China’s New DuClaw AI DESTROYS OpenClaw?

China's new Duclaw AI destroys OpenClaw. So China just solved the biggest problem with OpenClaw. Not a security patch, not a new model. The biggest problem, the one thing stopping a billion people from using it.

Set up. Baidu just launched a product called Duclaw. It's a zero deployment service. You don't install anything.

You don't touch a terminal. You don't configure anything. You just pay $2.50 and your AI agent is live and working. $2.50 a month.

That is less than a bottle of water at the airport. For a fully managed AI agent that runs your inbox, books your travel, manages your calendar, and handles tasks whilst you sleep. That's Duclaw. I want you to remember that name because I think it's one of the most important product launches in AI this year.

Not because of the technology inside it, but because of what it represents. Duclaw is the moment AI agents stopped being a technical project and became a consumer product. Now, let me show you why this matters so much. OpenQuant grew by 925% from February to March.

The second highest growing AI product in the same period grew by 390%. The third grew by 72%. 925% month over month. Nothing else in AI is that close.

And yet, for most of that growth, the tool was still too hard for regular people to use. You needed to type commands into a black terminal window. You needed to navigate developer platforms. You needed to configure models and set permissions and troubleshoot errors.

And most people looked at that and walked away from OpenQuant for that reason. Duclaw is Baidu's answer to that problem, and it's a very good answer. Duclaw includes built-in access to Baidu Search, Baidu Bank, Baidu Scholar. It supports model switching across DeepSeek, Kimi, GLM5, and Minimax.

And the service is fully managed on Baidu's cloud infrastructure. Think about what that actually means. When you open Duclaw, your AI agent already knows how to search a web, already knows how to look things up. It can switch between different AI brains, depending on what tasks you need done.

And you didn't set any of that up. It was already there when you logged in for $2.50 a month. And that's not a small thing. That's the iPhone moment for AI agents, the point where the complexity disappears behind a product that just works.

And the company doing this has 700 million monthly active users on their main app. 700 million people just got handed a door directly into the agentic AI world. Not an invitation, a door already open, already cheap, and already set up for them. Now to understand why Duclaw landed so hard, you need to understand what's happening in China before it launched.

Over the past month, major Chinese cloud providers launched their own versions of OpenClaw. Local governments dangled grants to startups building OpenClaw apps. And a cottage industry sprung up helping users install the open source framework. China's users are now calling it raising lobster, a phrase that refers to OpenClaw's red lobster logo.

Raising a lobster, that's the phrase. And it's not a niche joke across tech workers, it's on merchandise. It's a vocabulary of retirees and students and finance managers who have never written a line of code. Of the more than 142,000 publicly visible AI agents on OpenClaw track globally, nearly half originated from China.

Nearly half of all visible OpenClaw agents in the world are running in China right now. The tool was built by an Austrian developer. It went viral in Silicon Valley first, and China is running away with it. OpenClaw has amassed over 250,000 GitHub stars in just a few months, making it the most starred project on the entire platform.

Chinese media frame this as surpassing Linus' all-star time count, a milestone that took Linux over 30 years to reach. 30 years. Linux took 30 years to get there. OpenClaw did it in a few months.

That's not a trend line. That is a detonation. But here's the thing, with all of that energy, with all that demand, with all the people who wanted in, most of them still couldn't understand or get it because the setup was too hard. The gap between wanting OpenClaw and actually using OpenClaw was a technical skills gap that most people couldn't cross on their own, and that gap created a whole new economy.

On the morning of March 6th, Tencent's cloud engineers set up free installation booths outside their Shenzhen headquarters. Students stood alongside retired engineers. Someone had flown in from Hangzhou the night before. Meanwhile, across the city, sellers on Taobao were charging up to 500 yuan per remote install, with the busiest stores logging over 1,000 orders.

A retired engineer and a nine-year-old kid standing in the same line to get an AI agent installed on their laptop. Someone flying in from just another city to get access. 1,000 orders on a second-hand marketplace just for someone to help you install free software. A software engineer named Feng posted an ad that read, no need to know coding or complex terms.

Fully remote, anyone can quickly own an AI assistant available within 30 minutes. And that ad worked because the need was enormous and the solution was scarce. Feng was making real money charging people to install a free tool. This is what a setup barrier looks like when it meets mass demand.

Duclor just made Feng's business obsolete. That's the significance of what Baidu did. They did not just make OpenClor cheaper, they destroyed the reason the whole cartridge industry existed. When you can get a fully managed OpenClor setup for $250 a month from one of the biggest tech companies in the world, you don't need to pay a stranger $72 to install on your laptop.

