Why is Baidu car-making not favored by the industry?

Baidu announced the construction of cars, the industry is boiling, but seems to watch more lively.

On January 11, Baidu officially announced that it would enter the automobile industry as a vehicle manufacturer. At the same time, it said that Zhejiang Geely holding Group Co., Ltd. will become a strategic partner of the new company, Baidu will hold a majority stake and absolute voting rights in the new company, while Geely, as a strategic partner, is the only management of the new company except Baidu, holding only a small stake.

After the Baidu official announcement, Yang Xueliang, vice president of Geely Automobile Group, said on social media that "the two sides will work closely together in smart car manufacturing related fields based on the vastness of SEA, the world's leading pure electric architecture newly developed by Geely, to build the next generation of smart cars."

On February 18, Baidu revealed again in its results for the fourth quarter of 2020 and last year's annual results that the newly formed car company will target the passenger car market, mainly smart electric vehicles, and is expected to launch a new car within three years.

Robin Li, chairman and CEO of Baidu, also said that the company's CEO, had been appointed to choose the brand name. It was later reported that Xia Yiping, co-founder and chief technology officer of mobike, would take up the post.

Baidu cross-border car building is supreme. However, since the announcement of car-building, although Baidu's share price has risen, coupled with positive earnings data, many voices have expressed optimism about Baidu, but observations have found that most of the industry lacks confidence in its car-building.

So, as a leader in autopilot and intelligent Internet connection, why is Baidu not favored to build cars?

Baidu had no choice but to build a car in person.

"Baidu built a car to catch up with the next tuyere and to land its own artificial intelligence." A senior figure in the auto industry told GAC that capital favors areas with higher technology density, especially before a complete outbreak, where it needs to occupy space quickly.

The automobile is the aggregator of the latest technology, and the intelligent car integrates a large number of Internet technology and technology genes. it is also a new important growth point for the business of technology companies, and it is almost the only tuyere of cutting-edge technology.

As a result, Internet giants such as Apple, Huawei, Tencent, Ali and Meituan have invested or directly participated in car building. Just recently, there is also news that Xiaomi wants to build cars across the border, although Xiaomi officially responded, "wait and see, not yet", but there is no suspense, Xiaomi car building is imperative.

In 2020, Tesla's market capitalization rose nearly six times a year, ranking first among global car companies, and Musk, founder of Tesla, also became the richest man in the world. The share price of Lulai Motor rose 11 times in a year, 35 times higher than its lowest point at the end of 2019. Xiaopeng Motor and ideal car also rose 102% and 75% respectively.

Throughout the development of Baidu in recent years, it has missed the tuyere of e-commerce, social networking, takeout and other areas one after another. It had the opportunity to acquire a 5 per cent stake in Meituan for 2 billion yuan, but chose Nuomi.com, and then missed the tuyere of short videos and live broadcasts in 2018. Now, Baidu All in's AI, self-driving business has invested a lot of money, but the progress is slow.

After the announcement, Baidu's share price has soared to $340 from $140 in December, and the company's market capitalization has reached $115.9 billion. Xiaomi's market capitalization has also soared by HK $50 billion since it announced that it would build a car yesterday.

Baidu chose to build its own car, mainly because its self-driving technology can not really hit the ground. Zhang Xiang, an expert at the Talent Exchange Center of the Ministry of Industry and Information Technology and an automobile industry analyst, said that since the launch of the Apollo platform in 2017, there has been a great breakthrough in the commercial landing of the Apollo platform, while building its own car will be able to better verify and promote the platform and its unmanned vehicle business.

At this point, Li Yanhong also said that many traditional automakers are relatively conservative in this respect and are unwilling to take the initiative to try some new technologies, so Baidu can only pre-empt and take the initiative to build electric vehicles that integrate cutting-edge technology.

Two years after setting up Baidu's self-driving unit (ADU), in 2015, Baidu unveiled its Apollo plan, and Robin Li even drove his own driverless car, Shangwuhuan, to promote Baidu's self-driving project.

In April 2020, Apollo released its first full-stack intelligent transportation solution-ACE traffic engine. At present, the project has been put into practice in more than ten cities such as Beijing, Changsha, Cangzhou and so on.

Data show that Baidu's self-driving Apollo test fleet has reached 500 vehicles, with 1800 patents, a total of more than 6 million km of test mileage, and more than 170test licenses, including more than 120manned test license. obtained the first batch of T4 self-driving test licenses issued by the Beijing Autopilot Test Management Joint team.

But so far, Baidu's autopilot business is still unable to land, and it is even more difficult to make a profit.

"in 2019, both Xiaopeng and ideal went to Baidu for financing, but Baidu insisted on tying its Apollo plan to each other and not letting them do autopilot, and finally broke up in discord." Zhang Xiang said that no matter who it is, these core technologies must be developed by themselves, and no one wants to be strangled.

