For most of the nuclear age, enhanced strategic situational awareness (SA)—the ability to characterize the operating environment, detect attacks of strategic significance (whether conventional or nuclear), and discern real attacks from false alarms—has been viewed as an unqualified benefit to strategic stability.1 By improving the accuracy and timeliness of warning, increasing visibility and clarity as to adversary actions, and extending decision time in crisis, enhanced situational awareness reduces the risk of miscalculation and alleviates the use-or-lose pressure that could incentivize a nuclear first strike. Moreover, for the conventional war fight information dominance had no imaginable downsides. But can there be too much of a good thing?
Today, potential and existing technology offers the prospect of previously unimaginable visibility into adversary actions and intentions. The combination of new sensor technologies, platforms for their deployment, high-bandwidth networks, and artificial intelligence (AI) tools are transforming the potential field of view at the conventional and strategic levels of conflict. As the strategic SA ecosystem evolves, it is becoming increasingly possible that actions to improve strategic SA may actually increase the risk of escalation and upset strategic stability. On the other hand, concerns about escalation may cause reluctance among decision-makers to use capabilities that could better illuminate a crisis and reduce the risk and consequences of war.
The transformational nature of this emerging strategic SA landscape suggests a relook is necessary to consider the risks and challenges of these emerging capabilities, especially when employed between nuclear-armed states in crisis or conflict. To effectively manage crisis escalation, decisionmakers must understand the changing relationship between mechanistic improvements to strategic SA and crisis stability.
The Traditional Strategic Situational Awareness Environment (1950-1990)
The traditional strategic SA environment was first developed during the Cold War and was focused on understanding an adversary’s nuclear systems and nuclear force posture. It consisted of early warning radars and satellites, as well as hydrophones to detect ballistic missile submarines. Specific systems in the traditional environment included the Defense Support Program (DSP) Satellites, the Ballistic Missile Early Warning System (BMEWS), and the Navy’s sound surveillance system, or SOSUS.2 Together, these systems were central to providing situational awareness in the nuclear domain.
These passive systems, as well as other systems in the traditional strategic SA environment, were viewed as stabilizing—designed to detect attacks, not to predict them. For example, early warning radars detected incoming missiles after they were launched; they were able to alert decisionmakers that an attack was incoming but could not provide enough warning to prevent an attack. Furthermore, systems in the traditional strategic SA environment often passively received information instead of actively seeking it out. For example, SOSUS was comprised of hydrophones that were fixed on the ocean floor. These hydrophones could detect an adversary submarine as it that passed their location but could not actively roam the ocean to find or track adversary assets.
Furthermore, the traditional strategic SA ecosystem was stratified. During the Cold War, strategic SA capabilities were almost exclusively focused on collecting information on nuclear systems. Conventional assets, which were not yet capable of posing a strategic threat, were monitored in times of conflict by tactical SA capabilities like manned aircraft.3 The bright line between systems used for nuclear and conventional SA reduced the possibility that inadvertent escalation due to entanglement of nuclear and conventional systems would occur.
The clear line between SA systems used for conventional and nuclear missions also meant that strategic SA assets were secure and compartmentalized. Operating at in space or remote geographic locations, these systems were traditionally difficult to target kinetically. Other parts of the strategic SA system, such as command and control, contained substantial redundancies and were expected to survive even a nuclear attack. Moreover, in the traditional SA environment, countries would have limited incentives to target strategic SA systems in a conventional conflict, as doing so would not limit an adversary’s ability to conduct conventional operations. As such, if an adversary did target strategic SA systems, the attack would likely be interpreted as a precursor to a nuclear attack, inviting immediate nuclear retaliation. Hence these systems have traditionally been considered off limits to attackers.
Generally speaking, the secure and compartmentalized nature of the traditional SA environment led to high confidence in the information these systems provided, limited their vulnerability to attack and manipulation, and reduced the chances of miscalculation. As a result, these systems came to be viewed as contributing positively to strategic stability by ensuring confidence in the durability of the overall nuclear deterrent and reducing risks of premature or miscalculated nuclear use.
The Transitional Strategic SA Environment (1990-2020)
In the traditional SA environment, critical warning-related information could be obtained without incurring significant escalatory risk, specifically, without driving parties closer to the nuclear brink. However, the security and strategic SA environments have changed dramatically in the last two decades, potentially upsetting past assumptions about the compartmentalization of nuclear and conventional SA systems and the stabilizing nature of transparency.
