The End of Deception

Vice Admiral Horatio Nelson once said, “A ship’s a fool to fight a fort,” but in a potential future war with China, that’s what U.S. ships would be required to do when they entered the second island chain and came within range of China’s coordinated sensor networks and arsenal of land-based antiship missiles.¹ These shore-based missile sites, like forts of old, have greater range, firepower, and resiliency than their sea-based counterparts; Chinese missiles are capable of striking targets more than 3,000 kilometers away, outranging both the ship-based Tomahawk land-attack missile (1,700 kilometers) and U.S. carrier air wings (1,000 kilometers).² Furthermore, their land-based nature means they can be rapidly replenished. 

An effective Chinese reconnaissance–strike complex would decimate technologically advanced but fragile U.S. warships far before they were within range to carry out a strike. As a result, if required to destroy China’s shore-based missile sites, the U.S. Navy likely would rely on the few stealthy platforms in the current inventory to move in undetected and strike key Chinese targets prior to bringing in the rest of the fleet. Until now, the Navy has been poised to accomplish this because of its monopoly on stealth capabilities, but advances in commercial technology, when leveraged by near-peer militaries, are rapidly eroding the effectiveness of stealth. Ultimately, the Navy must rethink how it structures and trains its forces. 

How We Hide

U.S. precision strike warfare in denied environments hinges on stealth. During the initial phase of the First Gulf War, for example, stealthy Air Force F-117s delivered a first strike against Iraqi air defenses to provide clear skies for less stealthy fighter-bombers.³ In a potential conflict with China, one can imagine an analogous scenario where the stealthiest U.S. assets—Virginia- and Ohio-class submarines and Zumwalt-class destroyers—attempt to sneak undetected into the first island chain to deliver a first strike against Chinese radar and missile sites, clearing the way for conventional forces to launch follow-on strikes.

U.S. submarines and Zumwalt-class destroyers could use traditional emissions-control techniques—set sail under the cover of night, operate autonomously without calling back to headquarters, and steer visually rather than using radar—to decrease their probability of detection before launching a critical first strike. Traditional techniques to avoid detection, however, are effective only against traditional forms of scouting. Turning off a ship’s lights at night might fool nearby ships, and shutting off radars and other electromagnetic emission sources, using radar-absorbing materials, or spoofing your automatic identification system signal to represent yourself as a fishing vessel can minimize the chances of being found by more advanced electronic means. But there is no hiding from satellites. 

Eyes in the Sky

The astronomical price of highly capable satellites has meant no military has been able to maintain enough assets in orbit to achieve broad-area, persistent surveillance—especially over large ocean areas. But recent advances by the commercial sector are causing a paradigm shift. Companies such as Planet have launched constellations of hundreds of small, inexpensive satellites that collect images of every square foot of the globe multiple times per day, providing persistent surveillance capabilities for pennies on the dollar.⁴ In addition, companies have deployed signals intelligence (SigInt) satellites that are able to detect minute radio emissions and synthetic aperture radar (SAR) satellites that are capable of seeing objects on the surface of the earth during the day, at night, and in all weather conditions.⁵ 

Of course, all that space-based data would be useless without a way to process the imagery, fuse multiple sources, and sort the signal from the noise. The ocean is vast, and locating a ship might require looking through thousands of images of open ocean. In the past, human analysts had to pore over those images and then look across all sources (imagery, SigInt, open source, etc.) to confirm their hypotheses.⁶ Artificial intelligence (AI) eliminates the need for armies of analysts and drastically accelerates the speed at which analysis can be completed, all while enabling a much more robust fusion of multiple, disparate data sources. 

AI can sift through millions of square miles of SAR, electro-optical, and other images—as well as open source data points such as fuel or supply orders in ports—in minutes to quickly locate and track a ship, even in the vast expanse of the Pacific. In addition, because the proliferation of satellites and advances in AI largely are driven by the private sector and academia, where research is shared, any nation can replicate or purchase capabilities once reserved for global superpowers—at a fraction of the price. 

