Defense Investment in Artificial Intelligence

Back in April, the Congressional Research Service produced a report for legislators orienting them on the issues currently surrounding artificial intelligence (AI) and military operations. The report, titled “Artificial Intelligence and National Security,” raised a number of concerns that the CRS believes Congress ought to consider when debating oversight of the Department of Defense’s investment in and use of AI for military purposes. Among these concerns is the unpredictability of AI, as well as its vulnerability to manipulation (i.e., hacking), because the DOD is investigating how AI can “facilitate autonomous operations [and develop] more informed military decision-making.” As the report’s authors point out, Great Power competition with China and Russia will continue to be the major driver of DOD investment. The think it is incumbent upon the United States Government, therefore, to develop a viable strategy for investment in and use of AI that ensures sufficient legal oversight.

DOD’s AI Investment

Congressional interest in DOD’s use of AI comes none too soon as the department’s investment in AI-related technology has increased 44% since fiscal 2017. According to Deltek’s analysis of the DOD’s Budget Request for Fiscal Year 2019, the portion of defense programs using AI or machine learning technology will grow from $1.25B in FY 2017 to more than $1.8B in FY 2019. The Research, Development, Test, and Evaluation (RDT&E) and Procurement programs included in this total do not dedicate their entire budgets to AI investment, but they do intend to use the technology in one way or another. Parsing out exactly what the DOD will spend on AI is impossible given the way that programs report their annual budget requests. Despite the lack of clarity, investment is trending up and increased spending attracts Congressional interest.

Broken out by Defense organization we can see that the highest levels of spending have been at Air Force in recent years. This makes sense given the service’s involvement in Project Maven and it’s use of remotely piloted surveillance vehicles. Investment is also ramping up significantly at the Defense Advanced Research Projects Agency (DARPA), which also makes sense given the priority allocated to AI in the 2018 National Defense Strategy. DARPA (and the Office of the Secretary of Defense, for that matter) tend to take the lead in researching cutting edge technologies for the DOD as a whole.  

Navy’s employment of AI spans the range from informing decision making to creating autonomous sea-going vehicles, both surface and undersea. The Army is following a similar path, expressing interest in AI for its next generation combat vehicle, which can operate either manned or unmanned. Army’s AI investment also focuses on the intersection between electronic warfare, cyber, and positioning, navigation, and timing (PNT). Lessons learned from the Ukraine’s ongoing conflict with Russia are being leveraged to determine the nature of a potential future threat to U.S. forces.

The revelation of active Congressional interest in DOD’s AI investment validates the fact that rising levels of funding are going to the technology. AI and machine learning may remain somewhat niche technologies in the near-term, but tens of millions of dollars are now flowing into it, offering business opportunities for those companies that are correctly positioned in the market.