The Era of Equal Access to Smart Driving: Balancing Safety in Technology Development

The era of “Intelligent Driving Equality” is accelerating, highlighting the need for both hardware and software to enhance technological safety. In February this year, BYD took the lead by proposing the concept of “Universal Intelligent Driving,” shifting the focus of the automotive industry to the realm of intelligent driving. Following this, automakers such as Geely, Chery, and Changan quickly joined the movement. Additionally, companies like Huawei’s HarmonyOS, GAC, and Xpeng are hastening the implementation of intelligent driving, particularly in advancing the mass production of Level 3 (L3) autonomous driving.

However, a recent serious traffic accident involving a Xiaomi SU7 has intensified market scrutiny regarding the safety of intelligent driving systems. Amid the controversy, the industry is questioning how to advance intelligent driving and establish a safety firewall, which has become a focal point of concern.

The acceleration of “Intelligent Driving Equality” is closely tied to BYD’s strategy for smart technology, which has nearly become a new trend in the automotive sector. Previously, BYD’s Chairman and President Wang Chuanfu stated that 2025 would mark the beginning of universal intelligent driving, asserting that vehicles without intelligent driving capabilities will become a “minority.” He believes that in the next 2 to 3 years, advanced intelligent driving will become as essential as seat belts and airbags. Wang identified high costs as the main obstacle to the widespread adoption of intelligent driving, hence BYD’s initiative aims to expedite its popularization.

BYD’s launch of the “Eye of the God” system marked the first significant move towards “Intelligent Driving Equality.” Subsequently, Geely introduced the “Qianli Haohan” intelligent driving solution, while Chery unveiled the “Falcon Intelligent Driving” platform, and GAC launched new vehicles equipped with the “Star Spirit Intelligent Drive” system. Furthermore, Changan, Great Wall Motors, Leap Motor, and BAIC have all revealed their respective intelligent driving strategies, with the core objective of capturing the market share for intelligent vehicles priced between 100,000 to 150,000 yuan.

In addition, the exploration of higher-level intelligent driving by companies like Huawei’s HarmonyOS, Xpeng, and Tesla is driving a rapid reshuffling in the intelligent driving sector. According to Open Source Securities, there is significant room for growth in the penetration rate of advanced intelligent driving, necessitating companies like BYD to educate consumers through large-scale deployment while leading first-tier industry players like Huawei and Xpeng in developing advanced driving technologies.

Yang Jing, Director of Asia-Pacific Corporate Ratings at Fitch Ratings, indicated that the upcoming competition among automakers regarding intelligent driving features will inevitably lead to a new wave of consumer education, helping promote the adoption of advanced driving configurations while eliminating brands that lack the capability to develop intelligent driving solutions or have not partnered with mainstream suppliers.

Yang predicts that in the next 3 to 5 years, the direction for advanced intelligent driving will progress from L2+ to L3 capabilities, although applications at L3 and above will still face legal and regulatory limitations. Therefore, for some time, the technology will operate within the confines of existing regulations, which, while not fully liberating drivers, aims to enhance the driving experience through refined hardware designs and software development.

Industry insiders analyze that the trend of “Intelligent Driving Equality” will transform the market landscape across various segments, including cameras, intelligent driving domain controllers, LiDAR, millimeter-wave radar, and domain control chips, all of which stand to benefit significantly. Zheng Jun, founder and CEO of Ding Technology, believes that “Intelligent Driving Equality” will accelerate the popularization of automotive intelligence, necessitating more cost-effective chip solutions. However, this will also lead to a rapid industry shake-up, requiring automakers, algorithms, and chip manufacturers to collaborate deeply, leveraging their respective advantages to offer consumers high-quality, low-cost intelligent driving solutions.

The penetration rate of advanced intelligent driving continues to rise. At the China Electric Vehicle 100 People Forum (2025) held in late March, Xin Guobin, Vice Minister of the Ministry of Industry and Information Technology, stated that efforts will be intensified to promote the industrialization of autonomous driving, advance the access and road testing of smart connected vehicles, and improve the standard system while conditionally approving the production access of L3 autonomous vehicles. On April 1, the “Beijing Regulations on Autonomous Vehicles” officially came into effect, allowing private vehicles equipped with L3 level and above intelligent driving systems to legally operate on roads for the first time. Prior to this, Beijing had already begun road testing L3 systems, with companies like BMW, Mercedes-Benz, Changan, and BYD obtaining testing licenses.

