The rapid advancement of autonomous driving technology has significantly accelerated and has become a widely recognized fact in the passenger car field by 2024. In just two years, we have witnessed the application of LiDAR, the popularization of urban NOA pilot assist driving systems, to discussions on the value of Automatic Emergency Braking (AEB) features. Emerging automakers like Huawei, XPENG, NIO, and Li Auto have pushed intelligent driving to become standard configuration in passenger vehicles.

These companies have adopted various marketing strategies, including technology cross-platform comparisons, high-level marketing, and diversified corporate competition, which have effectively promoted the recognition of intelligent driving systems and profoundly influenced consumers' psychology and car purchasing decisions. According to IDC's forecast, by 2024, the number of new vehicles on the Chinese market that meet the L2 autonomous driving standard will reach 14.78 million, accounting for 59.8% of the entire new car market, whereas this proportion was just about 30% two years ago.

The development of autonomous driving needs to "cross the commercial chasm", meaning that technology must shift from concept validation, prototype development to a marketable, profitable finished stage in the market. The passenger vehicle field has taken the lead in this leap, gathering a geek-like user group and gradually accumulating enough market momentum to shift to the mainstream market. The core driving force behind this development is showing the real value of the technology to users.

When observing another area closely related to socioeconomic development—mainline logistics, we find it possesses vast commercial space and incomparable economic and social value. A company called Winning Innotek has demonstrated the ability to "cross the commercial chasm" in this sector, with its intelligent driving solutions having achieved 100 million kilometers of commercial driving distance.

How did Winning Innotek achieve this feat when the commercial vehicle field has not yet reached the same level of commercial maturity as passenger vehicles? In fact, passersby car users' purchase of intelligent driving systems may not be entirely rational; some of it is to alleviate the driving burden, with a portion stemming from imitation and vanity of other users. Nevertheless, in actual driving, the usage rate of intelligent driving functions may not be high, with actual usage scenarios possibly not exceeding 10%.

In stark contrast, trucks, as a production tool, have their economic value directly reflected in the orders of logistics enterprises, greatly alleviating the burden of long hours of work for freight drivers and consequently bringing a significant improvement in business efficiency. The market has an accurate cost-benefit calculation formula for the commercial value of this intelligent driving technology.

The reason Winning Innotek has been able to successfully overcome the difficulties is because its technology and products can meet corporate customers' expectations for Return on Investment (ROI). Winning Innotek's business expansion has comprehensively covered the entire mainline logistics industry, including the express delivery market with a significant demonstration effect, and the large-scale and structurally rigorous LTL dedicated lines and contract logistics markets.

The technology of autonomous driving in smart trucks has been rapidly developing in recent years, and its commercial value is being recognized by global logistics enterprises. As one of the industries that focus most on efficiency and cost, logistics enterprises are now quickly adopting autonomous driving technology to alleviate driver burdens, save on labor costs, and achieve fuel savings.

The business model of smart trucks lays the foundation for a sustainable and robustly growing business prospect. While enhancing industry efficiency, smart trucks significantly reduce the workload of drivers. Long-haul routes that previously required two drivers to take turns can now be completed by a single driver alongside an autonomous driving system. Drivers now act more as safety supervisors rather than full-time operators.

At present, the autonomous driving mileage of smart trucks accounts for 90% to 95%, which means that for routes that originally required long driving times, the actual operating time for drivers is greatly reduced. Additionally, the precision driving of the machines also saves a significant amount of fuel costs for enterprises.

Financially, smart trucks can bring significant cost savings to logistics enterprises. A simple and intuitive calculation is the cost recovery period; most enterprises can recoup their investment costs within 12 to 18 months. More detailed calculations show that after depreciation of additional costs, companies can save about 0.2 to 0.4 yuan per kilometer, which means a monthly cost savings of up to 6000 yuan.

With the continuous maturity and popularization of autonomous driving technology, more logistics enterprises are starting to seek products equipped with smart driving systems. Currently, almost all leading express delivery companies in China have begun to adopt this technology, which after strict verification and scrutiny, proves its commercial value.

From an industry perspective, smart trucks have become the benchmark in the trunk logistics field. Their strong demonstration effect has driven the transformation and upgrading of the entire industry. According to forecasts, this year will be a year of rapid commercial growth for smart heavy-duty trucks, and it will become another important milestone in the era of intelligent logistics.

In the past six months, a significant change is that the recognition of smart heavy-duty trucks has extended from the industry's top tier to the mid-to-high-end user groups, such as contract logistics enterprises and less-than-truckload service providers. They are beginning to realize the role that smart heavy-duty trucks play in enhancing user value. The benefits of smart heavy-duty trucks are multifaceted: they improve efficiency, reduce labor costs, enhance safety, and for practicing drivers, the increased daily safe driving mileage also helps to increase income. The recognition smart heavy-duty trucks now have in the entire industry signifies that a major turning point in commercialization has arrived.

So what is driving the widespread adoption of intelligent driving technology in small and medium-sized logistics companies? This natural progression is reflected in the following aspects: with more application cases, people's awareness of smart heavy-duty trucks continues to increase; the safety and stability of smart heavy-duty trucks are validated by an increasing number of people. First, small and medium-sized logistics companies are very concerned about transportation safety, especially those fleets that have had major accidents in the last three years, for them, the cost-performance ratio of investing in smart heavy-duty trucks is very high. Secondly, smart heavy-duty trucks can save a significant amount of fuel, which is especially beneficial for drivers of small and medium-sized enterprises with varying levels of technical skills. It can save at least 30,000 yuan in fuel costs for the enterprise each year. In addition, smart heavy-duty trucks can also reduce driver fatigue and lower recruitment costs. For individual drivers, smart heavy-duty trucks can increase the number of transportation trips per month, correspondingly increasing their income. For different customers, the benefits that smart heavy-duty trucks bring can form a comprehensive value system. At the same time, the popularization of intelligent driving technology in the passenger car field is also promoting the acceptance of intelligent driving throughout the industry.

