Post ContentOpenClaw, an open-source autonomous AI agent, has drawn growing interest from developers, companies, and regulators amid rising security concerns. (Express Image)
OpenClaw, the open-source, self-hosted autonomous AI agent, appears to have attracted scrutiny from the Chinese authorities. Authorities in China are reportedly prohibiting state-run enterprises and government agencies from installing OpenClaw AI applications on their office computers.
The Chinese government is taking swift measures to dispel any threats owing to the security concerns related to the AI agent. In the last two months, Chinese consumers have been voraciously experimenting with the AI assistant for various applications.
According to a report in Bloomberg, the government agencies, which also include some of China’s largest banks, received notices asking them to not install OpenClaw on office devices. Many of these agencies were directed to report to their superiors in case they had already installed the apps, and to start security checks and possible removal of these apps.
Also Read | Why OpenClaw is attracting even 60-year-olds in China
Several employees from state-run banks and some government agencies were prohibited from installing OpenClaw on office computers and even on their personal phones that operated on company networks. The ban is also reportedly extended to the relatives of Chinese military personnel. Some of the notices did not imply an outright ban but stated that prior approval is needed to use it.
Following the development, shares of companies with tools based on OpenClaw lost their earlier gains. Tencent’s stock, which was on the rise earlier, gave up most of those gains. On the other hand, stocks of MiniMax declined further, wiping out its earlier hike and ending up more than one per cent lower.
Why is China concerned?
The latest stance of the Chinese government highlights its concerns about OpenClaw, which requires broad access to private data and possesses the ability to communicate externally. This could likely expose it to external attacks.
Regardless of the potential risks, companies like Tencent and JD.com Inc. are extending OpenClaw apps to cash in on the ongoing excitement about the agentic AI phenomenon. Several local government agencies have announced millions in subsidies for companies that are building on top of the platform.
Story continues below this ad
Earlier this week, 60-year-olds were seen lining up to try their hand at OpenClaw during installation drives in Shanghai that was organised by companies such as Tencent. Many users were seen visiting pop-up stalls where technicians helped them install the agentic AI software on their personal computers.
Also Read | Inside Moltbook, the ‘Reddit for AI’ where bots are building their own society
The increasing interest in OpenClaw is reflective of China’s fast-moving tech ecosystem, which is fuelled by a fear of missing out among entrepreneurs and developers keen to build products and services backed by AI agents.
Ever since OpenClaw was introduced, Chinese startups have been experimenting with innovative ideas such as AI-agent social networks, recruitment platforms, and automated influencer management. Major tech firms like Alibaba, Tencent, and Baidu are also supporting this ecosystem with cloud services, allowing developers and hobbyists to experiment with AI agent-driven automation.
What are the security concerns?
OpenClaw, launched in November 2025, has been highlighted as one of the most important AI advancements in recent times. The AI agent operates autonomously and runs locally. However, the security concerns surrounding OpenClaw have been growing. Several cybersecurity experts have warned that the tool may pose a combination of risks related to autonomy and deep system access, such as prompt injection, malicious plugins, data leaks, and supply-chain attacks.
Story continues below this ad
In China, there are also apprehensions about loss of control of a potentially ground-breaking technology. China’s Communist Party has been tightening oversight of major technology companies such as Alibaba, which have gone on to amass vast amounts of user data and become deeply embedded in everyday life there, raising concerns in Beijing about the growing influence of private tech companies.
Moreover, the Chinese authorities have largely been vigilant about AI, expressing concerns about the potential dangers of AI agents to key sectors such as finance and energy.
