi:

Hello everyone, I’m Misi. I mainly focus on on-chain research and “grinding” (collecting alpha). If you’re interested in early alpha projects, follow me. I’m glad to discuss OpenClaw with you all. I’ve been researching it recently and have some insights to share.

Host Jesse:

Thanks again to all three for being here. Let’s get straight into today’s topic.

Many friends might still be confused: what exactly is “OpenClaw” igniting?

I’d like to ask each of you, in the simplest terms, to help our viewers understand the essence of this “autonomous AI agent.”

And when you’ve experienced or researched these autonomous AI agents, have you ever encountered moments that made your hair stand on end or seem unbelievable? Any failures? Surprises? I’m very eager to hear your firsthand experiences.

Crypto.0824:

Alright, to understand this wave of enthusiasm, we first need to dispel a common misconception—that people might think AI agents are just smarter search boxes. But that’s not the core.

The core of autonomous AI agents is shifting from passive answering to active execution.

Before, we asked it a question, and it responded. That’s a chatbot—only responsible for chatting. But now, give it a goal, and it will break down tasks, call tools, and perform actions. This is what we call an AI Agent. That’s the experience OpenClaw brings.

OpenClaw is an open-source AI agent framework. Its popularity comes from making autonomous action accessible and practical, allowing everyone to get involved. It’s not just in Web3 circles talking about “lobsters” (Xiaolongxia), but also among mainland users—this is very hot.

Building an autonomous AI that can operate traditionally required huge engineering effort and resources. Now, OpenClaw significantly lowers the barrier, enabling more people to quickly build AI systems that can actually do things for you. The most shocking part I experienced is that you can just give it a vague goal—like “monitor large transfers on ETH today, and alert me if there’s abnormal activity.” It will automatically decompose the task, query data sources, set thresholds, and trigger alerts. You only need to wait for the result. The whole process is hands-off—you just get a notification on your phone, and then check the results. That’s a completely different experience from using traditional tools.

On that note, I want to mention that Gate has launched two AI products: one is Gate Claw, based on OpenClaw, and the other is Gate for AI, a comprehensive AI infrastructure. I’ve tried both; Gate Claw is very smart. If you’re interested, give it a shot.

Mark Ruo Mcrowe:

In my view, the biggest difference between this autonomous AI and traditional AI is that it’s more like a human. How so? Because traditional AI agents can only think—they can’t act. For example, GPT can chat with you on any topic, help you generate content, and give you a relatively concrete output or process. But it can’t execute actions.

OpenClaw, on the other hand, can automatically perform tasks—like opening emails or browsing web pages—by directly controlling your computer. That’s the biggest leap: it can help you do things directly.

Previously, AI had ideas but couldn’t act. Think of it as having a brain but being paralyzed—only half-abled. Now, OpenClaw can connect to all your interfaces and execute what it thinks out. It’s like having a wise person who can act.

For example, I recently used OpenClaw to monitor social media buzz for a project. It quickly fetched the info because I connected its API to Telegram. When it detects a spike, it sends me a message. Then I can react immediately. It even asked me, “I see the buzz, should I place an order?” Its speed and ability to act are much faster than previous AI.

Some KOLs have shared their OpenClaw use cases—like trading, monitoring “golden dogs” (pump-and-dump schemes). It’s gaining popularity because it directly helps make money, which makes it more relatable.

Recently, Gate released three AI products, and I’ve tried them. Gate AI has become more powerful, especially tailored for crypto. I also tested Gate Claw, which now offers free usage tiers—very smart. Gate for AI is aimed at crypto trading infrastructure, with all interfaces open for easy deployment.

Overall, this is a revolutionary change. Its earning and execution capabilities are much stronger than before. That’s why OpenClaw is so hot.

Mr. Misi:

Before answering, everyone should know that when ChatGPT first appeared, crypto folks were among the earliest users. Everyone thought it was powerful—because previous AI tools I used for content creation were often garbage, even producing unreadable text. But ChatGPT changed the game—writing, image generation, and more.

Now, with OpenClaw, it’s clear that AI is gradually achieving autonomous interaction with reality.

To intuitively understand “autonomous AI agent,” I think it’s more than just a tool; it’s like a real assistant.

Earlier AI like GPT or DeepSeek was mostly “ask and answer.” But autonomous AI agents are different—they’re given a goal, then they plan steps, execute tasks, and keep optimizing results.

From this perspective, why is OpenClaw so popular? I believe it’s not just about the technology itself but about a realization: AI is shifting from “answering questions” to “helping you do things.” This is a crucial transformation.

In the past, we saw posts about hiring assistants to monitor markets or manage contracts. But with OpenClaw, tasks that required human effort can now be done by AI.

In Web3, information density is high, and the pace is fast—new projects, narratives, opportunities emerge daily. Relying solely on humans to filter, research, and track is very time-consuming. When I try autonomous AI agents, I feel their efficiency is greatly amplified.

For example, researching a new project used to take hours—reading whitepapers, browsing Twitter, analyzing on-chain data. Now, an AI agent can do all that simultaneously, providing a summarized view. It might not be perfect every time, but it’s very fast at filtering and initial screening.