Run Silent, Run Deep

Submarines, once nearly immune from detection, also are in danger. Advances in underwater acoustic sensor networks and acoustic detection algorithms are improving militaries’ ability to detect and identify submarines hiding below the surface.

Imagine you are hosting a party. Friends are chatting and an Amazon Echo is playing background music. You want to switch up what’s playing and say, “Alexa, play Hall and Oates.” The Echo detects a voice saying “Alexa” among all the party noise then understands, or “classifies,” the specific acoustic signal from your voice—“play Hall and Oates.” Similar technological principles can be applied to acoustic submarine detection.

The ocean contains a lot of noise: some of it natural, like biologics; and some mechanical, such as the turning of a ship’s propellers or the running of engines. Detecting a specific piece of equipment is analogous to the Echo hearing “Alexa” at a party. Figuring out the equipment type is a classification problem, similar to the Echo understanding what “play Hall and Oates” means. Navies collect sound signatures of each other’s submarines and ships and thus already have the data needed to train a machine-learning acoustic detection system.⁷ 

Although using algorithms to classify sounds already is achievable, the sensors need to be relatively close to the submarine. This makes detecting submarines in the waters beyond the second island chain unlikely. Once the operational area shrinks, however, sensor networks can be deployed much more effectively. For example, sonobuoys and towed arrays could be combined with highly capable military-specific autonomous systems—such as the Office of Naval Research’s Sea Hunter and Bluefish’s Knifefish—and commercial advances in distributed maritime sensor networks. Saildrone, a start-up company already under contract with the Navy, offers a fleet of persistent, solar-powered, unmanned surface vehicles that easily could be fitted with acoustic sensors or towed arrays.⁸ This surface capability will be enhanced by the proliferation of long-endurance autonomous underwater assets powered by recent commercial leaps in battery technology.  

The U.S. Navy must prepare for a world where commercial SAR satellites can detect and track ships anywhere, anytime, with precision and ease and SigInt satellites can recognize even the smallest radio emission or breach of emissions control. These new technologies are leading to the extinction of stealth. 

Destroying Enemy Defenses

If maritime stealth no longer exists in its current form, stealth-enhanced surface ships and guided-missile submarines may not be able to perform their vital function of eliminating the first line of enemy defenses with a “hard kill.” This leaves a second option: disabling the enemy’s first line of defense by leveraging nonkinetic means.

To do so, the U.S. military likely will rely on its capabilities in the electromagnetic spectrum—cyberspace. The 2007 spoofing of Syrian air defenses, which allowed Israeli fighter aircraft to sneak past and attack a suspected nuclear materials site, proves this practice can be used effectively on the battlefield.⁹ Assuming the Navy has the capabilities to accomplish such an attack, however, conducting a battle damage assessment will be challenging. It will be hard to know with certainty that a cyber attack was successful in disabling enemy defenses.¹⁰ Furthermore, a cyber attack may not be repeatable if an enemy applies software updates or security patches. U.S. leaders would be hesitant to put high-value assets, such as a carrier strike group, in harm’s way without a high degree of certainty that Chinese defenses were down. 

If neither stealth nor electromagnetic operations can guarantee an effective first strike, the Navy’s current force posture will not secure victory at sea. In rethinking the way it fights, the Navy must turn to distributed maritime operations. 

A Fleet for the Future Fight

In a world where everything can be seen, the advantage lies in having the ability to overwhelm the enemy. This means distributing lethality. 

Today’s ships are mind-bogglingly expensive—a single guided-missile destroyer costs more than $1.8 billion to build and equip—and fragile.¹¹ A single enemy missile that achieves a mission kill on a destroyer’s fire-control radar could effectively render its nearly one hundred missiles useless. Dispersing the U.S. Navy’s existing missile loads to more shooters would force China to defend against missile attacks from more locations while simultaneously needing to attack more ships. In addition, each Chinese missile hit would be less effective, since each U.S. shooter would represent a smaller percentage of U.S. firepower. 

