The Western Response to Weaponized Technology and AI

An independent commission in the U.S. that was tasked with making recommendations for the U.S. President and Congress on how to advance the development of AI and machine learning to address the national security needs of the United States completed its two-year long study. The findings made quite a splash when the commission co-chairs, former CEO of Google Eric Schmidt and former Deputy Secretary of Defense Robert Work wrote, “America is not prepared to defend or compete in the AI era… China’s plans, resources, and progress should concern all Americans. We take seriously China’s ambition to surpass the United States as the world’s AI leader within a decade.”

The U.S. and China are by far the world leaders in AI technology. But the Communist Party wants China to be the world leader in AI by 2030, and has committed to spending $150 billion to achieve that goal. The concern is not just that China is investing in AI, but how it has already weaponized the technology for political control. AI is also an effective tool for spreading disinformation quickly by generating automated, carefully tailored and personalized disinformation campaigns through social media platforms. 

This is not only about the Chinese government using this technology on its own citizens, which is a concern in its own right. Research by the Carnegie Endowment for International Peace and Security found at least 25 different countries where Chinese AI technology is already being deployed, touching every continent.

There are at least 25 different countries where Chinese AI technology is already being deployed, touching every continent.

The question is how can Western democracies not only counter or limit these dangerous applications, but also harness the good that exists and form a proactive agenda. Here are three important steps.

First, is an investment in the intellectual infrastructure needed to drive tech policy and harness AI. This includes the necessary investments in engineering and computer science. But beyond the technical and scientific investments, it’s critical to build the ties between policymakers and the scientific world. There is a cavernous gap between those working in the sciences and the political leaders who govern, regulate, and often fund relevant research. A better understanding among our leaders in government, especially in the national security space, will be important in both building the resources we need and understanding how to use them. 

Second, western democracies should work together to develop the international guidelines and norms around the use of new and emerging technologies, particularly AI. The West has largely forgotten how to think big and work together to forge the rules of the road. But it was the transatlantic alliance that largely built the post-World War II international architecture that has governed for the past 70 years. It is what built the norms around the use of nuclear technology and the prohibition of chemical weapons.

Finally, Western democracies can invest in the societal resilience necessary to withstand and counter the weaponization of AI or other technologies for disruptive or political purposes. Civil society can play a crucial role to help democracies resist authoritarian tools of surveillance. Organizations focused on issues like privacy, human rights, public health, and free speech can help spot and communicate to the public some of the threats and challenges AI technologies can pose. Investments in building these organizations both at home and abroad will go a long way to counter the effects of malicious uses of new technology.

Much of the conversation around AI developments and uses by China and other authoritarian states is alarming and anxiety-inducing. But democracies have survived through previous technological shifts, including existential, planet ending weaponry obtained by Joseph Stalin, one of the most ruthless authoritarians in history. They can do so again, with the right investments, resources, and maybe some creative thinking.