The security and strategic SA environments are being transformed primarily by three key trends: the rapid pace of technological innovation, the increasing likelihood of conflict at the conventional and sub-conventional level between nuclear armed states, and the increasingly dual-use4 nature of military and surveillance technology across all levels of conflict. Critically, whereas the traditional strategic SA environment contained systems that were either focused on nuclear warning (“nuclear” strategic SA systems) or on providing intelligence to commanders about the conventional battlefield (“conventional” strategic SA systems), in the transitional strategic SA environment, dual-use strategic SA capabilities may be tasked to conduct both missions.
The beginnings of this dynamic can be traced back to the 1990s and the end of the Cold War. Technological developments throughout the second half of the 20th century culminated in the networked battlefield of the Gulf War. The Gulf War saw the employment of effective communications, command, control and intelligence (C3I) which gave commanders dramatically improved situational awareness by making use of strategic systems for conventional purposes—especially in terms of precision targeting. For example, satellite systems were fully employed to give commanders in the field intelligence regarding the adversary’s troop locations, air defense and command and control installations.
The reliance on strategic SA assets in conventional conflicts is only intensifying in the transitional strategic SA environment. Advanced, long-range, and often dual-use missile systems have proliferated dramatically in recent decades, including among a range of nuclear-armed adversaries, and now must figure significantly into planning and execution of conventional conflicts. This means that adversaries may have strong incentives to target nuclear warning systems early in a crisis in order to ensure conventional dominance. However, as James Acton has argued, this type of action could easily lead the victim to believe that a nuclear strike against them was imminent, and to escalation because of this “misinterpreted warning.”5
At the same time, countries have invested in SA capabilities that were designed for conventional missions but may have utility for nuclear ones. For example, the Global Hawk was initially intended “to support joint combatant forces in worldwide peacetime, contingency and wartime operations.”6 However, as Keir Lieber and Daryl Press suggest, UAVs like the Global Hawk may also be useful to track a small country’s nuclear-tipped mobile missiles.7 This and other dual-use capabilities contribute to the blurring of lines between the conventional and nuclear sphere.
The blurring between the nuclear and conventional SA spaces is particularly dangerous because the likelihood of both conventional conflict and sub-conventional activities between nuclear-armed adversaries is growing. Nuclear-armed states are relying more on gray zone tactics such as economic coercion, cyber techniques, proxy forces, terrorism and information operations to achieve their geopolitical goals. At the same time, conflict dynamics in Syria, Ukraine, Kashmir, the Korean Peninsula, the Baltics, Taiwan, the South China Sea, and Doklam all demonstrate the potential for conventional conflict between nuclear-armed states. In both sub-conventional and conventional conflicts, nuclear armed states will rely on SA tools to gain an advantage. However, the increasingly interdependent, dual-use nature of the SA ecosystem means that the SA space may provide an unrecognized or unexpected pathway to nuclear escalation.
The Emerging Strategic Situational Awareness Environment (2020 forward)
The emerging strategic SA environment will be even more networked, dual-use and codependent than the transitional one. Distinctions or firebreaks between conventional and strategic situational awareness will all but disappear, creating a highly networked, real-time, dual-use landscape that is both murkier and more complex across all levels of conflict—sub-conventional, conventional, and strategic.
The blurring of lines between the conventional and strategic domains will only intensify as new SA systems come online. In the emerging SA environment, not only will conventional weapons rely on strategic SA assets for targeting data, countries will also rely on traditionally conventional systems for strategic warning. For example, hypersonic systems, boost-glide systems, long-range cruise missiles and other capabilities are designed to elude traditional U.S. early warning systems (i.e. radars and satellites) and thus defeat U.S. missile defenses. To counter these new delivery systems, the United States may have to rely on conventional situational awareness systems, including systems that are more visible or intrusive, to complete strategic missions and supplement strategic SA. If an adversary were to discover and target such systems, would such an attack be considered conventional or strategic in intent and implication? As such, the days of clear delineations between strategic and conventional situational awareness capabilities in ways that contribute to a sharp firebreak between conventional and strategic conflict seem limited at best.
Similarly, increasingly blurred lines between nuclear and conventional command, control, and communications also contribute to this dynamic. Conventional missile warning currently relies on these dual-use surveillance capabilities, increasing the risk that they could be targeted in a conventional conflict for conventional purposes but with profound strategic implications. For example, an attack by an adversary on an early warning system used to provide warning for both nuclear and conventional attacks could very well be interpreted as a precursor to a nuclear attack, regardless of the intent.
