It’s no secret that the United States’ relations with China have deteriorated over the last few years, over issues as varied as human rights for the Uyghurs, intellectual property theft, access to the South China Sea, and the sovereignty of Taiwan. But even in this context, the first sentence of a New York Times Magazine article from July, 2023 is eye-opening:
Last October, the United States Bureau of Industry and Security issued a document that — underneath its 139 pages of dense bureaucratic jargon and minute technical detail — amounted to a declaration of economic war on China.
What did this “declaration of economic war” revolve around? Computer chips. Specifically, high-end computer chips that facilitate the running of artificial intelligence (AI) programs and algorithms. And the US doesn’t mean to just slow China’s development of these chips — it means to strangle the Chinese chip industry.
A little background on computer chips
Computer chips are everywhere — not just your smartphone, laptop, TV and stereo, but in cars, trucks, appliances, and in everything that makes the internet and financial systems of the country run. They are also in practically everything the military uses these days.
In concept, chips are simple: they are collections of transistors — tiny (even microscopic) semiconductor on-off switches that together convert digital ones and zeros, the basic units of software, into high-level programs that accomplish real-world tasks, like converting digital radio signals into a voice on a cell phone call or images on a smartphone screen.
In practice, however, chips are incredibly complex, and getting more so each year. A single chip may have hundreds of millions to billions of transistors within it, arranged in perfect order. As it turns out, manufacturing the high-end, cutting-edge chips has become incredibly difficult. So difficult that only a tiny handful of companies in the world can make them, in highly complex factories called fabs. An example of how difficult they are to make:
[One machine critical for chip manufacture is the] extreme ultraviolet (EUV) lithography machine made by ASML, a Dutch manufacturing conglomerate, which is used to print out the layers of a chip. In 1997, ASML hired Jos Benschop, a young engineer with a Ph.D. in physics, to spearhead the creation of a new system, one that would help ASML’s customers in the semiconductor industry print smaller, faster and denser chips than ever before. It took four years to achieve the proof of concept necessary to even justify assigning a small team to the task, and then another five years for the team to build a prototype machine. In December 2010, at a research facility in South Korea, an updated prototype, a TWINSCAN NXE:3100, finally had its first successful test run. It would be nearly another decade before the first EUV-enabled products would go to market.
The newest version of the machine can craft structures as small as 10 nanometers; a human red blood cell, by comparison, is about 7,000 nanometers across. It uses a laser to create plasma 40 times hotter than the surface of the sun, which emits extreme ultraviolet light — invisible to the human eye — that is reflected onto a silicon chip by a series of mirrors. The laser is sourced from a German company and has 457,329 pieces; an entire EUV has more than 100,000 components of similar intricacy.
An EUV is just one part of the process: A cutting-edge fab can include more than 500 machines and 1,000 steps.
No one else has been able to replicate ASML’s achievement. Not even the Chinese, after years of trying.
So what did the United States Bureau of Industry and Security’s document say?
In essence, that no US microchip-related technology can be exported to China. And it defines “US microchip technology” very broadly. According to Chris Miller, author of the book “Chip War” and an associate professor of international history at the Fletcher School at Tufts University:
“The entire industry can only function with U.S. inputs . . . In every facility that’s remotely close to the cutting edge, there’s U.S. tools, U.S. design software and U.S. intellectual property throughout the process.”
. . .
[Under BIS’ new export rules,] China was cut off not just from importing the most advanced chips, but also from acquiring the inputs to develop its own advanced semiconductors and supercomputers, and even from the U.S.-origin components, technology and software that could be used to produce semiconductor-manufacturing equipment to eventually build their own fabs to make their own chips. . . . After Oct. 7, U.S. persons are no longer allowed to engage in any activity that supports the production of advanced semiconductors in China, whether by maintaining or repairing equipment in a Chinese fab, offering advice or even authorizing deliveries to a Chinese semiconductor manufacturer.
