China issues new safety rules for OpenClaw. Here are the dos and don’ts

China's Ministry of Industry and Information Technology (MIIT) has released new safety guidelines for OpenClaw, a widely used AI agent. Developed with AI providers and cybersecurity firms, the advisory outlines best practices and prohibitions for OpenClaw adoption to address potential risks. Recommendations include using the latest official version, minimizing internet exposure, granting minimal permissions, and exercising caution with third-party skills. Users are warned against using outdated versions, exposing instances to the internet, enabling administrator accounts, installing skill packs requiring passwords, browsing unverified websites, and disabling log auditing. The guidelines also provide instructions on restricting internet access, scanning files, and uninstalling the software, highlighting risks associated with connecting instant messaging apps due to excessive permissions.

Explore how the topics in this article connect to other news stories