Publisher’s Note: This column is the latest in a series by Don Lenihan exploring the issues around the use of AI, including the social, economic and governance implications. To see earlier instalments in the series, click here.
Is an AI Cold War imminent? TikTok could be an early warning signal. Critics say it gives the Chinese government access to sensitive user data, but the issue goes far beyond data and privacy. It's about China's use of artificial intelligence (AI) to advance authoritarian goals at home and abroad. It's about turning already-rising tensions into a new Cold War.
The New Cold War Narrative and the US Response
China’s willingness to defy global conventions is well-known. For instance, the Sharp Eyes program uses AI and facial recognition to monitor millions of citizens. And as Canadians know, China uses AI tools like ChatGPT to spread disinformation abroad, especially during elections.
In the West, this blend of ideological authoritarianism and advancing AI capabilities evokes Cold War memories. Many see China as the new Soviet Union seeking global dominance and want to restrict its capacity to use AI for this end. This is not just rhetoric.
In 2022, the Biden administration banned exports of semiconductors and chip-making technology to China. When Chinese leader Xi Jinping asked “Why?” Biden was blunt about the motives and the goal: “Because you use [the technology] for all the wrong reasons, so you’re not going to get those advanced computer chips.” In other words, the US aims to limit China’s progress on AI while the West advances.
Two years later, the world looks closer than ever to a new Cold War and an all-out AI arms race. To understand the tensions and the stakes, let’s consider three key points.
Taiwan: The Epicenter
First, Taiwan is the epicenter of these tensions. It produces 60% of the world’s semiconductors and 90% of the most advanced chips. Given China’s claim that Taiwan is part of its territory, Xi views Biden’s ban as an insult and a challenge to China’s sovereignty. Unsurprisingly, China’s military exercises around Taiwan have escalated along with Xi’s nationalist rhetoric. The West also depends on Taiwan for these chips and has pledged to protect its interests, ensuring Taiwan has the military resources it needs (the “silicon shield”). If an AI Cold War is coming, Taiwan is the new Berlin.
Effectiveness of Sanctions and China's Goals
Second, the sanctions are effective. China struggles to access crucial chips and faces significant manufacturing hurdles. As an observer noted, making advanced chips is “insanely hard,” and China “is not even close.” This inability seriously threatens Xi's plans for China.
Xi is committed to making China a global AI leader by 2030, a goal he is on track to achieve. In the past year, China has released state-of-the-art AI systems, including Baidu's Ernie chatbot, Vidu’s text-to-video system, Unitree’s advanced humanoid robots, and fleets of self-driving cars. These high-quality AI systems are closing the gap with their Western counterparts.
Domestically, China is also surging ahead. It is a world leader in the adoption and use of AI in health and education, and Xi has ambitious plans to leverage AI for other goals, including promoting food security, economic equality, smart cities, green technology, and transforming China’s economy into a digital ecosystem. However, Xi “needs the smallest and fastest chips to fulfill his dream of transforming China into a technology powerhouse” (The Atlantic). Without them, China’s progress could stall.
An AI Arms Race?
Third, the West is deeply engaged in its own AI race. Tech giants like OpenAI, Microsoft, Google, and Meta fiercely compete to dominate the global AI market, projected to reach $15 trillion by 2030. While there is consensus that this race shouldn't compromise public safety and ethics, the resurgence of the Cold War narrative is unsettling these priorities. Calls to slow AI development for safety concerns are routinely weighed against the fear that if the West pauses, China could overtake it. Thus, the idea of an AI race for market share is merging with an AI arms race to preserve the West’s geopolitical security.
Beyond the AI Arms Race
In hindsight, Biden’s ban is reactive rather than forward-thinking. It draws inspiration from the Cold War “containment strategy” where the US sought to confine the Soviet Union to its “sphere of influence.” Similarly, by slowing China’s AI development, Biden aims to limit its global market reach, effectively confining China to its own digital space. It’s a new twist on an old idea.
While it appears to be working, it is a temporary solution at best. One way or another, China will get the microchips it needs, and when it does, all its energy will go into catching up. What then?
Likely, a mad dash to win the AI arms race – and that is disturbing. No one knows where the finish line lies, but AI safety should always be the top priority. A runaway AI arms race where both sides ignore safety to get to the finish line first is a formula for disaster that rivals Mutual Assured Destruction (MADD).
In the end, the Cold War narrative, with its binary worldview and strategies of containment, belongs to a bygone era. It misrepresents the nature of digital technologies and AI. Unlike ballistic missiles, AI doesn’t sit in its silos waiting to be launched. It is inherently dynamic, constantly seeking to cross boundaries and connect. A strategy aiming to confine China to its digital sphere or exclude it from the global AI market is both impractical and misguided. If two global spheres emerge, they won’t remain separate for long.
To deal with China effectively, the West needs a strategy for the digital age, one that combines significant economic and political pressure with ongoing dialogue on AI safety and market cooperation. The good news is that there is lots of common ground. Both sides want to leverage AI for economic and social goals, from healthcare to green technology. Western countries, including Canada, could benefit from China's experiences in these areas and China wants to be part of the global AI market. There is a basis for progress.
To open the way, the West needs to back away from the Cold War narrative, champion responsible AI development, and invite dialogue on the benefits of global alignment and cooperation. Might Canada find a significant role in such a strategy?
Don Lenihan PhD is an expert in public engagement with a long-standing focus on how digital technologies are transforming societies, governments, and governance. This column appears weekly.