Who is responsible for L3 level private car accidents on the road? The implementation of the first autonomous vehicle regulation for the expansion of personal passenger vehicles in China will start today, and Beijing will take the lead in allowing autonomous private cars above L3 level to go on the road. According to the autonomous vehicle Regulations of Beijing Municipality (hereinafter referred to as the Regulations) officially implemented on April 1, "personal passenger vehicle travel" has been added to the application scenarios of L3 and above intelligent driving levels, opening the "door" for ordinary car owners. At the same time, the Regulations also provide relevant provisions on how to determine the responsibility for autonomous vehicle accidents. At the recently held China Electric Vehicle Hundred People Forum (2025), relevant personnel from the Ministry of Industry and Information Technology also clarified that they will conditionally approve the production access of L3 level autonomous driving models. Therefore, the implementation of this regulation is regarded as an important watershed for the autonomous driving industry. Since the beginning of this year, automobile manufacturers have actively accelerated the equal rights of advanced intelligent driving, especially focusing on the mass production and landing of L3 level passenger cars for the private car market. Is there a reference answer for the division of responsibility in autonomous driving accidents? According to the classification of autonomous driving, levels L1 to L2 belong to assisted driving, while level L3 is the first level dominated by the system, defined as "conditional autonomous driving". In certain specific scenarios or environments, drivers do not need to closely monitor road conditions and can entrust vehicles to autonomously complete driving tasks. The reporter noticed that Beijing has taken the lead in launching L3 level road testing points, and car companies such as BMW, Mercedes Benz, Changan, and BYD have successively obtained testing licenses. So, who is responsible for accidents involving autonomous vehicles? Articles 31 and 32 of the Regulations point out that: when autonomous vehicle are on the road, if they violate the laws and regulations on road traffic safety or have traffic accidents, the public security and traffic management departments shall investigate and deal with them in accordance with the relevant provisions of the state, and the relevant enterprises and individuals shall cooperate in the investigation and handling of the accidents, and provide corresponding evidence materials as required. In addition, like other motor vehicles, autonomous vehicle must also comply with existing traffic safety laws and regulations. According to the Rule of Law Daily, relevant officials from the Legal Work Committee of the Standing Committee of the Beijing Municipal People's Congress explained that the determination of responsibility for traffic accidents belongs to the legislative power of the state, and local legislation in Beijing has no authority to make regulations; Therefore, the Regulation has been linked with national regulations in this regard, requiring the public security traffic management department to investigate and handle traffic accidents in accordance with relevant national regulations. Meng Xin, from Beijing Junzejun Law Firm, said that the current legal system faces many challenges in dealing with the identification of liability for autonomous vehicle accidents. At present, China still mainly divides the responsibility for relevant accidents based on the Road Traffic Safety Law and the Product Quality Law. This means that Beijing has pioneered the landing of L3 level, but further clarification is still needed in the division of responsibility for autonomous driving accidents. Driven by policies, the landing of L3 level has become a new goal for car companies. Since the beginning of this year, "advanced intelligent driving" and "equal rights for intelligent driving" have become the core "selling points" for car companies to compete in marketing. According to data from the Ministry of Industry and Information Technology, the proportion of new passenger cars with L2 level combined driving assistance functions will reach 57.3% in 2024. The implementation of the Regulation has to some extent encouraged the accelerated landing of domestic L3 level vehicles. At present, Huawei ADS 4.0, Xiaopeng XNGP and other systems are accelerating their adaptation to Beijing's road conditions, with the goal of achieving a dual breakthrough of "urban NOA+L3 highway release" within the year. Guangzhou Automobile Group has also set a very clear timetable. In March of this year, it announced that its first domestic L3 level autonomous passenger car will be mass-produced and launched in the fourth quarter, and it will simultaneously promote the production of the industry's first L4 level front-end production model. It is reported that as one of the first domestic car companies to be approved for L3 level road trials, GAC has accumulated over 4 billion kilometers of L2 level intelligent driving mileage and 40 million kilometers of L4 level Robotaxi operation data (relying on companies such as Qixing and Xiaoma Zhixing). At the recently held China Electric Vehicle Hundred People Forum (2025), Xin Guobin, Deputy Minister of Industry and Information Technology, once again announced that he will "conditionally approve the production access of L3 level autonomous driving vehicles, and promote the improvement of laws and regulations on road traffic safety, insurance, etc. Zhang Yaqin, an academician of the CAE Member and president of the Intelligent Industry Research Institute (AIR) of Tsinghua University, predicted that 2025 would become a key node for breakthroughs in automatic driving technology, and he even predicted that by 2030, about 10% of new cars in the world would have L4 capabilities. Industry analysis shows that L3 level driving accidents require car companies to take responsibility. "Measuring whether it is truly L3 depends on whether the car company dares to compensate after the accident." Yu Qian, co-founder and CEO of Qingzhou Zhihang, believes that when the responsibility for intelligent driving accidents shifts from the driver to the car company, it is the core symbol of the transition to the level of autonomous driving. He pointed out that L3 should not be a marketing slogan. No matter how redundant the technology is or how regulations require it, if car companies dare not promise accident compensation, it is essentially still L2 level. In the view of Ouyang Minggao, an academician of the CAS Member, ensuring the safety and reliability of the large model "boarding" is still a key issue to be tackled. Not only does L3 level autonomous driving need to solve relevant legal and regulatory issues as soon as possible, but L4 level autonomous driving also needs to accumulate experience gradually. At present, it is not appropriate to propose "national autonomous driving". The popularization of autonomous driving is destined to be a protracted battle, from breaking the ice of policy to gaining the trust of the whole nation. (New Society)