The barrier is gone, the friction is gone, and the reason most people are not using it is gone. And Duclor is not alone. Two days after Baidu launched it, Alibaba fired back. Alibaba launched JVS Clor, a dedicated mobile app for iOS and Android.

No coding knowledge required. Install and deploy OpenClor in minutes, free for the first 14 days. And now it's on your phone. The thing that is already in your pocket, the thing that your grandmother uses to video call you, AI agents are on that device now, ready to go.

Tencent launched WorkBuddy, an AI agent for workplace tasks, fully compatible with OpenClor. ByteDance rolled out ArtClor, a version that runs directly in a web browser with no local setup at all. Minimax, Zed AI, and SenseTime all launched their own integrated products within days, and every major Chinese tech company is now in this race. And every product they launched is solving the same problem that Duclor solved first.

Remove the setup, remove the friction, make it so simple that anyone can have an AI agent by lunchtime. Duclor set the template, and everyone else is now following it. JD.com launched a page where users can pay $58 to get remote help from Lenovo's IT team to deploy OpenClor. Meituan announced a similar partnership.

Even the food delivery and e-commerce giants got in as well. They looked at what Duclor is doing and said, we can sell that too. Set up as a service, onboarding as a product, that's how far Duclor's core insight traveled in just days. The insight that the real product is not the AI, the real product is getting people into the AI.

Now, let me tell you what Duclor and this whole wave did to the stock market. Minimax has seen its shares rise 640% since listing just two months ago. It's now worth about $49 billion US dollars, surpassing Baidu in market value despite Baidu generating $18.5 billion in revenue in 2025 compared to Minimax's 79 million. The market is not buying the current numbers, it's buying the trajectory.

It's saying, if Duclor and products like it make AI agents mainstream at scale, the companies at the center of that ecosystem are going to be enormous. Minimax powers one of the models inside Duclor. That connection alone helped turn a $79 million revenue company into a $49 billion market cap story. Tencent shares rose 7.3% in a single day after launching Workbody.

That was Tencent's best single day in over a year. And one product announcement built on the same model, Duclor pioneered. One day, billions added to Tencent's market value. The market price is, and the market is pricing in the Duclor effect across every company that's moving in the same direction.

Now, let me talk about the human side of this because behind all the stock numbers and the product launches, there are real people feeling real pressure. A product manager named Cindy Wang works at OpenCore's Chinese, one of China's largest finance groups. Before Lunar New Year, her company had just adopted OpenCore and managers were demanding employees prove competitively that they could use it. She said, since OpenCore came along, things have gotten insanely competitive.

We are practically being crushed by the grind. Managers warned staff they would be immediately replaced if they didn't use AI. That's not a headline. That is a real person under real pressure.

Before Duclor, you could at least argue that the tool was too complicated for most people to use. You could say, I want to learn this, but the setup is too hard. Duclor just removed that excuse. 250 a month, no setup, no technical knowledge.

If you're not using an AI agent now, it's not because it's too hard, it's because you haven't started yet. That's a very different situation. And companies are starting to notice. This is where I want to be real with you for a second because the gap between the people using these tools and the people watching these stories is getting wider every single week.

Duclor is significant, not just because it's a good product, it's significant because it removed the last real excuse for not starting. Inside the AI Profit Boardroom, we talk about exactly this, not just headlines, the actual tools, the actual workflows, how to use products like Duclor to build systems that run whilst you sleep, how to take $2.50 a month and turn it into hours of reclaimed time and real competitive advantage. If you want to be in that room, the link is inside the comments description or go to the AIProfitBoardroom.com. Come and join us.

Now, here's the wrinkle in the Duclor story because this is not a clean win. There's actually a darker thread running through all of this. The same week Baidu launched Duclor, the Chinese government told its BC institutions not to touch OpenClor at all. Government agencies and state-owned enterprises, including the largest banks, received notices warning them against installing OpenClor software on office devices.

Some employees were instructed to notify superiors if they'd already installed the apps and submit to security checks and possible removal. The ban was extended to personal phones and using company networks, and in some cases to the families of military personnel. So Duclor makes OpenClor effortless for consumers on Tuesday. And the government bans it from banks and agencies on Wednesday.

Same week, same tool, two completely opposite official responses happening simultaneously. That's not confusion, that's a government that understands both the opportunity and the danger at the same time. And the security concerns are legitimate. Duclor's managed approach actually does address some of them.

