AI and National Security: Is the Future of War in the Hands of Silicon Valley?

• In Event: Policy Complexity of Technology and Innovation

Abstract

Growing consensus among civil innovators and government leaders indicate that artificial intelligence (AI) will critically shape the future of work, society, military affairs, and international relations. As the revolutionary potential of AI is increasingly recognized across private and public sectors, policymakers and military planners are scrambling to better understand how AI might change the balance of military power and warfare. Recently, Russian President Vladimir Putin proclaimed that the leader of AI development will “rule the world” and China pledged to invest $150 billion to become the world’s AI innovation center. With world powers entering what may be the next arms race for AI, it is increasingly critical to study the implications AI will have for national security and global power dynamics.
Research in this area has advanced theories of diffusion of military technology and examined how such diffusion has the potential to restructure power relations on the international stage (Gilli & Gilli, 2016; Horowitz, 2010; Goldman, 2006; Goldman & Eliason, 2003). In general, these studies indicate that theories about military innovations must continue to be adapted to the advent of AI, which still remains in early developmental stages. However, the existing research on the popular commercialization of AI and its relation to the military application of AI has not been thoroughly addressed in the literature. In this regard, this paper will examine the Public Private Partnerships’ (PPP) impact on the military’s adoption of AI. For this purpose, we will focus on the defense contracts with the industry’s leaders, such as Google and Microsoft. Some existing literature has noted the historical effects of commercial incentives on military development, namely that rapid diffusion of technology in the commercial sector makes it difficult for countries to gain sustainable advantages militarily (Herrera, 2006). Yet, research specific to AI remains scarce. Emerging studies are focusing more on how concurrent development of AI in the private sector could crucially limit or expand advantages for existing military powers (Horowitz, 2018).
As the commercial sector increasingly draws the best and brightest AI engineers away from the defense sector, more research must shed light on how commercial investments in AI will impact military innovation, and how private technology companies will play a role in the future of national security. In this context, Public Private Partnerships play a significant role. PPPs have transformed “the nature” of governance in various ways. These collaborations combine “public purpose and private authority” that bring together multiple actors such as firms, foundations, and governments. (Andonova, 2018). PPPs have been implemented in public policy areas ranging from of public transportation to environmental and waste-disposal services, and AI development has been no exception: both in the US and in Europe industry partners, academic institutions, and governments collaborate to advance AI. However, PPP faces hurdles in this area due to the contentiousness that AI development leads to, especially in the case of military sector.
One of the first direct applications of AI in the US military is Project Maven. The Department of Defense (DoD) created the Algorithmic Warfare Cross-Functional Team to oversee the development of Project Maven in collaboration with Google. Project Maven focuses on computer vision to study imagery used to improve drone strikes. Google employees responded with a petition to pressure Google to drop the project. Google announced that it will not renew their contract with the DoD, and will not collaborate in the future. The leaders and employees of Google sent a strong signal: “Google should not be in the business of war.”
The Pentagon is highly dependent on tech companies for the development and implementation of AI on the battlefield. As one DoD official explained, “There is no ‘black box’ that delivers the AI system the government needs, at least not now… Key elements have to be put together … and the only way to do that is with commercial partners alongside us.” ((https://dod.defense.gov/News/Article/Article/1254719/project-maven-to-deploy-computer-algorithms-to-war-zone-by-years-end/). This relationship is one the DoD has a significant stake in creating and fostering into the future. But the efficiency and efficacy of PPP in AI is still developing. Based on elite and expert interviews, this research will make a significant contribution to the literature by exploring how these partnerships will shape the future of national security.

Authors

• Lena Trabucco, Northwestern University

• Merih Angin, Koç University