Moving forward, the highly networked nature of conventional systems, as well as the dual-capable nature of many of them, may increase the potential for conflict to bleed from the conventional into the nuclear realm. In the emerging strategic situational awareness ecosystem, there will be ample potential for inadvertent escalation through miscalculation. By employing an invasive capability to collect information on an adversary’s systems, actions, or intent, the very nature of that collection could trigger an unwanted response. For example, deploying unmanned underwater vehicles into a known port in an attempt to gauge whether an adversary is flushing its nuclear submarines might make an adversary suspect intent to target its sea-based deterrent, and thereby lead to a nuclear escalation. There are a number of technologies and capabilities that are contributing to the emerging environment, but each brings unique advantages and risks.
Technology’s Impact on Strategic Situational Awareness
Technological advances are transforming strategic situational awareness with the potential to provide higher quality information more quickly, more reliably, and with greater strategic effect. The new strategic SA ecosystem has the potential to help decisionmakers not only detect and react to strategic attack but possibly even predict and prevent one by applying a larger range of emerging technologies, combining them to achieve networked effects, and enabling them through advanced data management, communications, and automation. Future strategic SA architectures have the potential to provide unprecedented insight into an adversary’s capabilities, actions and intentions such that decisionmakers can not only react to crises but also anticipate them. Taken together, these capabilities offer a range of characteristics – namely vantage and range, speed, detectability, precision, persistence, and resiliency and reliability – that allow them to detail the strategic operating environment in new and transformative ways.
Characterizing the Capabilities – Benefits to Strategic SA
- Vantage and range concern the position from which new information can be ascertained. Vantage denotes a new position from which information can be gathered, while range indicates the ability to gather information from a greater or lesser distance from the target. For example, super high-altitude unmanned aircraft and pseudo-satellites provide a unique range, between that of a traditional unmanned aerial vehicle (UAV) and a satellite, with which to view the battle space. 8 Similarly, UUVs provide a new vantage from which to detect submarines.9
- Speed refers to the shortening of time between an adversary’s action or decision to act, detection of that action, and the conveyance of information to the decisionmaker. Speed is focused on shortening the observe, orient, decide and act loop to the greatest possible degree. A number of technologies focused on the rapid acquisition and analysis of information flows are aimed at increasing this rate of action. For example, AI can be used to process vast quantities of information much more quickly than otherwise possible.10
- (Un)detectability concerns the degree to which an adversary can ascertain that information is being collected. As an example, next-generation stealth capabilities could allow a UAV or manned aircraft to fly over targets within another country without being detected. Similarly, some cyber capabilities could collect information while remaining virtually undetectable.
- Precision refers to the level of detail and quality of the information collected or alternatively, a heightened degree of confidence in the information collected. Many advances in sensor technology contribute to enhanced precision. For example, hyperspectral imaging technology is useful for performing change detection and movement analysis. While an infrared or multispectral sensor can indicate objects of interest, hyperspectral imaging provides a previously unattainable level of detail including an object’s material, color, and even moisture level.11
- Persistence is concerned with the extent to which the capability can continuously collect data. For example, a pseudo-satellite gives operators the maneuverability of an unmanned system, but with the persistent coverage of a satellite: it can stay deployed for over three weeks and relay data back to operators over a prolonged period of time.12 Other capabilities include underwater unmanned vehicles (UUVs) that can loiter near adversary submarine patrol routes for long periods of time to collect data on adversary patterns of behavior.
- Resiliency and reliability refer to the ability of a technology to employ redundant and robust systems for situational awareness in a contested environment. For example, microsatellites can be launched in constellations of tens or even hundreds to complete a task previously done by one exquisite system.13 Because of the disaggregated nature of such microsatellite constellations, they are able to continue to fulfill their mission even if some individual satellites are damaged or destroyed.
Scoping the Technologies
Today’s strategic SA ecosystem includes a wide and diverse range of platforms, sensors and support capabilities that are networked through complex and high-throughput, computer driven communications architectures. New technologies enable not just the platform on which a situational awareness tool rests, but also the situational awareness tool itself as well the means by which the data is collected, transmitted and analyzed. Broadly speaking, these technologies can be sorted according to the following categories: unmanned vehicles, platform control and information support, sensors, cyber, electronic warfare, space platforms, and new materials. Each of these categories represents an area where technological and operational improvements are enabling the development of capabilities with the potential to greatly improve situational awareness via the characteristics outlined above.