I also use AI to gather big news events, which sometimes gives unexpected insights—like correlating projects or market sentiment. It’s like having a constant assistant watching the market for you.

I think current autonomous AI is still early-stage; it can’t fully replace human judgment yet, but it significantly amplifies individual capabilities. For those of us doing content, research, and info filtering in Web3, it’s already a powerful helper.

So, the reason OpenClaw is so hot now is because people are finally realizing—AI can do more than just answer questions; it can participate in real workflows.

Host Jesse:

After hearing everyone’s insights, it’s clear that AI’s evolution isn’t just a technical upgrade but the birth of a new “species.”

This leads to our second core topic: when AI is no longer just a tool but an independent “subject,” what does that mean for crypto?

Especially considering products like Gate for AI, when exchanges become an “action layer” accessible directly by AI, how will the game change?

Will it reshape industry rules?

Crypto.0824:

That’s a very key perspective shift. Previously, we discussed AI × crypto mainly as “AI helping decision-making.” But now, AI is transforming into a subject—holding permissions, accessing accounts, touching funds.

This means: future on-chain participants won’t be just humans but AI Agents. They will trade independently, manage positions, and even participate in DeFi protocols autonomously.

First, at the trading level: AI Agents can monitor markets 24/7, respond in milliseconds, achieving high-frequency, precise actions beyond human capability. Arbitrage windows will shrink, market efficiency will improve. Retail traders without AI support will be at a disadvantage in speed and information.

Second, at the infrastructure level: traditional APIs are designed for human developers. AI agents need standardized, semantic interfaces. Whoever becomes the trusted on-chain entry point for AI will dominate the next-generation infrastructure.

Third, at the responsibility and accountability level: this is the biggest unresolved issue. If an AI makes a mistake in trading, who bears the responsibility? Developers? Users? Protocols? This will push the industry to rethink compliance, account systems, and on-chain identity mechanisms.

Here, I must mention Gate’s AI products—these address the three dimensions. Gate for AI is the world’s first unified AI trading gateway, integrating CEX, DEX, wallets, and on-chain data into a standardized interface system. When integrated with ChatGPT, Claude, Manus, etc., AI can perform the entire crypto trading process—from market data, account management, to executing orders—within a unified architecture. This infrastructure is laying the groundwork for AI to become a true “subject” in crypto automation.

Mark Ruo Mcrowe:

I think the exciting part is that AI can directly access more funds and permissions—via various APIs.

While AI can do many things, the current hype and practical use focus on its ability to help you make money. It can connect to your WeChat, Feishu, Telegram, etc.

Compared to before, it’s more integrated and real-time. Previously, AI agents could call APIs but couldn’t receive real-time info, give feedback, or act immediately. Now, it’s a disruptive change.

For example, if you give OpenClaw access to your wallet, it can automatically top up or move funds. Just give it your seed phrase, and it can operate independently—very convenient. It’s like an autonomous economic entity. That’s a huge leap. Earlier AI agents didn’t have this capability.

Second, the logic itself is changing. As Mr. Misi said, previous AI agents often lacked full execution ability. OpenClaw can follow your predefined logic to complete entire workflows—covering data collection, decision-making, and execution—minimizing complexity for users. No need to learn coding; it automates the entire process, lowering the entry barrier.

Third, for industry practitioners or professionals, this could mean a reshuffle. Someone who understands the logic and workflow can set up comprehensive automation. As long as the program runs and tokens are sufficient, it can operate 24/7, potentially replacing certain roles. That’s the core advantage.

Mr. Misi:

When we say “AI is becoming a new subject,” the key point is the role change.

Previously, in crypto, decisions and actions were made by humans. AI was just an assistant—analyzing data, automating tasks. Now, AI can act based on goals, plan steps, and execute independently. That’s a major shift.

From industry perspective, it could change two things: speed of information and decision-making. Crypto is a fast-paced, information-asymmetry market. Opportunities often appear in short windows. Humans can’t monitor everything 24/7, but AI can.

Second, it could alter participant structures. Previously, complex strategies or large-scale arbitrage required teams or institutions. With mature AI agents, individuals could have powerful “digital assistants” managing multiple strategies, wallets, and research workflows—something hard to do before.

In the long run, I believe AI as a “subject” in blockchain aligns naturally with the infrastructure—wallets, smart contracts, on-chain identities—designed for automation. We might see AI managing wallets, participating in protocols, making decisions, and collaborating across DeFi. It’s a vivid future to imagine.

Of course, we’re still exploring. But I think AI’s emergence will gradually transform the crypto industry from a “human-led” market into a “human + AI” economy. The way we compete, collaborate, and design projects will change significantly.

Host Jesse:

That’s a lot to digest. The future of AI sounds exciting but also complex. Thanks to all three for your brilliant insights. From the explosive growth of OpenClaw to the evolution of crypto infrastructure, we see clearly: this is an era of rapid technological change. For crypto, it’s both a rule-restructuring and a huge opportunity.

Thanks again to everyone watching and to our guests. If you’re interested in Gate’s AI products, feel free to try them out after the show and experience what it’s like to have AI trade for you.

See you next time!