China’s OpenClaw Boom Is a Gold Rush for AI Companies

George Zhang thought OpenClaw could make him rich, even if he didn’t really understand how the AI ​​viral agent software worked. But he saw a video from a Chinese social media influencer demonstrating how it could be deployed to manage stock portfolios and make investment decisions autonomously. Zhang, who works in cross-border e-commerce in the Chinese city of Xiamen, was intrigued enough that he decided to try installing OpenClaw in late February.

Zhang is one of several people in China who have recently been swept up in the OpenClaw craze. Workshops teaching people how to use the AI ​​agent have sprung up in cities across the country, attracting hundreds of people. Tech companies are rushing to integrate OpenClaw into their platforms, while local governments have announced grants for entrepreneurs who build products with it. Late last week, images of grandfathers and grandmothers lining up to install the software went viral on the Internet.

After renting a cloud server from Tencent and purchasing a subscription to the large Chinese language model Kimi, Zhang was able to start chatting with his OpenClaw agent, or his “lobster,” as many Chinese call theirs. At first, Zhang told me, he was impressed by the AI ​​agent as he watched it quickly generate a lengthy market analysis based on the latest news. But a few days later, his lobster began to loosen, and it only produced an overview of market trends instead of a detailed report. He asked OpenClaw to generate something like what he did on day one, to which the agent perpetually responded that he was “working on it” before never returning any results.

Zhang’s conclusion is that OpenClaw is not designed for people like him who have no coding skills. “That would tell me that I needed to configure the API port. But that’s a technical task, which I can’t do without having a tutorial that guides me step by step,” he says. Ultimately, he gave up on letting his lobster stocks be traded, settling for asking him to aggregate AI industry news, which he used to create a social media content farm on WeChat.

This week, I interviewed half a dozen OpenClaw users in China about their experiences with the agent, and a clear picture of the divide emerged between adopters who are technologically savvy and those who are not. AI-savvy people see OpenClaw as a revolutionary productivity tool, but those without a technical background feel like they were promised a miraculously powerful AI product that ultimately didn’t deliver. But by the time the bubble burst, they had already started paying for cloud servers and LLM tokens.

The real driving force behind the OpenClaw craze in China is not ordinary users, but rather the Chinese companies that will benefit financially from its widespread adoption. Big tech companies like Tencent, Alibaba, ByteDance, Minimax, Moonshot, and Z.ai have all seen AI productivity FOMO as a rare chance to get normal people to start paying for AI services, and they’re reaping the biggest rewards.

“A chatbot only uses a few hundred tokens per conversation; a single active OpenClaw instance can consume tens or even hundreds of times more tokens per day,” explains Poe Zhao, technical analyst and founder of the newsletter. Hello China Tech. Every new OpenClaw user is someone who pays 24/7 for LLM API calls. “That’s why Tencent engineers set up tables outside the headquarters to help people install the software for free,” he explains.

“I didn’t understand any of this.”

Song Zhuoqun, a student in China, says she started experiencing problems with OpenClaw as soon as she tried to install it. Song is a social media intern at an AI startup, but has no programming experience. So it proved difficult to figure out how to make OpenClaw work. She asked Doubao, ByteDance’s popular AI chatbot, to generate a step-by-step tutorial for her, but it didn’t help much.

“There were pages full of code, and I didn’t understand anything. I kept asking the AI ​​to generate a response for me, and then I would paste it in, run it, and it would encounter an error, so I would try a new response,” she says. The installation ended up being the most frustrating part of trying OpenClaw for Song, and she didn’t feel like she learned anything from it.