When Baidu hosts your OpenClor instance on their infrastructure, you're not running a vulnerable local installation. You're not one of the 42,665 OpenClor instances that anyone could walk into without a password. Baidu is responsible for patching vulnerabilities and managing your security configuration. Security researchers found critical flaws in the OpenClor that allow full gateway compromise and authentication token theft.

One vulnerability was rated 8.8 out of 10 on the cybersecurity severity scale. Over 800 confirmed malicious add-ons have been found inside OpenClor's own plugin registry. Duclor's cloud managed model means Baidu controls which plugins you can access and how your agent is configured. You trade some flexibility for a lot more safety.

For most regular users, that's a very good trade. For government agencies handling sensitive national security data, a managed consumer product is still not going to cut it, hence the ban, hence the contradiction. Kendra Schaefer, a technology policy researcher at Trivium China put it plainly, Chinese regulators typically respond with extraordinary speed to threats from emerging technologies. The rate of adoption of OpenClor and other agent tools is still outpacing them, still outpacing the fastest regulators in the world.

And Duclor is a big reason why. Every time a new managed product drops the barrier further, millions more people come in faster than any regulatory framework can keep up with. Now, let me zoom out and connect all of this to a bigger picture. Duclor is not just a product, it's a signal.

It's Baidu saying the era of OpenClor being only for technical people is over, it's done. We're building the infrastructure to make this work for everyone. And once Baidu said that, every other major company had to respond. JVS Clor from Alibaba, RClor from ByteDance, WorkBuddy from Tencent.

Each of these products exists because Duclor forced the question, if Baidu can put OpenClor in 250 subscription, what's your excuse for not doing the same? That's how innovation waves work. One product to make something possible. Every competitor has to match it or fall behind.

And the users win because the price drops and the ease increases every single time someone new enters the race. China-based usage of OpenClor has already topped that of US whilst driving demand for Chinese lower cost AR models. And the country that invented this technology is being out-adopted by the country that moved faster to deploy it. And Duclor is the central reason why.

Not the only reason, but the product that showed what was possible. The 250 subscription that told 700 million people, you can have this now, right today, without any help. Chinese models consumed 61% of global AI processing tokens in late February, powered by OpenClor demand. Kimi K215, one of the models inside Duclor's model switching feature, and more in 20 days than Moonshot's entire 2025 revenue, the company behind KimiClor.

And that last sentence is wild. One model in 20 days, more revenue than the entire previous year. Because Duclor and products like it are funneling millions of users through these models every single day. The infrastructure that Duclor built is generating that kind of economic activity downstream.

Now, let me bring this home, because I want to make sure this lands in a way that's useful for you. The Duclor story is a story about what happens when the last big barrier to a technology disappears. Set up was a barrier for OpenClor, 250 a month and a fully managed service just removed it. When barriers disappear, adoption goes vertical.

We are watching that happening real time in China, and it's coming to the rest of the world fast. If you're a business owner, Duclor is the proof that AI agents are no longer a technical project. They're a consumer product. Your competitors can now have a fully functioning AI agent for the cost of a cheap coffee subscription.

Ask yourself honestly, are they already using it and you're not? If you're a freelancer or creator, Duclor and products like it just handed you the same productivity leverage that used to require a whole operations team. The question is whether you pick it up. If you're an employee, the Cindy Wang story is coming to your company too.

The pressure to learn this stuff is not optional anymore. And the great news is that Duclor just made the learning curve almost flat. 50 a month, no setup, no coding, the only thing standing between you and your first AI agent is the decision start. The window where learning this gives you a real edge is still open, but Duclor is the product that started closing it because every single day more people use it, more of the advantage gets distributed and the gap between early movers and everyone else gets harder to close.

Feng, the engineer who built a business charging people to install OpenCore is now using his earnings to fund his own AI venture. His goal is to run a one person company using only AI agents. He's given himself a year. One person, one year, AI agents doing the work of a team.

That's a bet. And Duclor is part of what makes that bet credible because when your AI agent runs on Baidu's infrastructure for 250 a month and comes preloaded with search, research, multimodal, switching, the one person company is not really a fantasy anymore. It's the business model. Lobster has arrived, Duclor opened the door.

And the question is whether you walk through it. I'll see you in the next one. If you haven't already, check out the AI Profit Boarding, link in the comments description or just go to the AIProfitBoarding.com.