Introduction

In a July 2020 report, New America identified 38 states with armed drone programmes, 11 of which had used armed drones in combat.^[1] Twenty-eight more states have programmes in development.^[2] Global exports enable the rapid proliferation of drones, as states share drone technology with others. The US, Israel, and China have been the largest exporters.^[3] Iran has also helped non-state allies—Hezbollah, Hamas, and the Houthi rebels—acquire drones.^[4] The expanding, broad commercial drone industry and simple do-it-yourself ingenuity have also helped the Islamic State and various lone-wolf actors acquire and use them.^[5]

Single drones are increasingly being integrated into collaborative drone swarms. Precisely defined, drone swarms are “multiple unmanned systems capable of coordinating their actions to accomplish shared objectives.”^[a][6] These are proliferating quickly as well. Armenia, China, France, India, Israel, the Netherlands, Russia, Spain, South Africa, South Korea, and the US all have drone swarm programmes under development.^[7] In May 2021, Israel became the first state to use a drone swarm in combat, collecting and relaying information on Hamas militant locations for follow-up attacks.^[8] In most cases, the drone swarm consists of homogeneous, aerial drones, such as the 103 Perdix drone that the US Strategic Capabilities office launched out of three F/A-18 Super Hornets in January 2017.^[9] Others, such as Elbit System’s Torch-X, integrate diverse ground and aerial vehicles.^[10]

Although global proliferation may be likely, drone swarm proliferation should not be expected to be even or immediate. Some states may be capable of rapidly developing massive, armed drone swarms; for others, this is a farther possibility. Over time, the holdouts may change their views as they come to recognise the military value of drone swarms. Motivated states can also put their finger on the scale, implementing national export controls to limit transfers of drone swarms to states without them. Groups of states might work together to expand the scope of those restrictions, and encourage larger international norms around drone swarm development, use, and transfer.

Issues Around Drone Swarm Proliferation

Drone swarms are a technological innovation. Any military innovation requires the financial resources to buy or build the technology and organisational capacity to incorporate changes into recruiting, training, and operations.^[11] Incorporating drone swarms into larger operations requires identifying and assessing the most effective use cases to know where the technology is of the most value, and incorporating those uses into doctrine. Success depends on leadership delegations of authority, effective control over training, and independent doctrinal assessment mechanisms.^[12] For drone swarms, the unique organisational challenges are likely to revolve around sustaining drone mass over time in an area of operation, requiring integrated production, logistics, and sustainment systems. States may differ in how they do that and how well. For example, in the US, doctrine frequently drives technology, while in China, it would be the reverse.^[13]

Drone swarms are a sub-type of drones, so the proliferation dynamics of drones apply. For example, hobbyist, mid-size military and commercial, large military-specific, and stealth combat drones will all differ in how quickly and broadly they proliferate.^[14] Although hobbyist drones are broadly and easily available for only a few thousand dollars, the US appears to be the only operator of stealth drones.^[b],[15] Organisational and laboratory resources place constraints on state and non-state actor’s ability to create drones.^[16] Adding swarming must necessarily increase the constraints: a state developing a swarm of stealth combat drones has all the constraints of building the stealth combat drone, and the additional challenge of integrating them into a swarm. However, that increase will depend on the system in question. MIT engineering students designed the Perdix drone swarm, suggesting a simple drone swarm requires relatively limited capability.^[17] However, a swarm of MQ-9 Reaper drones would require not only the capability and resources to build an MQ-9, but the capability to build, deploy, and test the software and hardware needed to integrate them. Some factors that have encouraged drone proliferation, such as reduced risk to human operators, will similarly apply to drone swarms.^[18] Drone swarms probably reduce the immediate operator risk even more because their use necessarily implies larger adoption of drones.

The Demand for Drone Swarms

Demand drives the proliferation of drone swarms. States must perceive that they gain some meaningful economic, political, tactical, strategic, or other benefit from the use of drone swarms. The advantages of drone swarms include cheap mass, limited operator requirements, distributed complexity, and applicability to various military missions. Additionally, those gains must be greater than any economic, political, tactical, or strategic risks and requirements. In the case of drone swarms, those include fears of remote warfare, questions over their reliability, countermeasures from adversaries, and building and integrating necessary infrastructure and support systems.^[19]

Drivers and Advantages

Cheap mass: Drone swarms allow militaries to generate and manage large numbers of drones. In a 2012 master’s thesis at the Naval Postgraduate School, Loc Pham and co-authors found that eight drones were enough to overwhelm naval destroyer vessels.^[20] Typically, four drones would hit the ship.^[21] Cheap mass is useful not only for overwhelming defences but also for depleting munitions over time and fixing platforms in undesirable locations. In the ongoing war in Ukraine, Russia has employed cheap Shahed-131 and 136 drones extensively in attacks against Ukrainian critical infrastructure. Although the drones were shot down at high rates, the attacks still forced Ukraine to deploy air defences to protect infrastructure instead of forward-deployed units waging the war.^[22] If defenders employ expensive surface-to-air missiles to shoot down cheap drones, missile stocks will not be available for use against more valuable targets like manned aircraft.