How should a company respond to the various needs and unexpected situations put forth by clients during the collaboration process? Indeed, each day in the collaboration process brings new challenges, yet fundamentally no substantial issues have arisen. The clients' requirements are very clear, primarily focusing on three aspects: cost reduction and efficiency improvement, enhanced safety, and increased profits. These three aspects are of the greatest concern to clients, who expect us to perform better in satisfying these core needs. Surprisingly, major clients have longer validation cycles than expected, partly due to the pandemic, and partly because they are very cautious. However, new technology is always fascinating as it quietly transforms clients' concepts, behaviors, and operational modes. In the past, clients needed to manage aspects such as driver behavior, transportation safety, and fuel consumption. Now they realize they should focus on a core indicator: the proportion of autonomous driving. Assuming the precision, stability, and safety of autonomous driving surpass manual operation, the higher its proportion, the safer, more fuel-efficient, and more efficient the transportation becomes. Meanwhile, clients are also creating new management models, such as using new routing plans, fleet deployment, and relay methods to achieve efficient long-distance transportation with a single driver. New technology not only frees current productivity but also promotes business model transformation, and this is the joy that technological advancement brings.

We have established a data-driven R&D mechanism, which originates from the "shadow mode". In the global application within the smart truck industry, we have reached the most mature level and the widest application.

In terms of continuous product iteration, our update speed is very fast, with an average of one version every two months. Although this means significant R&D investment, we still effectively control costs. Training costs are not our main expense. The driving force of data comes from how we tap into and utilize data, which is a significant challenge for any company, especially startups. To address this challenge efficiently, we adopted several strategies:

Firstly, we focus on scenario recognition, only collecting truly valuable data to avoid unnecessary accumulation; secondly, we implement the automation of data labeling to increase efficiency and reduce costs; lastly, we rely on cloud training, which not only tests our own algorithm capabilities but also teaches us how to use the newly developed tools within the industry. At the same time, we are exploring how to effectively use the currently popular large models. We do not blindly invest a huge sum to develop our own models but prefer to use the open-source models from the industry, carry out lightweight processing on them, and then apply them to autonomous driving.

Regarding the long-term strategy of our business model, we believe that only services that can continuously and consistently provide high-quality experiences can become the cornerstone of user loyalty. Besides providing technology services to OEMs and smart driving capabilities to logistics companies, we are also exploring subscription services for autonomous driving technology. The initial user base of this business model is mainly small and medium-sized fleets, contract logistics companies, less-than-truckload, and dedicated line companies.

Specifically, when we saw Tesla reducing the monthly fee for its full self-driving software from $199 to $99, we recognized that price is a very important consideration for cost-sensitive logistics companies. However, the final decision on whether customers are willing to pay for the subscription of intelligent driving services will depend on the actual value of the service and its user stickiness. Looking at consumer behavior in other services such as music, video, and telecommunications, as long as the product is attractive enough, users will have a strong willingness to pay. Therefore, the long-term profitability model of intelligent driving will be based on whether it can consistently provide a quality experience under any circumstances.

Commercial vehicles, especially those designed for long-haul logistics on main routes, are expected to grow in attractiveness and customer commitment once their technology matures and becomes widely adopted in the next one or two years. The prevalence of autonomous driving is an important indicator of user stickiness: once the average rate of autonomous driving feature utilization exceeds 85%, the willingness of users to pay for subscription services will significantly increase, with price being the only barrier. Therefore, the focus at the current stage of the business model should shift to how to enhance the universality of the product, such as ensuring that autonomous driving can operate stably under varying road conditions and different types of goods, in order to maintain a high rate of use of autonomous driving. With the maturity of these technologies, subscription models are expected to grow rapidly.

For the commercial application of autonomous driving, particularly L4 level autonomous trucks, widespread use is expected by 2030. This goal is constrained by two main factors: first, the technical challenges. Technology is truly considered to have met safety standards only when both the OEMs and users deem it as safe. Second, regulation and oversight specific to this technology are needed, which will require assessments and validations by transportation and public safety departments to confirm the safety of the technology, thus enabling relevant products to gain road access rights.

In achieving this goal, the intelligent driving truck industry is expected to develop. Initially, there must be a sufficiently large market demand to ensure the work undertaken is substantively meaningful. The industry might first see the wide commercialization of L2+ level autonomous driving vehicles, followed by the gradual opening up of policies by the state supporting the large-scale commercialization of L3 vehicles. During this process, the industry's total driven mileage might increase to 1 billion kilometers, thus accumulating enough technology and data to lay a solid foundation for technological and commercial advancement.

The significance of 1 billion kilometers is extraordinary. Calculated based on the length of China's highways, 1 billion kilometers is equivalent to traveling China's entire highway network 1000 times. Such a wealth of foundational data provides a learning process equivalent to a human's driving experience for autonomous driving vehicles.

Once such mileage milestones are reached, competition within the industry may become further intensified. However, the key lies in the overall size of the market; one would prefer to have a share of a vast market rather than dominate the entirety of a smaller one. We look forward to seeing an even more prosperous and highly competitive industry.