The US got the Netherlands, Japan and Taiwan — all key players in the high-end microchip supply chain — to go along. That meant that China was effectively cut off from the most important parts of that worldwide supply chain. (China was not happy about this move.)
On its face, this is an economic move that resembles the US’s decisions in 1940 and 1941 to stop exporting scrap iron, fuel, and oil to Japan (along with other economically important goods) and to freeze the dollar-yen currency exchange — a set of decisions that arguably led directly to the Pacific was in 1941.
So what was the end goal of these export controls? Why hobble China’s microchip industry?
In two words: Artificial intelligence.
Artificial intelligence (AI), as a concept and as a body of software, is not new, but it is becoming ever more pervasive in our modern economy. From biomedicine to drug development to linguistics to driverless cars and trucks to content generation, AI is a software revolution that will change the economic landscape worldwide.
But that alone does not explain all of the rationale behind the BIS’ new export controls. The other piece of the puzzle is that AI is becoming essential for the US’ military strategy.
Everyone has seen how drones have become the newest weapon in a land war; witness how Ukraine’s military has effectively used drones against not just Russian tanks and armored personnel carriers, as well as Russian troops hiding in trenches. And not just in Ukraine; Ukrainian drones have hit targets outside Ukraine, in Russia. (And conversely, autonomous Russian Lancet drones have hit targets inside Ukraine.)
Every age has its advance in military technology that proved to be a game-changer. In the Middle Ages, it was the English longbow against armored knights; more recently it has been cannon and artillery against forts, the machine gun against soldiers on foot, tanks against soldiers and machine guns in trenches, aircraft carriers against battleships and land military installations. (Perhaps ship-killing hypersonic missiles will be a similar game-changer, but that hasn’t been effectively tested yet.) This era’s game-changing military technology is likely to be drones and similar autonomous vehicles.
And not just individual drones against individual targets — multiple drones capable of mass, coordinated attacks without human “pilots”. This is exactly the point of the U.S. military’s new Replicator program. In an article in The Hill, dated August 29, 2023, Brad Dress writes that
The Pentagon announced Monday a new program focused on building out thousands of autonomous systems, including drones, as the U.S. seeks to better counter China’s vast military buildup.
The U.S. is pledging to build out the self-operating drones and defense systems at a more rapid pace to counter China’s mass procurement of conventional military resources…
The U.S. has already invested in autonomous weapons systems, including self-piloting ships and uncrewed aircraft, but Hicks said the Pentagon would now be taking it “to the next level,” producing new systems and weapons at a more aggressive pace.
These autonomous systems will be driven by AI. China has spent immense resources over the last decade building up their military, particularly the PLAN (People’s Liberation Army Navy). Current concern is that this buildup is not just an effort to establish supremacy in the South China Sea — home to critical shipping lanes — but may also be a prelude to an invasion of Taiwan. (Some speculate that China’s renewed interest in Taiwan is based not just in the history of the Republic of Taiwan, established at the end of the Chinese Civil War, but in the fact that Taiwan is home to one of the biggest chip manufacturers in the world — Taiwan Semiconductor Manufacturing Company (TSMC). TSMC produces around 40% of the world’s chips, including most of the high-end chips. It is possible that China sees an invasion of Taiwan as a way to control a key player in the worldwide chip supply chain, thereby gaining enormous leverage over the rest of the world (Some have even suggested that, should China actually invade Taiwan, the West should immediately destroy TSMC’’s fab.)
Currently, the U.S.’ main instrument of force projection is aircraft carriers and ships supporting them. However, ship-killer missiles could make carriers as irrelevant as the Prince of Wales was in WWII. Autonomous drones, on the other hand, would be a relatively cheap and easy-to-deploy force against an Chinese invasion force, or if armed appropriately, Chinese naval vessels.
China had been very active in developing its own advanced microchip and supercomputer capabilities, but the BIS’ new export rules are likely in the process of stunting those efforts.
The next few years are bound to be interesting. It would be worth paying more attention to what appears to be a new cold war between The U.S. and China.