In the long run, self-driving requires long-term data accumulation, and short-term cooperation with car companies will not help Apollo self-driving platform reach the L5 level of self-driving.

At the same time, the development of RoboTax I, which is mainly promoted by Baidu, is not smooth, and the industry does not have much confidence in it. It is understood that European car companies and Uber have at least temporarily abandoned this road.

The above-mentioned industry analysts said that whether it is AI technology or self-driving technology, no matter how precipitated it is, it will be difficult to enter the mass market without hardware as a grasp. Now the car is still in the tuyere period, Baidu does not take action, then its self-driving system, which has spent so many years, may be in vain, and what it will miss is a trillion-class market.

Although Baidu chose to build cars in person, in Baidu's plan, the newly established automobile company will only cooperate deeply with Apollo, the newly established car manufacturing company will be independent of Baidu, and Apollo will continue to adhere to the open platform strategy to help auto partners "build cars."

What are the odds?

For Baidu car-building, the American stock goddess Cathrine Wood, known as the "female version of Buffett", bought Baidu shares on a large scale, and its ARK fund increased its position in Baidu for six consecutive trading days. Wood said it was because it valued Baidu's accumulated strength in self-driving and the huge imagination brought by entering the bureau to build a car.

Mo Chase even said that the Geely joint venture was strategically similar to Google's acquisition of Motorola's mobile phone division after raising its target price from $155to $290.

It is understood that in cooperation with Geely, Baidu will adopt the "SEA vastness" architecture, which took four years and invested 18 billion yuan in research and development, which takes into account the trinity of hardware layer, system layer and ecological layer, and can meet the development of A-class to E-class models.

In addition, "vast" can shorten the software development time by 50%, which can greatly shorten the vehicle development cycle. Lectra ZERO concept based on this architecture will also be mass-produced and listed in 2021.

Jia Xinguang, chief analyst at China Automotive Industry Consulting and Development Corporation, said: "despite several twists and turns, it is still not too late for Baidu to build cars in 2021. On this basis, Baidu can give full play to the technology of autopilot Apollo to achieve the complementarity of 'software + hardware'. "

A securities analyst also told Galaxy that as an Internet technology giant, Baidu can support its rapid growth and become a moat with core technologies such as artificial intelligence, Apollo self-driving, Xiaodu car and Baidu Maps.

But for Baidu car, the industry also has a lot of doubts about its success, and some people even think that "Huawei may be able to build a car, but Baidu is very difficult to succeed."

It is worth noting that according to a recent industry survey on Baidu car-making released by Galaxy, more than half of the participants in the survey of more than 1400 people thought that the probability of Baidu car-building success was less than 50%. Compared with other technology giants, only 6% of participants are optimistic about Baidu car-building, only higher than Alibaba, while the proportion of participants who support Huawei and Apple are 44% and 31% respectively.

Most of the reasons for not being optimistic are that Baidu is not familiar with the law of the development of the automobile industry and lacks the technology for sustainable large-scale landing. Up to 68% of the participants believe that the cooperation between Baidu and Geely is nominally dominated by Baidu, and that the core car-building technology is still in Geely's hands.

"now to develop an intelligent electric vehicle, the calibration and design of the whole vehicle need to be made forward-looking planning, and both Baidu and Geely need to be deeply involved before the model development," an engineer from China Auto Research told Geoshi. However, Baidu lacks uniform verification in the design, development, manufacturing and even sales service environment of smart vehicles.

In addition, the survey report believes that in C-end sales, because car sales have always been directly to C-end consumers, but Baidu does not have any C-side sales experience.

The above analysts said that although Baidu car is made by Geely, but in vehicle process design, supply chain management, sales channel operation and other aspects, the two sides need in-depth running-in, and Baidu also lacks experience. Some commentators are even worried about Geely's suppliers, saying that Geely has always been squeezed on the supplier side, and it is unknown that there will be a war of words in the future.

In addition, at the technical level, after a year and a half of cooperation with BAIC, Xu Heyi, former chairman of BAIC, said at an annual strategy meeting within BAIC that "after working with Baidu for a period of time, Huawei has made rapid progress." now the deepest cooperation with Huawei also makes people worry about Baidu car building.

There is also a view that Geely has a wide range of strategies, and it is also unknown whether and how much energy it has to take care of the cooperation with Baidu.

It is worth noting that Geely's partnership with Daimler, also based on the SEA architecture, will launch its first new car this year, and its other high-end electric car brand, Polestar, has already partnered with Waymo. In addition, Geely has its own cooperation with Volvo in self-driving technology, its own service provider of intelligent self-driving solutions for cars, and has also reached strategic cooperation with Tencent in smart cockpits, self-driving and so on.

However, the above analysts also said that "software-defined cars" have gradually become an industry consensus, and Baidu's technology reserves such as AI and self-driving are the important driving force to achieve this change.