The Biden Administration is pursuing a two-pronged strategy to slow, or ideally halt, China’s domestic development of its advanced semiconductor sector. Its central strategy is to impose unprecedented, broad and unilateral restrictions on U.S. exports of advanced semiconductor technology to China.
Underpinning this core approach is the ad hoc tactic of periodically pestering major semiconductor countries to comply with U.S. export controls so that their semiconductor companies cannot easily grab the large market share in China that U.S. semiconductor companies would have to give up.
In January of this year, I predicted that this outreach tactic would fail, and in April, that’s what happened.
The Netherlands and Japan rejected requests by U.S. officials to tighten export controls because the two major players in the semiconductor supply chain wanted additional time to evaluate existing export controls and to determine who would win the upcoming U.S. presidential election. Germany’s response was neutral, neither supporting nor rejecting the requests by U.S. officials.
The crickets have been chirping ever since.
That leaves only the core tactic, which is self-defeating in itself: Competing semiconductor companies in these three major semiconductor countries, along with South Korea and Taiwan, will simply soak up the U.S. semiconductor market share abandoned in China. In the embarrassing parts of the advanced semiconductor supply chain where no semiconductor companies are present in the U.S. at all, sales to China will still be viable.
China is a “pace challenge” for the US As mentioned above, semiconductors are not just another important technology, or even a first-class technology among peers. Semiconductors are uniquely positioned as a first-class technology because they are the foundation upon which all other advanced technologies are built.
Unfortunately, the Biden Administration doesn’t even have a strategy for this most advanced technology, and headlines like “Blacklisted Chinese Chipmaker SMIC Becomes World’s Second-Largest Pure-Play Foundry by Revenue, Surpassing GlobalFoundries, etc.” demonstrate the real-world consequences of this lack of strategy.
As if this were not already dire enough, certain advanced technological weapons from China, such as the hypersonic missiles we face today and the autonomous killer robots we may face in the future, have many additional physical components that are physical objects and therefore subject to U.S. or other U.S.-allied export controls. But with artificial intelligence, once you have high-end silicon, all you need is a software algorithm and a large data set.
Technically, such advanced AI technology would likely be subject to export controls from the United States or other U.S. allies, but because no physical goods are involved, blocking the transfer of such AI technology would pose real and practical challenges that are nearly insurmountable. All that is needed to transport this AI technology to China is a power outlet, a laptop, and internet access.
This is especially true when the AI software is open source rather than proprietary. For example, Meta Platforms, which owns Facebook, has adopted an open source AI software business model that purportedly includes national security safeguards. However, this has not stopped China from using this advanced AI software as the basis for most of its homegrown AI models. This makes China’s advanced semiconductor situation even more important as a last line of defense.
Thus, if China is successful in purchasing advanced semiconductor technology from other countries that the United States has banned its own companies from selling to China, or that it would ban from companies located in China, we can expect to see Chinese cyber attacks, biological weapons, robotic submarines, warships, fighter jets, swarms of aerial drones, and ground combat vehicles powered by AI.
And of course, now that AI is considered a watershed moment in human history on a par with fire, electricity, the internet, and nuclear weapons, its impact on the economies of the United States and its allies vis-à-vis China cannot be overstated. As historian Michael Mastanduno succinctly explains, “military power rests on a foundation of economic power, both qualitatively and quantitatively.”
During the second presidential debate in 1992, Ross Perot famously said that the North American Free Trade Agreement would put a “giant suck” of U.S. jobs in Mexico. China is putting a second giant suck by buying as much advanced semiconductor technology as it can from America’s overseas competitors. This time, China is putting a giant suck because there are no export control treaties; there is only a failed voluntary export control regime known as the Wassenaar Arrangement, of which Russia is the most notorious member.
So it is a bit of an exaggeration to say that if we don’t get our semiconductor export control strategy right, little else matters when it comes to the technological arms race with China.
The deepening cooperation between China and Russia, especially regarding semiconductors, does not limit the US semiconductor export control strategy to China. Since Russia’s full-scale invasion of Ukraine in 2022, more than 80% of the semiconductors purchased by Russia have been imported directly from China. This shocking figure is another example of how important it is for us to have a strategy, and to have the right strategy. This deep connection between China and Russia in semiconductors is another arrow in the quiver of US authorities for European countries such as the Netherlands and Germany, which have been ambivalent about strengthening semiconductor export controls to China.
I have long argued that we need a semiconductor export control strategy, not a failed and self-defeating tactic. Our relationships with the Netherlands, Germany, South Korea, Japan and Taiwan are multifaceted and deep. There is a lot of potential deal-making going on if we were to commit to a binding semiconductor export control treaty between our countries that focuses on China, Russia, North Korea and Iran.
The People’s Liberation Army recently surrounded Taiwan, Matsu, Kinmen and other small islands as a “punishment” for the inauguration of Taiwan’s democratically elected President Lai Ching-te. According to Taiwanese military experts, this military exercise by the People’s Liberation Army is the first to simulate a full-scale attack, not an economic blockade. There is no time to waste.
André Brunel is an international technology lawyer at Reiter, Brunel & Dunn. This commentary was written by his article It was published in the Journal of Business & Technology Law. The views and opinions expressed in this commentary are his own and do not necessarily reflect the views or positions of the law firm or its clients.
