（Singapore, 11.03.2026）Chinese authorities have begun limiting the use of a rapidly growing artificial intelligence tool known as OpenClaw in government agencies and state-owned enterprises, citing potential cybersecurity risks. The move comes just months after the technology sparked a surge of excitement across China’s technology industry.
According to people familiar with the matter, Bloomberg reported that government departments and state-owned companies, which include some of the country’s largest banks — have received notices warning employees not to install OpenClaw software on their office computers. The instructions were issued in recent days as officials sought to address potential security risks linked to the emerging AI system.
Some organizations have gone further, asking employees to report whether the software has already been installed so that security teams can review and possibly remove it. The individuals familiar with the internal notices spoke on condition of anonymity because the instructions were circulated privately.
In certain institutions, staff members were also prohibited from installing the software on personal mobile phones if those devices are connected to the organization’s internal network. One person familiar with the guidance said the restrictions could even extend to family members of military personnel.
Other agencies have not imposed a full ban but have told staff they must obtain official approval before using the tool.
Authorities including China’s Ministry of Industry and Information Technology and the State-owned Assets Supervision and Administration Commission have not publicly commented on the new restrictions.
OpenClaw Sparks Rapid Adoption Across China’s Tech Sector
OpenClaw is part of a new generation of agentic AI systems, designed to perform tasks autonomously rather than simply responding to user prompts. The software can connect to various apps and services to carry out actions on behalf of users.
The tool can automatically manage email inboxes, organize schedules, make restaurant reservations or even check users in for flights. Users can control the AI agent through messaging platforms such as WhatsApp or Slack.
Originally known as Clawdbot and later Moltbot, the software was created by Austrian developer Peter Steinberger and launched in November. Since then it has attracted a passionate global user base, especially among programmers experimenting with autonomous AI.
In China, enthusiasm has been particularly strong. Tech companies and startups quickly began building services and tools based on the platform, creating what many observers describe as a fast-growing ecosystem.
Major Chinese technology companies have moved quickly to embrace the trend. Firms such as Tencent, Alibaba and other cloud providers have introduced services that allow developers to easily deploy OpenClaw on their platforms.
At the same time, AI startups including Moonshot and MiniMax have released their own customized versions of the technology in hopes of boosting adoption of their models.
Local governments have also shown interest in promoting the technology. Cities such as Shenzhen and Wuxi have proposed policies offering millions of yuan in subsidies to encourage companies to develop applications based on OpenClaw.
The enthusiasm has also been reflected in financial markets. Shares of some companies linked to the technology have surged as investors bet that AI agents could become a major new business opportunity.
For example, AI developer MiniMax has seen its share price jump sharply since its stock market listing just two months ago. The company recently launched a new AI agent called MaxClaw, built on the OpenClaw platform.
Security Risks Raise Alarm Among Regulators
Despite the excitement, cybersecurity experts have warned that autonomous AI agents may carry serious risks.
Unlike traditional software, OpenClaw requires broad access to user data in order to perform tasks. The AI can also communicate with external systems and interact with content from the internet, potentially exposing computers to outside attacks.
Experts say this combination of deep data access, external connectivity and interaction with unverified information can create significant vulnerabilities.
One cybersecurity researcher described the mix as a “lethal trifecta.”
In one widely shared incident, a user reported that an AI agent connected to their iMessage account began sending hundreds of spam messages after behaving unexpectedly.
Chinese regulators have already flagged similar concerns. Last month, the Ministry of Industry and Information Technology warned that certain OpenClaw deployments could lead to data leaks or cyberattacks if they are configured improperly.
The cautious response from Beijing reflects a broader focus on data security in China. President Xi Jinping has repeatedly emphasized that data is a key component of national security.
Over the past decade, the government has introduced strict rules governing data protection, cybersecurity and the operations of major internet companies. Officials are particularly concerned about sensitive information, including financial data, geospatial mapping data and genetic databases being exposed to foreign actors or cyber threats.
Against that backdrop, technologies like OpenClaw that can access large amounts of information and communicate with external networks have raised alarm among policymakers.
Tech Stocks Slip as Policy Uncertainty Grows
News of the government warnings had an immediate impact on Chinese technology stocks.
Shares of Tencent erased most of their earlier gains during afternoon trading, while several AI-related companies including MiniMax and Zhipu, also known as Knowledge Atlas Technology, fell more than 6%.
The decline reflects growing uncertainty about how regulators might approach the rapidly expanding AI agent sector.
For China’s leadership, the challenge now is balancing technological innovation with national security concerns.
On one hand, tools like OpenClaw are seen as potential drivers of productivity and digital transformation. Some investors believe autonomous AI agents could eventually reshape the way businesses operate.
On the other hand, the technology remains relatively new, and experts say its risks are not yet fully understood.
For now, China’s approach appears to be one of careful monitoring rather than an outright ban. But the latest restrictions suggest regulators are moving quickly to ensure that rapid adoption of powerful AI technologies does not come at the cost of data security.