The proactive strategies of automakers, combined with the continuous enhancement of legal regulations, have accelerated the commercialization of intelligent driving vehicles. According to data from the Ministry of Industry and Information Technology, by 2024, 57.3% of new passenger cars will be equipped with L2 level combined assistance driving functions. Recent forecasts suggest that by 2025, the penetration rate of L2 and above level assistance driving in China’s passenger cars will reach 65%, while the penetration rate of intelligent cockpits in domestic passenger cars is expected to hit 76%.

Zhang Yaqin, an academician at the Chinese Academy of Engineering and head of the Intelligent Industry Research Institute at Tsinghua University, indicated that 2025 will be a critical turning point for breakthroughs in autonomous driving technology, with an estimated 10% of new vehicles globally expected to possess L4 capabilities by 2030.

In the face of increasing competition in the intelligent driving sector, a tragic accident involving a Xiaomi SU7 on a highway has raised alarms across the industry, sparking in-depth discussions about the safety of intelligent driving and bringing to light numerous potential issues related to it. Geely Automobile Group’s CEO, Gan Jiayue, emphasized that a vehicle lacking a solid “safety” foundation renders its intelligent features unstable, as safety pertains to human life and societal public interests.

Further technical validation is needed. Pan Helin, a member of the Expert Committee on Information and Communication Economy under the Ministry of Industry and Information Technology, also stressed the necessity for further standardization in intelligent driving, as electric vehicles must enhance their technical safety. He highlighted that the industry places particular emphasis on robustness, which refers to the equipment’s ability to handle complex scenarios. Currently, the robustness of batteries in electric vehicles and intelligent driving systems is inadequate. Pan added that automakers should not rush the manufacturing process, as the key to product competition lies not in intelligent driving or design, but in safety.

Some industry experts suggest that certain automakers, driven by cost considerations, have omitted LiDAR from their vehicles, focusing solely on visual systems. However, a comprehensive approach that integrates cameras, millimeter-wave radar, and LiDAR through multi-sensor fusion is more beneficial for enhancing a vehicle’s perception of its surroundings. Moreover, continuously optimizing algorithms and conducting extensive testing is essential for improving the decision-making accuracy and response speed of intelligent driving systems in complex scenarios.

According to Qiu Chunchao, CEO of SOTERIA, “LiDAR is a baseline for ensuring the safety of intelligent driving; if companies intend to develop high-level intelligent driving, they must invest in this technology.” He noted that the price of LiDAR has decreased, making it necessary for automakers to consider safety in their technological approaches. Liu Le, Chief Analyst at Guoyuan Automotive, predicts that the development of intelligent driving will exhibit a rapid progression despite its challenges, as the current state of automotive intelligence has not yet achieved L3 level autonomy, indicating that we are still in the initial phase.

The “Intelligent Driving Development Report (2024) – Intelligent Driving Section,” released by the China Electric Vehicle 100 People Forum, reveals that advanced intelligent driving has surpassed the “trial phase,” with features like highway NOA and urban NOA becoming mainstream in the 100,000 to 200,000 yuan price range. It is anticipated that by the end of 2025, the penetration rate of NOA in passenger vehicles will reach 20%, nearly doubling from the first half of 2024. Chen Chuanhong, Chief Analyst for Automotive at Guojin Securities, believes that reducing costs, exploring new markets, and increasing market share have become the new trends in automotive intelligence development, with urban NOA poised to be the next competitive battleground. By 2025, urban NOA is expected to become a key milestone in its development, targeting the mainstream market segment priced between 150,000 to 200,000 yuan, which is likely to drive intense competition among manufacturers in the next 2 to 3 years.

As mid-to-high level intelligent driving accelerates, there will be a corresponding increase in demand for high-processing power chips. For domestic chip manufacturers, this presents a valuable opportunity for replacement. Chen Chuanhong analyzed that especially for intelligent driving SoC chips, which serve as the central brain for enabling intelligent driving functions, there is a need for unified real-time analysis, processing of vast amounts of data, and complex logical computations, thus placing high demands on their computational capabilities. As intelligent driving features evolve, the demand for processing power in intelligent driving chips will continue to rise.