Unmanned vehicles include aircraft, underwater, surface and ground vehicles that operate autonomously or semi-autonomously. Semi-autonomous vehicles, such as remotely piloted drones, can make minor adjustments autonomously, but maintain a human operator in the loop. Fully autonomous systems can adjust to changing situations and fulfill the commander’s intent without direct instructions from an operator.14 Unmanned vehicles can deliver sensors very close to targets, enabling them to provide warning and operational insight into adversary actions and intentions that cannot be gained through stand-off capabilities. Additionally, unmanned vehicles don’t place human assets directly in harm’s way, enabling them to be used in contested or dangerous environments.
Platform control and information support includes the AI systems that enable autonomy, sophisticated data analysis, and decision support. Improvements in computer processing power and massive investments in AI research and development have made AI a nascent game-changer in the strategic SA and other domains. Emerging AI capabilities will allow autonomous vehicles to complete a range of tasks—from navigating an environment without operator input to swarming, which allows numerous unmanned vehicles to share intelligence and distribute tasks amongst themselves to achieve a goal.15 AI can also be used to rapidly analyze large quantities of information—an increasingly important ability in a world where the proverbial “haystacks” that analysts must sort through to find relevant information are “growing exponentially.”16 Platform control and information support also includes robust communications links that enable transmission of information and control of strategic SA platforms even in contested environments.
Sensors are the devices or component of devices that can collect information about a target. They include everything from LIDAR and RADAR to biological and quantum sensors. While some of these sensors have existed for a while, technological advances have heightened their accuracy, precision, sensitivity, and resolution; improved their durability; and allowed for their miniaturization. These developments have expanded sensors’ applications and, in some cases, given them military utility for the first time. For example, the miniaturization of optical sensors has allowed for the development of small satellites that can capture images with high enough resolution to supplement and, depending on the mission, even replace those taken by exquisite assets.17 Furthermore, some developments in remote sensing provide entirely novel capabilities; for example, change-detection software allows synthetic aperture radar (SAR) to detect moving objects.18
Cyber capabilities include software and hardware that make it possible to gain access to and collect information from an adversary’s computer networks. This study will only focus on cyber techniques that are used to improve situational awareness; however, the authors recognize that some may consider them offensive capabilities they could potentially be used to coerce, impose harm, or degrade rival capabilities.19 Although cyberattacks and cyber espionage are not new, the increasingly networked nature of civilian and military capabilities is making them much more vulnerable to these exploits. At the same time, increased computer processing power and developments in artificial intelligence are allowing cyber espionage and cyber operations to become more useful for strategic situational awareness. By penetrating an adversary’s computer networks, cyber tools and capabilities may provide essential insight into adversary behavior, intentions, and decision-making not available through traditional means of strategic SA collection.
Electronic warfare technologies can collect information transmitted through the electromagnetic spectrum or prevent the transmission of said information. Military capabilities are increasingly reliant on radio, radar, and infrared technologies, which are used for communication, data transmission, command and control, and position, navigation and timing. As such, capabilities to collect information on or to degrade, disrupt, and deny electromagnetic emissions are becoming increasingly relevant on the networked battlefield. Countries have responded by investing in advanced offensive and defensive electronic warfare systems and by integrating these systems into every aspect of their militaries.20 Advanced electronic warfare systems can increase a countries’ strategic SA by allowing them to monitor and possibly intercept adversary communications. These technologies can also be used to degrade an opponent’s strategic SA by disrupting their electromagnetic communications or spoofing electromagnetic signals.
Space platforms are technologies deployed in space that enhance or degrade an adversary’s strategic situational awareness. In recent years, the development of commercial off-the-shelf satellite components (i.e. cubesat) and the miniaturization of sensor technologies have allowed countries and companies to more cheaply and rapidly launch space assets. The availability of easily deployed, less expensive space assets allows for redundancy or replacement of limited or vulnerable space assets and expanded coverage of high priority strategic SA targets.
New materials include synthetic materials with special characteristics that make them useful for the situational awareness mission, as well as novel platforms. New materials may enable missions that were previously impossible; for example, plant-based sensors can enable the ability to collect data in circumstances where outside monitoring would have previously been blocked by the party being monitored.21 Similarly, if new materials, such as ultra-high temperature ceramics (UHTCs) are developed, it would revolutionize hypersonic travel and allow for a hypersonic spy plane which could operate at speeds that would render it invulnerable to adversary air defenses.
