Don't want to overtake a corner? U.S. expansion of chip controls will hit China's self-driving industry:

After signing the $280 billion chip bill last month, the Biden administration is continuing to move to limit U.S. exports to China of advanced chips used in machine learning and artificial intelligence. Observers say the move will affect the development of China's many emerging technology industries, one of which is a huge sunrise industry: self-driving cars.

Reuters reported on Sunday (September 11), citing people familiar with the matter, that the Biden administration plans to expand export controls in October to bar U.S. companies from exporting semiconductors used in artificial intelligence and chipmaking to China.Earlier, NVIDIA and AMD received a letter from the U.S. Commerce Department requiring them to obtain permission from the Commerce Department if they plan to export chips that can be used in artificial intelligence to China in the future. Three other U.S. companies, KLA Corp, Lam Research and Applied Materials Inc., received similar letters, according to Reuters.

"Turning a letter to a single company into a rule would expand its reach, subjecting U.S. companies developing similar technologies to restrictions," Reuters said. Previously, the Commerce Department issued export control requests mostly to individual companies.

China is always looking for technologies that can overtake cars in curves, and self-driving cars are one of them,” James Lewis, director of the Strategic Science and Technology Program at the Center for Strategic and International Studies (CSIS), a think tank in Washington, told VOA. “The traditional car frontrunners Japan and Germany are not the frontrunners in autonomous driving technology, so China wants to lead the technology, but the export controls announced by the (US) will slow China down.”Frontrunners: US and China

US and China Currently leading the research and development in the field of autonomous driving.

In the US, major developers include Ford, General Motors and Tesla. The U.S. Department of Transportation is cautiously encouraging autonomous driving technology, and is still studying the possible safety issues. At the local level, Arizona allowed self-driving trucks to hit the road, while California allowed GM-developed self-driving taxis to start charging in June.

On August 9, the Biden administration signed the "2022 Chip and Science Act" (CHIPS and Science Act), a rare government high subsidy for a single industry. The bill would provide up to $52.7 billion in subsidies for U.S. semiconductor chip manufacturing and research and development, and add $200 billion to U.S. scientific research over the next decade. Analysts say the bill will power the U.S. self-driving car industry.

In China, the big names in autonomous driving include Baidu Apollo, Pony.ai, Xiaopeng Motors, Alibaba's AutoX, NIO and BYD. According to a report

last year by market analyst firm IHS Markit , driverless cars will account for 60 percent of China's ride-hailing market by 2030, reaching 1.3 trillion yuan. Baidu alone will account for 40% of the market share. Baidu CEO Robin Li said the company will expand its self-driving business to 30 cities in China by 2023, covering 3 million users. In recent years, the Chinese government has invested a lot of human, financial and material resources into the self-driving car industry. Just in the past August, intensive policy dividends hit the self-driving car industry. On August 1, Shenzhen implemented the Regulations on the Administration of Intelligent Connected Vehicles in Shenzhen Special Economic Zone , which is the first legal document in China to regulate the commercialization of driverless vehicles, laying a legal foundation for driverless technology in the city. On August 8, China's Ministry of Transport released the "Guidelines for Self-Driving Vehicle Transportation Safety Services (Trial)" (Draft for Comment), soliciting comments from the public and explicitly encouraging the commercialization of self-driving vehicles. Also in August, the governments of Wuhan and Chongqing issued a national unmanned demonstration operation qualification to Baidu, allowing the robotaxi developed by it to drive on the road and provide commercial services to citizens. On August 23, the Shanghai Municipal Government issuedThe "Shanghai Implementation Plan for Accelerating the Innovation and Development of Intelligent and Connected Vehicles" stated that by 2025, a domestic leading intelligent automatic vehicle development system will be formed, and the industrial scale will reach 500 billion yuan. Among them, the key tasks are to "break through key forward-looking technologies" and "promote key components to tackle key problems."

Lewis: Without chips, there would be no autonomous vehicles

. Lewis, a strategic technology expert, said that these core components are about chips."(The core component) is about chips and sensors, and often that's also determined by the chip. Think of a driverless car, it's a car, but at the same time the computer makes various decisions based on the sensor input, when Turning, how fast to drive,” he explained, “These require chips. Like all high-tech products, without chips, there is no self-driving car.”

Zach Edwards, an independent cybersecurity researcher, told US Voice says that these advanced artificial intelligence chips are equivalent to a brain dedicated to a certain field. "When you integrate these chips into computer networks, these artificial skills accelerator chips can solve all kinds of specific problems," he said.

At present, the largest producer of high-end chips in the world is Taiwan's TSMC. Global foundry giant TSMC accounts for more than 90 percent of global production of high-end chips, according to a 2021 Reuters analysis . The Reuters report noted that Beijing is also investing heavily, but its chip industry is "about a decade behind Taiwan" in many key areas, and that the gap will continue to widen in the coming years.

"Taiwan, South Korea and the U.S. are at the forefront of making chips in the world, and you'd be hard-pressed to find alternatives," Lewis said. "You have to think about the design of the car, you have to think about what these chips can do. China is not completely dependent on other countries, but Beijing is dependent enough on foreign countries that it will cause problems for Chinese automakers."

Shelly Palmer is president of the Palmer Group, a technology information company. He told VOA that on the US government side, the main consideration is to limit the A100 chips produced by Nvidia, which can optimize the algorithms of smart cars. He also pointed out that in addition to being used in autonomous driving technology, these high-end chips can also be used in various other high-end technology industries.

"Both (the US and China) are likely to weaponise AI chips. So really, it's part of an arms race to see who can win the AI ​​battle," he said.

Reuters reported on Sunday, citing a Commerce Department spokesman , that the department "is taking a comprehensive approach to implementing additional actionsto protect U.S. national security and foreign policy interests, including blocking China's access to U.S. technology suitable for military modernization." ."

How will China respond?

Lewis of CSIS noted that current export controls are not a complete ban. "For Nvidia, it's not a total ban, but a license (to export to China)," he said, adding that China could go to South Korea or Taiwan to find alternative manufacturers.

So for China's self-driving industry, "it's not a complete stagnation, it's a slowdown, which is an obstacle for them," he said.

Palmer Group's Palmer believes that the current US export controls are based on individual companies and cannot effectively slow down the development of China's autonomous driving field. It remains to be seen how the ban will follow.

Other experts say China is capable of dealing with U.S. export controls.

Marjory Blumenthal, a senior policy researcher at the RAND Corporation, a U.S. research think tank, told VOA that China is investing heavily in artificial intelligence.

“Policies have made AI a priority and linked to companies through measures such as designating 'national champions'. Government support in China is more comprehensive and cohesive than in the US, which contributes to China's leadership in computer vision , which is a core technology involving hardware and software," she said.

At the same time, she noted, China's investment in smart city and safe city technologies also supports the development of autonomous vehicles.

Lewis of CSIS, however, took a different view. He said it had been 20 years since the Chinese Academy of Sciences started discussing why China needed its own chip industry in 2003. In the past 20 years, South Korea and Taiwan have established their own chip empires, but China's chip industry has not developed yet.

He believes that the Chinese model has not worked in the chip industry.

The local Communist Party leaders got instructions from Beijing: We want a chip industry. So they spent a lot of money to build a new building in Guangzhou or something: chip factory No. 117. Local party leaders talked to entrepreneurs Stand in front and shake hands," he said, "but you walk into the building and there's nothing there."