The decisive moment between Tesla and Huawei has finally come

Golden Horn Finance

2024-06-19 17:25Posted in Guangdong financial field creators

原创首发 | 金角财经（ID: F-Jinjiao)

Author | Zelda

The battle of intelligent driving of domestic new energy vehicles has finally ushered in the most terrifying opponent.

Recently, rumors that Tesla's smart driving is about to land in China have been rampant.

On June 14, some media reported that an official of the Shanghai Lingang Management Committee revealed that 10 Tesla FSD (Full-Self Driving) vehicles are being promoted to be tested in the area.

But on June 15, a screenshot of the official's statement was circulated on the Internet, saying that he did not know "the relevant information about the pilot of Company T's autonomous driving"; On the same day, according to Caixin, a person close to Shanghai's policymaking department said that the Shanghai autonomous driving demonstration zone had issued a road test license to Tesla, and FSD may be testing.

FSD is an evolution of Tesla's intelligent driving assistance system Autopilot, and Tesla hopes that FSD will be gradually upgraded and iterated to eventually achieve autonomous driving.

Since Musk arrived in Beijing in April, the talk of FSD's entry into China has begun to fly all over the sky, coupled with the recent news of "landing in Shanghai", which means that FSD's entry into China may not be groundless, but only a matter of time.

The pattern of the global electric vehicle industry is also on the eve of change due to Tesla's upcoming entry into China.

Cry wolf

In fact, before the recent rumors of "landing in Shanghai", the pace of Tesla's FSD landing in China has accelerated significantly this year. Especially after stepping into the second quarter, heavy news came one after another.

As early as April 20, Tesla CEO Elon Musk mentioned in response to questions from Chinese car owners that FSD may soon enter China.

Three days later, Musk said on the company's first-quarter earnings call that FSD can operate normally in almost all countries without changes, "just like you can drive well if you rent a car abroad."

He also said that regardless of the country in which it receives regulatory approval, Tesla will roll out a supervised FSD system locally, "and I think that includes China."

The rumors fermented to the extreme on April 28, and Musk arrived in Beijing on the same day. According to CCTV News, Musk visited China at the invitation of the China Council for the Promotion of International Trade, and Ren Hongbin, president of the China Council for the Promotion of International Trade, met with Musk that afternoon to exchange views on topics such as the next step of cooperation.

According to relevant media reports, Musk's surprise visit to China is aimed at promoting the implementation of Tesla's intelligent driving system FSD in China, as well as seeking permission to transfer the data collected in China to foreign countries for training autonomous driving technology algorithms. On the same day, according to the notice issued by the China Association of Automobile Manufacturers, Tesla passed all four requirements for China's automotive data security.

A day after Musk's arrival in Beijing, the text description on Tesla's China FSD purchase page changed from "to be launched later" to "coming soon". There is also news circulating on the Internet that Tesla China is recruiting FSD early adopters.

In addition, a person familiar with the matter told the media that Tesla and Baidu reached a cooperation in the field of maps, which met the requirements of the regulatory authorities. On April 22, Baidu publicly said that Tesla would be the first to carry the 3D leading version of Baidu Maps in May. "The implementation of FSD in China will be a historic moment, with an impact comparable to that of local production in China in 2020," the person commented. ”

After months of foreshadowing, the recent rumors of "FSD landing in Shanghai" indicate that it is only a matter of time before Tesla's FSD enters the Chinese market. And for domestic electric vehicle companies that are also betting heavily on intelligent driving, this means that the wolf is finally coming.

At present, Tesla's FSD is in a clear leading position in the field of intelligent driving. FSD has been launched on a small scale since October 2020 and has undergone 11 large-scale updates so far, with the latest version being FSDV12.

Cao Xudong, CEO of autonomous driving supplier Momenta, said that Tesla's FSDV12 performance is very impressive, compared with V11, the performance is between 10 and 100 times, and FSD is still evolving.

He also pointed out that the hardware cost of Tesla's FSD is only 4,000 yuan, and the ability to achieve such an effect on this basis reflects the competitiveness of FSD.