These technology categories, while perhaps not comprehensive, capture the most impactful technologies in the emerging SA ecosystem. They present countries with numerous and myriad ways to improve their strategic situational awareness, and decisionmakers may be attracted to them in a crisis or conflict. However, these technologies may also introduce new stability risks; ones that may be dangerously overlooked due to the fact that historically, SA technologies haven’t been considered as dangerous or high risk.
Technology and Escalation Risks
Emerging technologies that improve situational awareness have the potential to be stabilizing if they offer decisionmakers more clarity into developing crises and reduce the possibility of miscalculation. However, these technologies may also generate new risks. Of particular concern are three escalation pathways—provocation, entanglement, and complexity based on the quality and quantity of information—that may be triggered or exacerbated by the use of emerging SA-enhancing capabilities.
The active nature of the emerging strategic SA ecosystem means that states have the capability to penetrate adversary territory (land and sea), airspace, and networks to gain highly precise and potentially actionable information. However, these capabilities are provocative because they directly challenge legal and political concepts of sovereignty, may not always be readily identifiable as surveillance (rather than offensive) assets, and may intentionally or unintentionally approach vital strategic assets as they conduct surveillance.
For example, if the United States used a UAV to collect information within North Korean airspace, it could introduce significant escalatory potential. The mere presence of a U.S. asset in North Korean airspace could trigger retaliation if it was detected; the likelihood of escalation would increase if North Korea was unable to determine that the asset was a surveillance platform. Furthermore, if the UUV was operating near a missile launch facility, North Korea might assume that its nuclear assets were being targeted. In a crisis, this could create a strong incentive for North Korea to disperse and even use its nuclear weapons early.
Furthermore, these concerns are particularly acute in the cyber domain, where the line between espionage and attack is very thin (as the same techniques can be used to deploy cyber probes or to launch cyberattacks). As such, the victims of cyber espionage may be particularly inclined to interpret any intrusion into their networks as an attack or the precursor to an attack. For example, China believes that the United States may attempt to use a cyber-attack to prevent adversaries from launching nuclear weapons in a crisis. This type of cyber-attack would require constant cyberespionage during peacetime to identify vulnerabilities and ensure that previously identified vulnerabilities had not been patched.22 With this in mind, China might feel especially threatened if they detected any type of cyber probing of their nuclear command, control, and communications systems early in a crisis. This could create an intense incentive to escalate a crisis, either by using nuclear weapons before they can be compromised, or by escalating within the conventional realm.
Furthermore, the rapid, precise, persistent, and predictive nature of emerging SA capabilities also increases the chance for escalation through provocation by potentially creating new opportunities for a counterforce strike, increasing the first mover-incentive and undermining crisis stability. For example, Keir Lieber and Daryl Press suggest that the United States could plausibly use a combination of SAR-equipped satellites, standoff UAVs and penetrating UAVs to track North Korean mobile missiles.23 U.S. decisionmakers might be tempted to conduct a counterforce strike if they believed this was possible; however, the consequences of miscalculation would be dire, as North Korea might use remaining nuclear weapons against the United States or allies in the region in retaliation. Furthermore, if North Korea believed that the United States had the capabilities to carry out a counterforce strike, it could be incentivized to use its nuclear weapons early in a crisis.
The blended or dual-use nature of the emerging SA ecosystem contributes to the potential for escalation through entanglement. As defined by James Acton, entanglement occurs when nuclear delivery systems, forces, and support structures are co-mingled, or when non-nuclear weapons are able to threaten nuclear weapons and their command, control, communication, and information systems (C3I).24 Entanglement in the strategic SA space occurs when conventional SA systems intentionally or unintentionally collect information on nuclear assets, or when dual-use SA systems become military targets during a conventional conflict. Entanglement can lead to escalation by convincing one or more countries in a crisis that their nuclear assets are at risk.
First, since many of the SA capabilities in the emerging ecosystem are inherently dual use, they cannot exclusively collect information on conventional military assets. As such, targeted states must assume that when these assets are deployed, they are collecting information on both nuclear and conventional military assets. This may cause targeted states, especially those that co-mingle their nuclear and conventional forces, to fear that an adversary is preparing for a counterforce strike (either conventional or nuclear) against its nuclear assets, and to respond in an escalatory manner. For instance, Caitlin Talmadge highlights that in a U.S.-China crisis, powerful conventional SA capabilities and forces “can threaten nuclear forces in ways that generate pressures to escalate,” particularly when the target possesses a relatively small and fragile nuclear arsenal.25 Additionally, advanced strategic SA tools combined with increased conventional capabilities may allow for non-nuclear weapons state counter nuclear force targeting.
