A mandatory national standard for the safety of autonomous driving systems in intelligent connected vehicles has been officially released, with implementation scheduled to begin on July 1, 2027.
On August 4, according to information from the official website of the Ministry of Industry and Information Technology (MIIT), the mandatory national standard Intelligent Connected Vehicles - Autonomous Driving System Safety Requirements (GB 44721—2026) was approved and published on July 30 by the State Administration for Market Regulation and the National Standardization Administration. Formulated and overseen by MIIT, the standard will take effect on July 1, 2027.
In recent years, as a new wave of technological revolution and industrial transformation deepens, China's autonomous driving technology has accelerated its iterative breakthroughs. The demonstration of intelligent connected vehicles on roads has progressed in an orderly manner, and in 2025, two L3-level (Conditional Automation) vehicle models received conditional access permits. To actively adapt to the higher demands for standardization posed by the rapid development of the industry, MIIT has accelerated the construction of the intelligent connected vehicle standards system and the development of key and urgently needed standards. Since 2025, it has organized the National Technical Committee of Auto Standardization to initiate the drafting of this mandatory national standard.
Applicability and Core Framework
This standard applies to Category M and N vehicles equipped with L3-level and L4-level (High Automation) systems, but does not apply to automatic parking systems. Based on the actual needs of China's industrial development and industry regulation, and considering technical feasibility, industrial adaptability, and practical operability, the standard establishes a safety requirement system with a complete dimension, clear requirements, and adaptability to national conditions.
Four Key Safety Requirements
First, it strengthens the enterprise's full life-cycle safety assurance mechanism to solidify the foundation of safety. The standard requires vehicle manufacturers to establish a comprehensive assurance capability covering product design, development, manufacturing, and post-deployment processes from four dimensions: safety policy, risk management, safety assurance, and safety improvement, thereby reducing potential safety risks of autonomous driving systems at the source. It also stipulates that vehicle manufacturers must organize and conduct corresponding simulation, test track, and road tests during the R&D and verification of autonomous driving systems, ensuring that relevant test resources and conditions meet the development and verification requirements.
Second, it enhances the system's ability to perform the dynamic driving task (DDT) to ensure operational safety. The standard requires that the safety level of an autonomous driving system should at least be equivalent to that of a competent and attentive human driver performing the DDT, and it must not pose unreasonable safety risks to users or other road users. Based on this, it specifies safety requirements for the system's execution of DDTs in various scenarios, and clarifies the trigger and execution requirements for the minimum risk maneuver (MRM).
Third, it standardizes human-machine interaction and user notification requirements to prevent the risk of misuse and abuse. The standard requires that the activation and deactivation processes of the autonomous driving system must be safe, and specifies the notification norms for status information and signals such as ready, activated, and deactivated. For L3-level systems, it also requires a driver takeover capability monitoring function. Additionally, vehicle manufacturers, besides providing product manuals, must clearly inform users of key information such as the level of driving automation, system capabilities, limitations, control strategies, and user role transitions through channels like official websites or in-vehicle displays, to guide users in correct understanding and use.
Fourth, it improves multi-dimensional inspection and testing methods to ensure the standard is implementable. The standard builds an integrated inspection and testing system of 'enterprise capability inspection + safety file inspection + confirmatory testing'. This system verifies the enterprise's safety assurance capability and the completeness and robustness of the product safety file. On this basis, third-party testing agencies will conduct confirmatory testing by comprehensively using methods such as track testing, road testing, and simulation testing, based on the recommended national standards GB/T 41798, GB/T 44719, and GB/T 47025 for intelligent connected vehicle autonomous driving function testing. This ensures a comprehensive inspection and verification of the system's actual capabilities.
Comparison with International Regulations and Significance
Concurrently, the United Nations Global Technical Regulation on Autonomous Driving Systems (ADS GTR), led by China, was officially approved and published in June 2026. Compared to the ADS GTR, China's national standard has more detailed technical requirements for L3 and L4-level autonomous driving systems, further clarifies the safety boundaries for different product levels, improves requirements for user notifications and operational training, and innovatively establishes a unified standard test scenario system. While maintaining consistency with international regulations on core technical requirements, it is more closely aligned with China's road traffic environment and industry management needs, making it more feasible for implementation.
The release and implementation of the Intelligent Connected Vehicles - Autonomous Driving System Safety Requirements standard establishes a unified baseline for the safety access of autonomous driving products. It is of significant importance for standardizing the R&D and application of autonomous driving technology, improving road traffic safety levels, strengthening the primary safety responsibility of enterprises, and safeguarding the healthy and sustainable development of the industry. Moving forward, MIIT will solidly carry out the standard's interpretation, promotion, and implementation, strengthen the access management of intelligent connected vehicle products, and ensure the safe and compliant application of autonomous driving systems. Concurrently, it will continue to improve supporting standards for autonomous driving test methods, safety event data interaction and management systems, accelerate the construction of a comprehensive intelligent connected vehicle regulatory system, speed up the industrialization and large-scale deployment of autonomous driving technology, and lead the high-quality development of China's intelligent connected vehicle industry.
Comments