Key members of NATO will say yes
Konaev and Chahal 2021 – research fellow and research analys with the Center for Security and Emerging Technology
Margarita Konaev Husanjot Chahal April 2021 “The Path of Least Resistance: Multinational Collaboration on AI for Military Logistics and Sustainment” CSET Issue Brief
This is not to say that adopting and developing, let alone collaborating on AI-enabled logistics will be an easy task for the U.S. military and allied defense organizations. The ML and deep learning algorithms behind commercial AI-enabled logistics are generally not optimized for military needs.39 And if the experience of the Department of Defense is any indication, there are multiple challenges with regards to the data needed to power AI applications—from lack of data to problems with traceability, access, and interoperability of data collected by different systems.40 Moreover, data security and privacy concerns as well as different legal frameworks for how personal data is collected, handled, processed, and stored remain a critical barrier to international collaboration. Lack of clarity surrounding how to implement the exemptions for research incorporated into the General Data Protection Regulation, for example, has stalled collaboration between the U.S. National Institutes of Health and some European counterparts.41
These and other technical barriers and privacy-related concerns are indeed significant. But developments in privacy-preserving ML techniques, including homomorphic encryption, secure multi-party computation, and federated learning offer opportunities for allies to share and pool data without compromising the privacy of individual users and organizations whose data is being used.42 The United States can also work with allies to develop technical standards and protocols for harmonizing data collection, formatting, storage, and archiving to ensure data security and integrity.43
Overall, the U.S. military and allied defense organizations will face nonnegligible technical barriers whether adapting commercial AI technologies or building AI-enabled systems and tools in-house. From a comparative standpoint, however, military logistics and sustainment applications that fall under the broader category of enterprise AI applications present “low hanging fruit” for the U.S. military (and presumably for other technologically advanced militaries).44 Moreover, international collaboration on AI-enabled military logistics and sustainment is likely more within reach than collaboration on AI integrated into weapons systems or applications that feed on sensitive data collected by proprietary weapons and sensor systems.45
With key U.S. allies like the United Kingdom, Germany, France, South Korea and Japan already pursuing efforts to leverage AI for military logistics and sustainment, collaboration in this area seems politically feasible.
The integration of AI into weapons systems has raised ethical concerns and opposition in some communities across the United States and in allied countries. Yet by focusing collaboration on AI applications for military logistics and sustainment functions, the United States and its allies could potentially sidestep the contentious “killer robots” debate. Collaborative efforts to develop and apply AI tools to areas such as defense supply chain management, personnel management, and equipment maintenance can improve existing processes and functions, save costs and increase efficiencies in defense organizations. Multinational collaboration around this set of goals and applications is less likely to galvanize widespread grassroots opposition than programs on AI-enabled drones or autonomous ground combat vehicles.
Moreover, some of the United States’ closest allies are already investing in AI and ML technologies for logistics and sustainment. The United Kingdom’s Ministry of Defense’s (MOD) Autonomy Programme, for example, identifies defense resupply and logistics challenges through the Defense and Security Accelerator as one of its key activities.46 In 2019, MOD also allocated £66 million (about $83 million) to accelerate robotic projects for the British Army, including autonomous logistics vehicles supporting resupply missions in conflict zones.47 Notably, the UK’s Defense Science and Technology Laboratory and the U.S. Army Combat Capabilities Development Command’s Ground Vehicle Systems Center have been working together since 2016 on the Coalition Assured Autonomous Resupply project, prototyping semiautonomous logistics convoys, along with ground and aerial autonomous resupply systems, and demonstrating the interoperability of the two nations’ armies with autonomous driving technology.48
France’s military AI strategy also views “logistics and operational readiness” as one of the priority areas for the defense ministry, including a focus on predictive maintenance.49 Notably, the strategy states that “mission performance and assisted maintenance applications, especially for cooperation with countries that have the same systems” as France pose no significant problems in terms of sharing classified data. And in addition to its key European partners, France is also open to collaboration with the United States given the similar approach to AI development. 50 Along similar lines, the German Army identifies AI for personnel and material management, including predictive maintenance, as one of the main areas for action on AI development.51
Japan and South Korea are also increasingly investing in military applications of AI, including for logistics and sustainment. South Korea’s National Strategy for Artificial Intelligence lists national defense as a key area for AI applications, including using AI to “quickly analyze and process large-scale defense data and develop and support common services such as medical care, logistics, and administration.”52 Meanwhile, Japan’s Acquisition, Technology and Logistics Agency (ATLA) has identified “logistical support technologies” in its medium- to long-term defense technology outlook back in 2016. More recently, ATLA has been working with private sector partners on research and development projects applying AI for defense logistics and “streamlining system maintenance work.”53