Similarly, many of the strategic SA capabilities that support nuclear weapons, either directly or indirectly, are also dual use. In a conventional crisis, a state may target these capabilities in order to “blind” the adversary and gain an advantage. For example, U.S. early warning radars both provide the U.S. with warning of an incoming nuclear strike and support U.S. missile defense capabilities, including short range theater defense systems.26 In a conventional conflict, a U.S. adversary might be incentivized to target these satellites to ensure the success of short- and medium-range conventional missile attacks. However, the United States would likely interpret an attack on an early warning satellite as preparation for a nuclear strike, inviting U.S. escalation to the nuclear level.
Finally, both the quantity and quality of information generated by the emerging strategic SA ecosystem have the potential to cause escalation in surprising ways. In the national security space, it is widely assumed that more and better information leads to better decisionmaking. However, this may not always be the case. The technologies in the emerging strategic SA ecosystem have the potential to provide vast amounts of information; however, this information must be analyzed and distilled in a way that is useful. While it may be possible for AI to assist human analysts with this task, the fact remains that the right questions have to be asked in order to render information useful.
Even if processed and presented to decisionmakers appropriately, more information does not necessarily lead to better decisionmaking. Some recent studies have suggested that instead of improving decisionmaking, more information can lead to decision paralysis, or cause decisionmakers to wait too long to make a decision in the hope that they will get a final, conclusive piece of evidence.27
Furthermore, the ambiguous and unproven nature of some of the new streams of strategic SA may lead decisionmakers to discount vital information if they don’t trust the source. For instance, many experts remain cautious about relying on artificial intelligence because AI systems cannot always explain to a human how they arrived at their conclusions. Furthermore, decisionmakers may not trust AI even if it is “explainable.” AI derives some of its unique advantage from being able to recognize patterns that human analysts cannot, but if the indicators that an AI system cites do not match a decisionmaker’s idea of relevant indicators, they may dismiss it.
Finally, the emerging strategic SA ecosystem has the potential to provide decisionmakers with highly precise or anticipatory information that could make previously untenable courses of action possible. For example, a state that detected an adversary preparing its nuclear assets and was confident that it knew the location of all of that country’s nuclear assets could launch a preemptive counterforce strike. At the same time, if a country believed that it was vulnerable to such a strike, it could be incentivized to use its nuclear weapons early in a crisis.
Stability Risk Factors
With these escalation pathways in mind it is possible to illuminate the key features or attributes of strategic SA technologies that may undermine strategic stability and increase the risk of escalation through one of the pathways described above. These factors include: intrusiveness, destructiveness, predictiveness, preemptive-ness, dual-use nature, clandestine nature, vulnerability, and action-enabling nature.
- Intrusiveness describes the extent to which a capability must enter into an adversary’s territory, airspace or networks. Some capabilities, such as UAVs, have the potential to be more intrusive, as information gathering might require violating an adversary’s airspace. Similarly, cyber techniques are often incredibly intrusive even though they don’t involve physically entering enemy territory.
- Destructiveness refers to the extent to which a capability destroys/degrades an enemy system, either temporarily or permanently, in achieving its objective. For example, kinetic anti-satellite weapons are highly destructive, while laser dazzling (a technology that temporarily “blinds” a satellite) is only somewhat destructive.
- Predictiveness assesses the degree to which a capability allows a state to anticipate adversary actions as opposed to merely reacting to them. For example, AI applications that use predictive analytics have the potential to unlock new insights and enhance strategic SA. While potentially valuable, predictive capabilities generate risks for stability because they may create a strong incentive for a country to take action to prevent an undesirable outcome. Furthermore, predictive capabilities may not always be right: AI-enabled systems can misinterpret signals and are vulnerable to “poisoned” data which can provide false information and affect the veracity of a system’s output. Reliance on predictive capabilities could thus lead countries to unnecessarily take escalatory actions.
- Preemptive-ness describes the extent to which a capability supports the capability to react to adversary actions before they can be completed. The AI-enabled predictive system described above is also a preemptive capability; however, capabilities can be preemptive without being predictive. For example, a system of satellites and UAVs that could reliably track mobile missiles could also be preemptive if it allowed a country to detect unusual activities and conduct a preemptive strike if they believed an attack was imminent.
