Longgang district of Shenzhen opened its AI and Robotics Safety Testing Center and Security Incident Response Center on June 25. The launch of two specialized institutions focused on integrated security scenarios for AI and robotics has created a full-cycle security service loop, featuring pre-incident testing and protection, as well as in-incident and post-incident emergency response, cementing a robust security underpinning for Longgang's strategic emerging AI and robotics industries.
The AI and Robotics Safety Testing Center has entered trial operation, delivering one-stop public testing technical services to local enterprises engaged in AI, intelligent robots and smart systems. Its service scope covers five major sectors, including standard compliance, AI model security, AI application security, intelligent robotics,and information security for internet of things devices.
Centered on the full life cycle of AI and robotic products, the center targets four core security links: environment, model, system, and application. It has fostered two core capabilities covering information security testing and testing platform research and development, and built a standardized comprehensive service framework covering personnel, equipment, workflows, and environments to help enterprises fully address their weaknesses in security protection. The center has unearthed multiple high-risk vulnerabilities that could enable remote control and trigger cascading damage.
Several such vulnerabilities have received official recognition from the China National Vulnerability Database, fully attesting to its robust capacity for in-depth risk identification and helping the industry defuse major security hazards, such as remote hijacking and robot system malfunctions.
The testing center has independently developed a suite of highly automated security assessment tools. These include the OpenClaw Security Assessment Platform, the Bluetooth Protocol Testing Platform for AI Robots, the Automated Testing Platform for Overseas Compliance under Standard EN18031 for Robots, the 4G and 5G Network Traffic Security Analysis Platform and the Firmware Vulnerability Scanning Platform. These tools enable fast, low-cost and in-depth security testing and greatly boost enterprises' self-inspection efficiency.
In the future, the center will continuously accumulate testing data and vulnerability cases to establish exclusive databases for security knowledge, industrial cases and vulnerabilities tailored to the AI and robotics sector. The databases will offer data support for the formulation of industrial standards and regular self-inspections by enterprises.
Unveiled on the same day, the AI and Robotics Security Incident Response Center functions as a special technical support platform for AI and robotics emergencies. It has built a full-process, closed-loop management system for industrial security risks, covering monitoring, early warning, response, and post-incident review. Focusing on the entire chain of research, deployment and operation for the AI industry, the center has established an integrated service system encompassing global situation awareness, risk research and early warning, incident reporting and acceptance, work order assignment and circulation, coordinated closed-loop response as well as post-event review and improvement. It provides six categories of public services including risk monitoring and early warning, round-the-clock emergency handling, vulnerability coordination and management, public opinion research, technical consultation and industrial empowerment.
Leveraging its dedicated emergency management platform, the center efficiently addresses 12 common security incidents. These include risks arising from model and algorithm safety, data privacy, network systems, content ethics, industrial robots, unmanned aerial vehicles, internet of vehicles, smart hardware and AI out-of-control scenarios. The whole workflow from graded risk assessment and expert dispatch to coordinated response and follow-up reviews is conducted online.
In addition, the center has established an analysis and research system for AI-related public opinion to capture trending industrial risks and common security issues in real time. It conducts regular vulnerability disclosure, risk alert releases, security training, and standard research. It shares the databases on industrial security knowledge, cases, and vulnerabilities with the testing center to enable two-way data exchange between testing records and emergency cases.
AI and robotics stand as priority strategic emerging industries for Longgang. The simultaneous launch and operation of the two centers mark a pivotal innovative move by Longgang to improve its regional service system for AI industrial security and advance the modernization of industrial governance. Moving forward, the two centers will act as core hubs to coordinate multi-party resources spanning security testing, emergency response, public opinion analysis, standard development and talent training. They will foster an integrated security service ecosystem centered on preventive testing and emergency backstops. The ecosystem will deliver a secure, stable, and trustworthy development environment for local AI and robotics enterprises and underpin high-quality growth of the regional AI and robotics industry through robust security safeguards.
From proactive risk inspection and automated security evaluation to round-the-clock emergency handling of sudden security incidents, Longgang now hosts the country's first two specialized institutions dedicated to AI and robotics security, filling a critical gap in industrial security services. Longgang will continue to roll out inclusive public security technical services in the future, help enterprises uphold their security and compliance bottom lines, and support steady, far-reaching innovation in AI and robotics sector.