In this fleet construct, high-end ships such as destroyers and cruisers should operate as command-and-control platforms, controlling swarms of unmanned air, surface, and subsurface vehicles deployed by attached amphibious assault ships, which already have the well decks and storage capacity necessary to facilitate rapid deployment of large numbers of autonomous vehicles. The drones would operate alongside traditional platforms, providing increased firepower, distributing the force, and making it hard for China to deliver a debilitating blow against a center of gravity. 

A strike group of five to ten ships could deploy drones just outside the range of enemy missiles, essentially creating 50 or more unmanned surface missile platforms, hundreds of autonomous airborne drones, and dozens of underwater unmanned vehicles to screen against enemy submarine activity. In the words of Rear Admiral Ronald Boxall, former director of Surface Warfare, these drones would form a “network of armed nodes that [China] has to deal with” so China is forced to contend with “the entire system: not just [a] ship, not just [a] strike group.”¹² 

In the future, stand-alone autonomous platforms that are able to transit with strike groups across oceans could be employed, increasing the overall size and capability of the fighting force. The size of this force, its geographic spread, and the dispersion of offensive capabilities would make it more difficult for China to prioritize targets, defeat their defenses, and inflict catastrophic damage across the entire network. 

Strategy and Infrastructure

The 2018 National Defense Strategy (NDS) Commission contends the Pentagon’s current strategy “does not articulate clear approaches to succeeding in peacetime competition or wartime conflict,” and “absent a more integrated strategy, the United States is unlikely to reverse its rivals’ momentum across an evolving, complex, spectrum of competition.”¹³ The Navy must lay the groundwork for a future conflict now or risk ceding supremacy. 

First, the Navy must articulate a technology strategy and operationalize it. China released its artificial intelligence national plan in 2017, aiming to become a world leader in artificial intelligence, machine learning, and autonomy by 2025.¹⁴ It already has started a $500 million venture fund in Silicon Valley focused on early stage technology related to autonomy and artificial intelligence and is working to develop a home-grown microchip industry.¹⁵ In developing its strategy, the Navy must reform its acquisition processes to capitalize on and integrate technical advances into the fleet faster than its adversaries.  

Second, the Navy must develop a pro-innovation mind-set and actively engage in efforts that will drive the future of the force. The Army and Air Force have “leaned in” with ambitious innovation projects such as the Army Futures Command and Air Force Kessel Run, and the Navy’s February launch of NavalX to connect innovators throughout the fleet seems promising.¹⁶ But while Strategy for Maintaining Maritime Superiority 2.0 discusses leveraging innovative technologies, the Navy has disbanded internal organizations such as the Chief of Naval Operations’ Rapid Innovation Cell and the Strategic Studies Group without replacement. 

Third, the Navy must build the technical infrastructure to ensure AI advances can be integrated rapidly into existing systems. While building a resilient and robust cloud environment or labeling data used to teach algorithms across the Navy might not please Congress as much as a new destroyer, it might be more impactful in the long run by developing the foundation for future technical capabilities in both autonomy and AI. 

Finally, the Navy must educate its people. Fleet-wide technical training is needed to ensure warfighters are ready to deal with future conflict. Technology must drive tactics, and while they need not be experts, sailors and officers must be familiar with trends and implications of cyber operations, AI, and autonomy. This can be achieved by expanding the Secretary of the Navy Tours with Industry program to expose junior officers to the latest technology, having specific weapons and tactics instructors programs focused on how to integrate new technologies into the fleet, and increasing the number of shore-based billets at Tenth Fleet and the National Security Agency. If the fleet is armed with knowledge of the latest technology, it can drive tactical changes from the lowest levels. 

The Navy has relied on its technical superiority, particularly the ability to elude detection, to fight and win. The rapid proliferation of sensors paired with growth in artificial intelligence, however, means stealth is quickly disappearing. The Navy must adopt new technologies and approaches, using warships as command-and-control platforms to direct fleets of drones on, above, and below the sea to strike effectively first. This approach is technologically possible, but the Navy will not be able to capitalize on it without cultural and infrastructure changes.