China’s OpenClaw craze tests Xi’s approach to regulating AI

issued Friday, March 13, 2026 · 09:04 AM

[SINGAPORE] The OpenClaw craze that has spread across China is testing the Chinese government’s approach to artificial intelligence (AI) as it seeks to balance driving cutting-edge innovation with concerns about data security.

In recent weeks, tech giants and consumers have rushed to embrace the agentic AI phenomenon, with hundreds lining up outside Shenzhen tech shops at peak times to get help installing the software.

But the runaway import of foreign technology has caught the attention of authorities, who on Wednesday (March 11) warned state-run companies and government agencies not to install the technology on office computers, according to people familiar with the matter.

China is keen to establish itself as a world leader in AI, pushing to integrate AI throughout its economy with the aim of transforming industry and daily life. The new directive highlights how its ambitions run counter to the ruling Communist Party’s instincts for stability and control.

“Chinese regulators typically respond with extraordinary speed to threats from emerging technologies, but the adoption rate of OpenClaw and other agent tools continues to outpace regulators,” said Kendra Schaefer, partner and director of technology policy research at Trivium China.

Software such as OpenClaw, an open-source autonomous AI agent developed by an Austrian that went mainstream this year, can perform tasks from email cleanup to calendar management to flight check-in. The act of installing AI agents also has the nickname “lobster farming” in China.

Tencent, Alibaba, MiniMax, and Baidu have all announced OpenClaw compatible tools. Local governments in cities such as Shenzhen, Wuxi and Hefei have announced large subsidies for startups building on the platform. However, it requires very extensive access to personal data and the possibility of communicating with the outside world, which can expose your computer to external attacks.

Ying Tongyue, chairman of Chery Automobile, one of China’s largest electric car makers, said amid the uproar that he had asked employees to refrain from using the software until a training program was implemented. “I said everyone should refrain from installations now so we can do a focused training program. We have to be willing to accept new things, but we can’t blindly follow the crowd…that can lead to some risks beyond our imagination,” he said.

Regulatory authority response

The Chinese government has previously sounded the alarm about foreign attackers targeting datasets containing geospatial and genetic information, and the speed with which OpenClaw has been deployed increases the urgency of regulatory action.

Unlike the European Union, the Chinese government has avoided enacting comprehensive AI legislation. Instead, it has imposed temporary measures to address specific issues, such as algorithmic recommendations and rules around deepfake content, starting in 2022. Last year, it became the first government to require AI-generated content to be labeled.

“Beijing’s biggest challenge in regulating AI is the same one facing any government: technology is moving so fast that regulations could be outdated before the ink is dry,” said Matt Sheehan, a senior fellow at the Carnegie Endowment for International Peace.

So far, there are no regulations specifically targeting this cutting-edge software, which grapples with questions such as who is responsible for the actions of AI agents. On Wednesday, the China Academy of Information and Communication Technology, a research institute under the Ministry of Industry and Information Technology, announced that it will test the reliability of AI agents such as OpenClaw starting in late March and will develop a set of standards for their use.

“I’m definitely concerned about security,” said Ryan Shi, a Chinese language teacher in the southern city of Jiangmen. He uses AI agents to assist with repetitive tasks such as creating presentation slides and filling out forms. “So I’m looking for workarounds, such as running Openclaw in a Docker container and sandbox, or configuring certain rules to avoid exceeding limits.”

In a sign of the growing importance of the technology sector in the economy, the Chinese government has set a target to increase the added value of the digital economy’s core industries to 12.5% ​​of gross domestic product by 2030, up from 10.5% last year.

That is where AI and “lobster fever” present another challenge: social stability. The country already suffers from a weak labor market and a youth unemployment rate that has hovered above 15% for the past six months, with automation threatening to displace the world’s largest workforce.

Lu Jianhua, an academic at the Chinese Academy of Sciences and a professor at Tsinghua University, said AI is being used to study low-altitude economic infrastructure that previously required small teams of researchers.

“AI can act as a very capable assistant, comparable to several human assistants,” said Lu, a member of the Chinese People’s Political Consultative Conference.

A recent study by Peking University that looked at more than 1 million job postings found that hiring numbers are already declining in the sectors most exposed to AI, such as accounting, editing, and programming. The Ministry of Human Resources announced in January that it was drafting policy guidance to address the impact of AI on employment, but did not provide a release date.

Victor Chen, a fintech worker in Guangzhou who has been using OpenClaw since early February to help with several projects, from developing a mobile game to writing a novel, is also worried about losing his job.

“The more important underlying factor is that governments, like the rest of the world, are unprepared to actually deal with the mass unemployment caused by AI and the social unrest it could cause,” he said. bloomberg