Geopolitical Power Play: AI Unleashed
As tensions between the US and China grow amid the rise of AI, the associated geopolitical risks are also swelling at a commensurate rate.
The rapid advancement of artificial intelligence technology brings forward a myriad of political and military risks. AI has the potential to influence public opinion, manipulate information through fake news and deepfakes, and escalate global threats if autonomous weapons systems fall into the hands of terrorists or rogue agents.
AI’s ability to provide tactical superiority in decision support, predictive maintenance, and target recognition also presents an existential threat on the military front. These same systems may respond by launching preemptive attacks and risk escalation as a calculated move in the geopolitical game of chess.
However, the greatest concern is as US-China tensions rise, the urgency to use AI with military applications will rise and create a constellation of novel risks. Both superpowers are currently engaged in what I call the Promethean War: a competition to secure hegemon-defining technologies to ensure global preeminence.
As a result, Washington and Beijing are engaging in a tech-for-tat trade war. The former has imposed restrictions of AI-interfacing semiconductors (like Nvidia’s AI100 chip) and the latter responded by restricting exports of key minerals used in strategic industries with commercial and military applications.
Regulatory Problems of AI
When it comes to AI global governance, the regulatory landscape is becoming increasingly balkanized amid a breakdown in global geopolitical relations. As a result, a lack of cross-continental coordination and alignment on policy may lead to asymmetric risks politically, economically, and financially.
Both the US and China have their own markets, customers, standards, technologies, and producers. Consequently, a divergence in policy will increase the costs of businesses trying to operate in each country’s techno-economic zone. But the US-Sino rivalry is not exactly inspiring hope for cross-Pacific collaboration.
Ironically, the US and China have collaborated more on AI policy than any other two nations, but that trend is now decelerating. A research paper from Stanford concluded that:
“Overall, U.S.-China collaborations on AI research have quintupled since 2010 and totaled 9,660 papers in 2021—much faster than the increase in collaborations between any other two nations[…but] the volume of published U.S.-China collaborations in AI has declined slightly from its peak in 2019”. This comes as no surprise: relations around that time were rapidly deteriorating amid the US-China trade war.
AI systems are being implemented across and within industries without adequate legal oversight or thorough consideration of their ethical implications. This discrepancy, commonly known as the pacing problem, has overwhelmed legislatures and executive branches, rendering them incapable of effectively addressing the issue.
China is a bit of a petri dish when it comes to AI governance, and their rivalry with the US adds another variable to the deployment of their regulatory regime. As a result, both superpowers are “de-risking” from each other ( i.e. decoupling) across numerous aspects of the value chain of AI-interfacing technologies.
The unraveling of relations, combined with the lack of standardization is a risk investors will now have to incorporate into their investment strategies. For example, there remains no clear international definition of what constitutes AI in the context of outbound investment from the US to China.
Washington has often repeated that it is designing a “high fence built around a small yard” in its Promethean War with China; that is, narrow and targeted. But that policy itself hinges on the definition of what constitutes AI, and which applications/hardware could be used for malicious purposes by a foreign adversary.
One clear example of a security threat - and surgical approach to remedying it - is the US restricting China’s access to Nvidia’s A100 and H100 GPU chips. Both are highly advanced, AI-interfacing technologies that could threaten the US’s position as the global leader in artificial intelligence if China was able to produce them at scale.
According to a recent report by Reuters, China has found a loophole in acquiring small quantities of these chips through alternative means. These avenues include importing the chips from other Asian nations such as India, Taiwan, or Singapore, as well as resorting to black markets.
But Beijing is not getting enough to match the computing power necessary for achieving technological preeminence as outlined in their targeted industrial policies.
The Promethean War
While the US leads in AI research and its deployment in products and services, China leads in specific areas of application such as facial recognition. The United States continues to receive more AI patents than any other nation, but China is now filing more than half of all the world’s patent applications in the field.
But the US should not rest on its laurels, for history is unforgiving. The same paper published in Stanford University concluded that:
“history has shown that the full economic impact of historic tech advances stemmed less from which nation pioneered a technology than from which ones were best at applying it across a broad range of industries”.
China is approximately 2-3 years behind the US in AI research, and faces a number of a headwinds. One major problem is how can China harmonize its stringent Great Firewall of censorship with the information-generating capabilities of generative AI?
Beijing’s top-down political economy will likely clash with its global leadership ambitions. And this is where the competition for hegemonic rivalry and intersection with AI becomes interesting - and dangerous.
US-China Military Confrontation & AI Deployment
Within the context of a military confrontation with the US, China cannot win through brute force. However, the advent of AI has led China to reconceptualize warfare to what is called “systems confrontation”.
A paper from Georgetown University defined it as “a duel between opposing military operating systems, with the center of gravity being the information architecture”. In short, data-dependent. AI’s ability to process vast amounts of data makes it therefore instrumental in the People’s Liberation Army’s (PLA) military strategy.
The framework moves away from wars of attrition that are limited by geography, physics, and manpower; And instead, systems confrontation leverages AI and data to analyze and exploit critical vulnerabilities in your adversary’s digital architecture. The RAND Corporation outlines this theory in greater detail below:
“Chinese military publications indicate that the PLA has recognized that war is no longer a contest between particular units, arms, services, or even specific weapons platforms of competing adversaries, but rather a contest among numerous adversarial operational systems. This mode of fighting is unique to modern warfare, as are the battlefields on which conflict is waged. This is referred to as systems confrontation [体系对抗]…[and] is waged not only in the traditional physical domains of land, sea, and air, but also in outer space, nonphysical cyberspace, electromagnetic, and even psychological domains. Whereas achieving dominance in one or a few of the physical domains was sufficient for war fighting success in the past, systems confrontation requires that “comprehensive dominance” be achieved in all domains or battlefields.”
The vast computing power of AI and its abilities are therefore indispensable to China’s military strategy, and is heavily dependent on the centralization of data. As such, it becomes easier to understand why Washington is limiting the exportation of strategic technologies to an adversary who intends to wield it against them.
The current geopolitical climate and the policies that are being deployed in it reflect a gradual shift from targeted economic measures to AI-enabled military operations. This final stage of the Promethean War would mark a direct conflict between the US and China, with AI as both their swords and shields.