- The dual-use nature of a capability refers to the extent it is used for conventional and nuclear missions. The dual-use nature of emerging technologies can create confusion as to the intentions of the surveilling party. For example, if UUVs are used to observe an adversary’s conventional submarines, which might be housed alongside its SSBNs, the surveilled state would be unable to tell which assets were being targeted and may deem their nuclear assets under threat.
- Clandestine capabilities, as defined by Austin Long and Brendan Green, are those that derive significant military advantage by being kept secret.28 Many cyber techniques derive benefit from being kept a secret, because once a country is aware of an adversary’s presence in its networks, it can take steps to mitigate the damage or provide false information.
- Vulnerability assesses the degree to which an adversary can deny the use of a capability. Adversaries are likely to disrupt or destroy strategic SA capabilities that are more vulnerable, thereby cutting off the flow of information. If one country in a conflict suddenly loses access to one or multiple situational awareness capabilities, opportunities for escalation through miscalculation may ensue.
- The Action-enabling nature of a capability refers to the degree to which it enables new military options. For example, a capability that allows the tracking of mobile missiles might enable a country to target all of its adversary’s nuclear weapons, a dynamic that currently does not exist for any nuclear dyad. This dynamic might create new risks as it allows for the finding out of information that was elusive under the traditional ecosystem.
Risk vs. Reward
The characteristics and risk factors identified by this project provide concepts and tools to better understand the potential risks and rewards associated with emerging strategic SA capabilities. In doing so, they may help analysts and decisionmakers identify strategic SA capabilities that can significantly improve situational awareness while mitigating escalation risks. For example, a smallsatellite constellation improves resiliency/reliability and persistence, and provides increased “vantage with the ability to ascertain new information concerning adversary space capabilities across low earth orbit (LEO), geosynchronous orbit (GEO), and highly elliptical orbit (HEO).”29 However, smallsats may be predictive, vulnerable, and dual-use. Furthermore, cyberespionage improves speed, resiliency/reliability, precision, persistence, and in some cases undetectability; but it is intrusive, predictive, preemptive, action-enabling, clandestine, and dual use, and can be vulnerable. Depending on a decisionmakers’ objectives and risk tolerance one capability may be appropriate while the other is not. As such, the characteristics and risk factors can help decisionmakers understand the potential impact of these capabilities on both situational awareness and stability to leverage them to successfully navigate future crises.
The strategic SA capabilities in the transitional and emerging strategic situational awareness ecosystem have the potential to provide more timely warning of attacks, unprecedented visibility into adversary actions and decisionmaking, and to extend decisionmaking time in a crisis. However, some capabilities may also have the potential to trigger escalation through provocation, entanglement, and information complexity. The goal of this work is not to discourage technical innovation or suggest the voluntary disarmament of valuable strategic SA capabilities. Rather, it is to seek to better understand the ways in which these capabilities improve strategic SA and crisis management and to better illuminate attendant risk pathways.
Tactical/operational situational awareness support to the warfighter is distinct from strategic situational awareness and is not addressed in this study. ↩
For more information on each of these systems, see: Guntar Krebs, “DSP 1,2,3,4 (Phase 1),” Guntar’s Space Page, Accessed May 23, 2019, https://space.skyrocket.de/doc_sdat/dsp-1.htm; “Ballistic Missile Early Warning System (BMEWS)” Global Security, https://www.globalsecurity.org/space/systems/bmews.htm; and University of Rhode Island, “The Cold War: A History of SOund SUrveillance System (SOSUS),” The Discovery of Sound in the Sea, https://dosits.org/people-and-sound/history-of-underwater-acoustics/the-cold-war-history-of-the-sound-surveillance-system-sosus/. ↩