In addition, Tesla has recently taken the lead in landing an "end-to-end" architecture. End-to-end refers to breaking the artificially demarcated different modules and rules, directly inputting information from the perception end, and the system gives the driving decision results. In this way, information loss is reduced, and the system's driving decisions are therefore less "rigid" and more human-like.

FSD, with strong technical barriers, is menacing. Pattern changes, survival of the fittest, and pros and cons have become issues that every participant in China's new energy vehicle industry has begun to think deeply.

Huawei is not convinced

For Tesla's FSD to enter China, there are also a variety of voices in the Chinese industry.

He Xiaopeng, chairman of Xpeng Motors, publicly said that Tesla's FSD is welcome to enter China. He also said that Tesla has very good self-driving technology and brand, and only when more good products and technologies enter China can consumers get a better experience and let the market develop benignly.

Cao Xudong also welcomed the implementation of FSD in China. He believes that Tesla will once again play the role of good money to drive out bad money, bringing a positive impact to the industry; He also added that with his understanding of Tesla's algorithm, there is no fundamental technical obstacle to the implementation of FSD in China, and it can be trained adaptively.

Dong Zhihua, head of AVATAR's intelligent driving business, said that Tesla's FSD is undoubtedly the industry benchmark, and the public and industry people are very curious about it.

Of course, there are also giants who are not afraid of Tesla, such as Huawei, Tesla's biggest competitor in the field of intelligent driving in China.

Recently, Yu Chengdong said when talking about Tesla FSD that Huawei's intelligent driving is better than Tesla FSD even if it does not have a version with lidar, and the performance with lidar is better.

"Tesla has a lot of vehicles, a lot of data, and FSD is doing a good job. We went to the United States, Canada, etc. to do tests and comparisons, and our experience was slightly better. They didn't use lidar, we did, and we made up for the perception ability, and in China we were slightly better than them. Regarding Tesla's FSD entry into China, Yu Chengdong said that he is willing to compete, and only after competition can he see who does better.

A similar sentiment is expressed by Gansha Wu, CEO of autonomous driving technology start-up UISEE, who believes that FSD is doing well mainly because it drives "more like a human", but that FSD is not "far ahead" when measured by takeover data. During the working process of the autonomous driving system, the number of manual takeovers corresponds to the technical level of the system. The fewer the number of overheads at the same distance, or the longer the distance between two overheads on average, the better the system will perform.

Gong Min, head of automotive industry research at UBS China, said that Tesla does have advantages in terms of car ownership, data, intelligent driving hardware standards and R&D capabilities, but it cannot be directly concluded that Chinese companies will be wiped out.

He pointed out that a few years ago, when Tesla was made in China, it had a more obvious advantage over Chinese electric car companies, and the results of competition in recent years have proved that Chinese companies still have a chance to win. Gong Min emphasized that Chinese car companies have a large number of excellent engineers, and open competition will make Chinese companies progress faster.

In fact, despite Tesla's obvious technological advantages, it is not easy to say whether it will be unsuitable after it really lands in China.

For example, the "end-to-end" architecture mentioned above, despite its powerful technology, faces equally severe computing challenges. Tesla's large amount of AI GPU hardware used in the development of FSD was purchased from Nvidia, but these state-of-the-art chips have been banned from being sold to China by the United States.

Under these chip bans, if Musk has no way to ship Nvidia's most advanced chips in, then Tesla's FSD performance in China will probably be just a "castrated version" of the deformed product like the H20 chips sold by Nvidia to China.

Of course, it is not excluded that Tesla's FSD has overcome certain computing power deficiencies in the algorithm, and it can maintain a mid-to-upper level of performance, which is enough to form a dimensionality reduction blow to the intelligent driving technology of most car companies.

In short, how the Chinese version of FSD will perform after landing will most likely determine the endgame of the intelligent driving war of domestic new energy vehicle companies.

Tesla's big test

In addition to having a far-reaching impact on China's new energy vehicle industry, FSD's performance in China is also a big test for Tesla, which is under increasing pressure on performance.

According to the financial report, Tesla's revenue in the first quarter was $21.3 billion, down 8.7% year-on-year and 15.4% month-on-month. Net profit attributable to the parent company was US$1.13 billion, down 55.1% year-on-year and 85.8% month-on-month, respectively, and non-GAAP net profit attributable to the parent company was US$1.54 billion, down 47.6% year-on-year and 38.2% month-on-month, respectively.

