A critical unauthenticated remote code execution vulnerability in Unitree G1 EDU firmware allows network-adjacent attackers to gain root access to the robot.

This vulnerability consists of an unauthenticated remote code execution chain involving an insecure WebRTC-to-DDS bridge on TCP port 9991, the use of a static AES-128 key with world-readable permissions, and a path traversal flaw in the chat_go knowledge upload API. These issues allow an unauthenticated attacker to inject malicious payloads into the bashrunner service and execute arbitrary commands as the root user.

A successful exploit grants an attacker full control over the affected robotic unit with root privileges. This could lead to complete loss of confidentiality, integrity, and availability of the device, potential physical safety risks if the robot is operational, and unauthorized access to internal network environments where the robot is connected. Given the CVSS score of 8.8, this represents a high-severity risk that demands immediate attention for all deployed units.

Immediate Action: Update the firmware on all Unitree G1 EDU units to the latest available version provided by the vendor. Ensure that any devices currently connected to public or untrusted networks are isolated until the update is applied.

Proactive Monitoring: Monitor network traffic directed to TCP port 9991 for anomalous patterns and review system logs for unauthorized service restarts or unexpected file creation within the bashrunner directories.

Compensating Controls: Implement strict network segmentation to ensure that robots are not accessible from untrusted network segments. Use firewall rules to restrict access to the WebRTC-to-DDS bridge (TCP 9991) to only authorized management workstations.

Public Exploit Available: Yes, a public proof-of-concept exists as documented in the research write-up by boschko.ca and the associated GitHub repository UniBLEed.

The severity of this vulnerability, combined with the availability of public proof-of-concept code, poses a significant threat to all G1 EDU users. Security teams must prioritize the verification of firmware versions and apply the vendor-supplied patches immediately. Failure to address this flaw leaves the robotic platform susceptible to complete compromise by any network-adjacent attacker.