Information in War: Military Innovation, Battle Networks, and the Future of Artificial Intelligence

  • Published

Information in War: Military Innovation, Battle Networks, and the Future of Artificial Intelligence by Benjamin M. Jensen, Christopher Whyte, and Scott Cuomo. Georgetown University Press, 2022, 266 pp.

Unlike its title suggests, Information in War largely provides interesting case studies of innovations made prior to war to suggest how the US military can prepare to use artificial intelligence (AI) to enhance current battle networks. The authors organize their work along four broad case studies of advancements in battle networks spanning about one hundred years: radar; the Semi-Automatic Ground Environment (SAGE) computerized air defense system; the more diffuse revolution in military affairs (RMA); and the global battle network, best thought of as the kill chain that arose during the Global War on Terror, particularly with the employment of the MQ-1 Predator. Those readers assuming they will be reading mainly about AI should know that Information in War instead focuses largely on these historical case studies, which rely primarily but not entirely on synthetic research into secondary sources.

This book had its genesis in 2015 when senior military officers asked Marine Corps Lieutenant Colonel Scott Cuomo to dig deeply into artificial intelligence by “review[ing] new AI military programs and defin[ing] how they would change the future of warfare” (xi). Professors Dr. Benjamin M. Jensen of Marine Corps University and Dr. Christopher Whyte of Virginia Commonwealth University subsequently came on board to support the project.

New technology does not guarantee military innovation. As such, central to the work is the notion that information cannot contribute effectively to warfare unless a battle network exists to allow the “fusing [of] disparate pieces of information to efficiently allocate scare resources” (6). Artificial intelligence will be unable to do this effectively if the US military assumes that the “commercial development of narrow AI applications will translate into new ways of warfighting” (45). Rather, it must cultivate the open sharing of ideas between a wide-range of individuals and organizations, even if that poses a challenge to current ideas about warfighting.

Borrowing from AI studies pioneer Claude Shannon’s theory of information to provide a theoretical framework, the authors first stress the importance of channel capacity, or the amount of information that can be transmitted. This factor posits that the “more information in circulation, the higher the potential for change” (30). The authors contrast closed organizational legacy structures—which include legacy software systems and the organizations that support them—to open ones, with the latter allowing information to flow throughout.

 Second, the authors borrow the idea of source coding, or the notion that those considering the adoption of new technology need “schema” to help them understand what is being proposed in a way that it becomes tacit knowledge. Within source theory, there are two categories: high or low resonance. High resonance enables an institution to meld new information technology with warfighting employment ideas, but innovation still may not occur because of “existing organizational design and legacy structure.” In a case of low resonance, the prospect of change can also be stymied by legacy structures in addition to a military unreceptive to changing its warfighting ideas in the first place (44). The authors’ differentiation between the two levels of resonance could be clearer, as could much of the theoretical section that precedes the case studies. Readers may find it helpful to review the authors’ key findings (on 194–195) to understand the theoretical framework before beginning the case studies.

Air Force readers will likely appreciate the first case study on how Great Britain, unlike the United States or France, successfully innovated and built a battle network thanks to advancements in radar technology. Whereas the use of the battle network during the Battle of Britain often receives the most attention, the authors here focus on pre-wartime developments. This case study shows successful innovation through a cadre of dedicated military officers and scientists who had the opportunity to thrive because “malleable institutions [gave] way to a tightknit cohort of persons with a collective vision of an alternative air defense strategy” (77). Sometimes, though, the actors become lost in the more dominant tendency of the case study to provide a clear explanation of how radar technology advanced, however interesting that might be.

Given the 43 pages devoted to radar, it is disappointing that the case study on SAGE, which allowed for North America’s air defense, receives a mere 12-and-a-half pages. This innovation worked, according to the authors, because the Air Force, as a newly independent service, had an “open structure.” SAGE also supposedly reaffirmed extant notions of ground control interception (103). Yet the authors simply assert that because the Air Force was “new” that it must have been “more open to new formations and missions” without providing compelling evidence. Readers familiar with the literature on the cognitive dissonance of Air Force leadership will likely be unconvinced. In the context of battle networks, however, the authors explain how SAGE represented the first digital battle network. Previously, humans had mapped and made decisions based on radar returns. Now, however, SAGE’s users would maintain oversight of an automated system that could not only identify aircraft but also send out aircraft to intercept incoming bombers (107).

Next, the authors fast forward to the revolution in military affairs, exploring advancements in information technology in both the US Army and the US Marine Corps. The key contribution that information made was to ensure that precision strike came to eclipse “massed formations” in importance (116). The authors offer the interesting point that the RMA, contrary to popular opinion, did not arise primarily from the DoD think tank, the Office of Net Assessment. Indeed, they point out how unusual it is for a sole individual or singular organization to merit credit for novel concepts or developments. They instead argue that these ideas develop out of “transnational, cross-service networks that connect defense policy officials in a continuing process of anticipation, debate and synthesis” (123). This idea is a compelling one, although some counterexamples may exist. The concept of Joint-all domain command and control, for example, arguably originated in a white paper between the Marine Corps and the Army, perhaps partially to undercut the Air-Sea Battle Concept dominated by the Air Force and the Navy.

This RMA chapter does not quite align with the other case studies, in part because it is not precisely clear how this represents a revolution in information technology; indeed, the authors acknowledge that change in this period can better be characterized as evolutionary, so framing this chapter as an RMA is somewhat confusing (118). For example, the authors appear to critique Marine Corps Commandant General Charles Krulak—in what reads more like an accidental compliment—for being “more attentive to building a force capable of adapting to the changing character of war than on integrating information technologies associated with the RMA” (144). Taking one further step back, the authors also do not explain clearly enough how the digital battle network transformed between SAGE and the so-called RMA. What is interesting, however, is how they highlight the difference in reactions between Army and Marine Corps leadership, with Army leadership embracing it even as the Corps rejected much of it.

In the work’s final case study, the authors blend the first global battle networks with developments in remotely piloted aircraft (RPAs). The US military could successfully innovate because of how high-level “demand” for these capabilities converged with those cognizant in information technology, which again required a “diverse set of connections across the defense establishment” (157). More unclear, however, is why the authors believe they are solving a mystery in explaining how the Predator came to be armed, since that story is well known (165).[1] And, although fascinating for its discussion of cultural blind spots in the airpower communities of both the Air Force and the Marine Corps, the chapter is very platform-centric on the Predator at the expense of a broader, holistic treatment of the global battle networks themselves. Moreover, the chapter relies on Richard Whittle’s Predator: The Secret Origins of the Drone Revolution (2015) for about 40 percent of its sources, which seems excessive.

Drawing on all of these case studies, the authors pull together the past and the future in their conclusion, returning again to the subject of AI. They argue that “military applications of AI will succeed best in environments where diverse, military and civilian, constituencies are allowed to experiment and challenge prevailing ideas about warfare” (193). While their stress on the value that “contrarians” bring to military institutions is refreshing, their conclusion is not particularly argumentative as it is difficult to craft a compelling argument for the converse: that technology succeeds where small, isolated groups cannot experiment or challenge dominant narratives about how best to employ force (217). Still, this book provides much to consider regarding the challenge of successfully innovating and adapting AI into the military.

Heather P. Venable, PhD

[1] See, for example, Walter J. Boyne, “How the Predator Grew Teeth,” Air & Space Forces Magazine, July 1, 2009, https://www.airandspaceforces.com/.

"The views expressed are those of the author(s) and do not reflect the official policy or position of the US government or the Department of Defense."