Tyler W. Morton, Eyes in the Sky: The Evolution of Manned Airborne ISR, (Montgomery, AL: School of Advanced Air and Space Studies, 2012), https://apps.dtic.mil/dtic/tr/fulltext/u2/1019401.pdf, 89-90, 121-145. ↩
Unless otherwise specified, dual-use refers to whether a technology can be employed for both conventional and nuclear purposes. ↩
James Acton, “Escalation through Entanglement: How the Vulnerability of Command-and-Control Systems Raises the Risks of an Inadvertent Nuclear War,” International Security 43, no. 1 (2018):58. ↩
“RQ-4 Global Hawk,” United States Air Force, October 27, 2014, https://www.af.mil/About-Us/Fact-Sheets/Display/Article/104516/rq-4-global-hawk/. ↩
Keir Lieber and Daryl Press, “The New Era of Counterforce,” International Security 41, no. 4 (2017): 37-46. ↩
Pedro Vincente Valdez and Paulina Wheeler, “High Altitude Pseudosatellites,” On the Radar, July 10, 2019, https://ontheradar.csis.org/issue-briefs/high-altitude-pseudosatellites/. ↩
Bethany Goldblum and Andrew Reddie, “Unmanned Underwater Vehicles for Submarine Detection,” On the Radar, May 5, 2019, https://ontheradar.csis.org/issue-briefs/unmanned-underwater-vehicle-uuv-systems-for-submarine-detection-a-technology-primer/. ↩
Lizamaria Arias and Nate Frierson, “Artificial Intelligence Analysis Applications,” On the Radar, April 3, 2019, https://ontheradar.csis.org/issue-briefs/artificial-intelligence-analysis-applications-a-technology-primer/. ↩
Bernadette Stadler and Meyer Thalheimer, “Hyperspectral Imaging,” On the Radar, November 15, 2018, https://ontheradar.csis.org/issue-briefs/hyperspectral-imaging-a-technology-primer/. ↩
Pedro Vincente Valdez and Paulina Wheeler, “High Altitude Pseudosatellites.” ↩
Bethany Goldblum and Andrew Reddie, “Smallsats,” On the Radar, May 4, 2018, https://ontheradar.csis.org/issue-briefs/smallsats-a-technology-primer/. ↩
Daniel Wassmuth and Dave Blair, “Loyal Wingman, Flocking and Swarming: New Models of Distributed Airpower,” War on the Rocks, February 21, 2018, https://warontherocks.com/2018/02/loyal-wingman-flocking-swarming-new-models-distributed-airpower/. ↩
Amy Zegart and Michael Morrell, “Spies, Lies and Algorithms: Why U.S. Intelligence Agencies Must Adapt or Fail,” Foreign Affairs 98, no. 3 (May/June 2019), https://www.foreignaffairs.com/articles/2019-04-16/spies-lies-and-algorithms. ↩
For example, SkySat, a commercial smallsatellite operated by PlanetLabs, has a resolution of 72 cm2, as compared to the open-source global standard of 31 cm2 achieved by DigitalGlobe’s WorldView-3 and WorldView4 satellites. See “Planet Imagery and Archive,” Planet, accessed July 24, 2019, https://www.planet.com/products/planet-imagery/; and Melissa Hanham and Jeffrey Lewis, “Remote Sensing for Arms Control and Disarmament Verification,” Federation of American Scientists, accessed July 24, 2019, https://fas.org/wp-content/uploads/media/Remote-Sensing-Analysis-for-Arms-Control-and-Disarmament-Verification.pdf. ↩
Lieber and Press,”The New Era of Counterforce,” 38-39. ↩
The nature of cyber—whether its offensive or defensive, for example—depends on the objectives of an operation. See: Belk and Noyes, “On the Use of Offensive Cyber Capabilities: A Policy Analysis on Offensive US Cyber Policy,” Belfer Center, (March 2012). ↩
For example, Russia is deeply integrating electronic warfare capabilities through it’s military. See: Roger McDermott, “Russia’s Electronic Warfare Capabilities to 2025: Challenging NATO in the Electromagnetic Spectrum,” International Centre for Defence and Security, (Tallinn: International Centre for Defence and Security, September 2017) 2-11, https://icds.ee/wp-content/uploads/2018/ICDS_Report_Russias_Electronic_Warfare_to_2025.pdf. ↩
Anthony Benjamin, ” Plant-based Sensors,” On the Radar, October 12, 2018, https://ontheradar.csis.org/ issue-briefs/plant-based-sensors-a-technology-primer/. ↩
James Acton, ed. Entanglement (Washington, D.C.: Carnegie Endowment for International Peace, 2017), 5. ↩
Lieber and Press, “The New Era of Counterforce,” 37-46. ↩
James Acton, Entanglement. ↩
Caitlin Talmadge, “Would China Go Nuclear? Assessing the Risk of Chinese Nuclear Escalation in a Conventional War with the United States,” International Security, 41 no. 4 (Spring 2017), 50-92. ↩
James Acton, Entanglement, 51. ↩
Austin Long and Brendan Green, “Invisible Doomsday Machines: The Challenge of Clandestine Capabilities in World Politics,” War on the Rocks, December 15, 2017, https://warontherocks.com/2017/12/invisible-doomsday-machines-challenge-clandestine-capabilities-deterrence/. ↩
Goldblum and Reddie, “Smallsats,”4. ↩