China’s OpenClaw AI Boom Turns Into a Security Crackdown

Authorities embraced the powerful new AI tool, only to warn that its capabilities could make it harder to control and riskier for sensitive systems.

Xiaohua Li

Published: 3/19/2026, 8:14:32 AM EDT

China’s OpenClaw AI Boom Turns Into a Security Crackdown — A man wears a lobster hat that represent the OpenClaw logo, an open-source AI assistant at the Baidu headquarter in Beijing on March 11, 2026. (Adek Berry/AFP via Getty Images)

An artificial intelligence tool known as OpenClaw has exploded in popularity across China in recent months, with developers, companies, and local governments rushing to “raise a lobster”—social media slang for running the open-source AI software named for its red lobster logo.

The boom has been fueled in part by Chinese cloud providers and local governments eager to promote AI development under Beijing’s “AI Plus” strategy, a national initiative introduced in the country’s 2024 policy report to expand artificial intelligence across industries and public services.

But as the software spreads through workplaces and campuses, Chinese authorities are warning it could expose organizations to cyberattacks and sensitive data leaks.

The warnings underscore a growing challenge for Beijing as it pushes to expand artificial intelligence across the economy to strengthen its technological rivalry with the United States, while trying to control the risks posed by increasingly autonomous AI systems.

OpenClaw’s Rapid Rise in China

OpenClaw was created by Austrian engineer Peter Steinberger and released in November 2025, according to developer reports. Unlike traditional chatbots that mainly answer questions, the software acts as an AI “agent” that can carry out tasks on a user’s behalf.

Once granted permission, it can interact with files, emails, and online services to complete multi-step assignments automatically—leading some developers to describe it as a kind of “digital employee.”

Technology executive Hsueh Tsung-chih, a former procurement manager at Taiwan Semiconductor Manufacturing Company (TSMC), compared OpenClaw to a more capable version of voice assistants such as Apple’s Siri.

Instead of simply answering questions, the system can proactively perform tasks—checking the weather, suggesting activities, or booking reservations, he told The Epoch Times’ Chinese-language edition.

Adoption in China has been rapid.

Major Chinese cloud platforms, including Tencent Cloud, Alibaba Cloud, and Baidu AI Cloud, have launched services allowing users to deploy OpenClaw on remote servers, Reuters reported.

Local governments have also promoted the technology. Authorities in Shenzhen and Wuxi have introduced draft measures to support companies developing OpenClaw-based applications, including subsidies and computing resources, said Reuters.

Developers said the technology could boost productivity by automating routine digital work.

Restrictions Hit Banks and Universities

Concerns about OpenClaw’s security risks began surfacing in early February.

On Feb. 5, China’s industrial cybersecurity response team warned in a security advisory that improperly configured OpenClaw deployments could expose organizations to cyberattacks.

The agency said the software’s ability to connect multiple tools, data sources, and AI models could allow attackers to exploit malicious instructions or supply-chain vulnerabilities to gain control of systems or extract sensitive information.

By March, those warnings had begun translating into restrictions across government-linked institutions.

Bloomberg News reported that Chinese agencies and state-owned enterprises, including major state banks, had received notices warning employees not to install OpenClaw on workplace devices for security reasons.

Some institutions instructed staff who had already installed related applications to report them to supervisors for security checks and possible removal, according to people familiar with the matter cited by Bloomberg.

In certain cases, employees at state-run banks and government agencies were barred from installing the software on office computers or personal phones connected to work networks. One person said the restrictions also extended to relatives of military personnel, Bloomberg reported.

Universities have also begun issuing bans. Several Chinese campuses, including Nanjing Institute of Technology, have ordered students and staff to uninstall OpenClaw from campus systems and warned against using the software in work involving teaching, research, or administrative data, according to university notices.

China’s AI Push Meets Security Control

OpenClaw’s rapid rise—and the subsequent crackdown—highlights a broader dilemma in China’s AI strategy.

“China wants to move quickly in AI and compete with the United States,” Wang He, a China affairs analyst and commentator, told The Epoch Times. “But when technologies raise security concerns, the government tends to prioritize control.”

Beijing has promoted artificial intelligence as a key driver of economic growth and technological competition with Washington. Official policy plans list AI, robotics, and advanced manufacturing among the priorities for China’s next phase of industrial development, according to Chinese policy documents.

That push is reflected in Beijing’s “AI Plus” strategy, which encourages the adoption of artificial intelligence across industries and public services, according to official statements.

At the same time, authorities maintain tight oversight of technologies that could affect cybersecurity or political stability, Wang said.

He added tools like OpenClaw, which can perform tasks across multiple digital systems, highlight both the promise and the risks of increasingly autonomous AI software.

For now, the OpenClaw crackdown illustrates the tension at the heart of China’s AI push: officials are encouraging rapid adoption of new technologies under the “AI Plus” agenda, while moving quickly to restrict them when control risks emerge.