Coupled with the previous layoffs, the financial data of the overall decline in revenue and profit in the first quarter has put a question mark on Tesla's future growth prospects.

The decline in revenue was mainly dragged down by lower volumes. In the first quarter, Tesla delivered a total of 387,000 vehicles, down 8.5% and 20.2% year-on-year, respectively, of which the wholesale sales of Tesla's Shanghai factory in the first quarter of 2024 reached 221,000 units, down 3.7% year-on-year, accounting for 57.1% of Tesla's global deliveries. In the three major auto markets of North America, Europe and China, Tesla's share has declined.

At the same time, Tesla is facing an increasingly competitive environment.

In 2023, six of the top 10 Chinese brands in the global NEV sales ranking are BYD, GAC Aion, SAIC-GM-Wuling, Li Auto, Changan Automobile, Geely Automobile, etc. Among them, BYD's sales share of the new energy market in 2023 will account for 21%, far exceeding Tesla, which accounts for 13% and ranks second.

Source: Gasgoo

In order to cope with the fierce competition in China, Tesla launched a new round of price reduction promotion on April 21, Tesla Model 3 / Y / S / X all series in Chinese mainland reduced the price by 14,000 yuan, on April 24, Tesla for Model 3 / Y further launched 0 down payment activities.

But this has also significantly affected Tesla's profitability. Since last year, Tesla's gross profit margin has continued to decline, and now it has declined for five consecutive quarters, falling to the lowest in four years, and the profit of a single car has also continued to decline.

Tesla's comprehensive gross profit margin in the first quarter was 17.4%, down 2.3 percentage points year-on-year and 0.3 percentage points month-on-month, of which the gross profit margin of the automobile business was 18.5%, down 2.6 and 0.4 percentage points year-on-year and quarter-on-quarter, respectively, if the carbon emission credit revenue of 442 million US dollars is excluded, Tesla's gross profit margin of the manufacturing end of the automobile business in the first quarter fell to 16.4%, down 2.6/0.8 percentage points year-on-year.

In the face of the decline in a number of core financial data, Tesla highlighted its future response strategy in the financial report and results conference - low-cost models, investment in AI infrastructure and commercialization in the field of autonomous driving.

Musk announced at the results meeting that the relevant models originally planned to be put into production in the second half of 2025 are expected to start production in early 25 or even late 24 years in advance, including low-cost models with new platforms, which will form a richer product line with existing models after delivery. The launch of the new model is expected to boost Tesla's sales to 3 million units a year.

However, under the haze of the industry's price war, Tesla's launch of low-cost models in the already very red sea market may only be a short-term measure to treat the symptoms but not the root cause of the growth prospects.

Obviously, from a longer-term time perspective, the progress of AI and autonomous driving is more critical for Tesla.

Currently, the use rate of FSD has exceeded 50% of the 1.8 million vehicles in the U.S. during the trial period. Cui Yan, an analyst at Minsheng Securities, believes that the implementation of the V12 end-to-end neural network algorithm will have a huge subversion of the autonomous driving industry, and the high valuation of Tesla is also expected to gradually recover.

In addition, Tesla plans to launch a Robotaxi on August 8 this year as promised, and will adopt an innovative "unboxing" manufacturing strategy.

However, Wang Shenyu, an analyst at Haitong International Securities, believes that the success or failure of Robotaxi still depends on Tesla's need to break through the two major bottlenecks of autonomous driving technology and regulatory regulations;

The implementation of unregulated FSD is expected to completely reshape the profit model and valuation system of the company and even the new energy vehicle industry.

China is Tesla's most important market, and FSD is Tesla's most important development direction in the future.

FSD's entry into China also means that the fate of the global new energy vehicle industry is turning again, will it be the outsider Tesla and Hengqiang, or the independent brand Huawei can overtake in the corner?

Resources:

Caixin "Tesla's FSD entry into China is gradually approaching, and industry professionals are looking forward to benchmarking"

Caixin "Tesla's performance declined in the first quarter, accelerating the commercialization of low-cost models and self-driving"

Golden Horn Finance "Tesla can't laugh"

View original image 215K

• 

•