Limited operator requirements: As drone swarms scale in size, they must necessarily be autonomous. General John Murray of the US Army Futures Command once said, “When you have little drones operating in different patterns and formations, all talking to each other and staying in sync with one another...imagine that with the ability to create lethal effects on the battlefield. There is no human who will be able to keep up with that.”^[23] That may mean no personnel needed to operate and manage the swarm. Instead, humans may be relegated to the task of oversight, including high-level command and control, supporting logistics and launch, and providing maintenance or support activities.

Distributed complexity: Drone swarms can integrate multiple capabilities and spread over a broad area. Combined arms effectiveness can be baked into the DNA of a drone swarm. A single drone swarm may contain multiple payloads operating together: sophisticated sensors to identify and track targets, electronic warfare equipment for jamming, and bombs and missiles for carrying out strikes.^[24] A couple of drones might suppress the target with electronic warfare, while other drones coordinate kinetic attacks from multiple axes simultaneously. Drone-to-drone communication allows drones to distribute widely to coordinate intelligence gathering and searches, providing situational awareness for follow-on attacks.

Broad military applications: Drone swarms can be used in a broad range of missions, from undersea warfare to countering anti-access/area-denial and amphibious warfare. Drones could be distributed throughout the ocean, identifying and tracking adversary submarines, bringing greater transparency to the ocean.^[25] That could create stability concerns by reducing the viability of ocean-based second-strike deterrence measures. The US also sees great value in drone swarms for defeating Chinese anti-access area-denial measures, using mass drones to exhaust missile defence magazines, suppress or destroy defensive sites, and protect more valuable manned assets.^[26] Although the value of drone swarms to any mission will depend on the state and their conflict environment, the broad range of applications means drone swarms can help many states tackle their security challenges.

Challenges and Risks

Fears of remote warfare: Some states hesitate to use armed drones out of fear that countries could more readily engage in violence and war because the costs of loss are lower.^[27] Others note that the combination of decreased risk to soldiers and physical distance from the battlefield makes discrimination between combatants and non-combatants more difficult—this issue is critical in the laws of war.^[28] For example, Germany only allowed military forces to deploy unarmed drones until April 2022, when the war between Russia and Ukraine nearby spurred a rethinking along with a broader defence bump.^[29] If states are unwilling to use armed drones, they certainly would be unwilling to use numerous drones networked together, especially if drone swarms exacerbate the ethical concerns. German concerns were associated with drone use during the War on Terror, where drone strikes were relatively narrow against pre-identified targets. Thousands of drones deployed against thousands of targets would exacerbate concerns about remote warfare thousands of times.

Reliability questions: As drone swarms scale in size, the swarm may become ever more unpredictable and unreliable in practice. In a truly massive drone swarm, a human operator could not plausibly maintain direct operational control over the swarm. This creates at least three reliability concerns. First, autonomous target selection and engagement using current artificial intelligence is unlikely to be reliable. A single pixel change is enough to convince a machine-learning system that a stealth bomber is a dog.^[30] Although such a radical error was only possible in a specific circumstance, battlefields are complex and dynamic, and those errors may arise unexpectedly. Second, because drones within a swarm communicate by definition, a mistake in one drone may cascade to the drone swarm as a whole. For example, if one drone attacks a school bus after mistaking it for an enemy tank, other drones may attack, too, because they are following the lead of the first.

Third, the interaction between the drones may produce collective error. That is, a drone swarm operating on a distributed, collective intelligence may collect accurate information from its sensors, but draw an incorrect inference about the location or existence of an adversary. Altogether, the reliability challenges coupled with the potential for mass harm mean that drone swarms are potential future weapons of mass destruction, with significant challenges in holding to laws of armed conflict around discriminating between civilian and non-civilian targets and applying proportional force to achieve a military objective.^[31]

Dealing with adversarial action: Integrating drones into a drone swarm provides new potential paths for adversaries to manipulate or disrupt the drones. Drone swarms depend on communication between the drones, which may be jammed or manipulated.^[32] Because the drone swarms must be increasingly autonomous, an adversary might also attempt to trick the swarm into hitting a friendly target or crash into a mountain.^[33] An adversary might also attempt to exacerbate the reliability concerns by using decoys or camouflage to induce more frequent and serious errors. Although testing, evaluation, and technological hardening may reduce adversarial concerns, it is likely to be extremely difficult to certify operations for large, complex drone swarms operating in a dynamic environment.^[34] If an adversary can readily defeat or manipulate the swarm, adding swarming capability may not be worth it.

Infrastructure and support systems: Battlefield drone swarms require logistics, maintenance, and production capabilities to sustain them. A state might be able to develop a swarm of 10,000 drones, all integrated and working together, but to be effective, that swarm still must be delivered to the battlefield. States are developing motherships—larger platforms to transport and deploy drones—but that necessarily means added cost and heightened vulnerability.^[35] If an adversary can find and destroy the mothership, all the drones it carries may be defeated or destroyed as well. If the 10,000 drones make it to the battlefield, many will likely be defeated or destroyed in combat. There are so many drones that even if half are lost, the swarm may still accomplish its objective. However, to sustain that momentum, the drones must be replaced. For example, in Ukraine’s war with Russia, Ukraine lost an estimated 10,000 drones per month.^[36] States need the production and supply capacity to replace the drones lost in combat.

The Supply of Drone Swarms

If a military desires drone swarms, they must acquire one. That means acquiring four things: the drone platform, payloads, control stations, and the swarm management system. The drone platform can range broadly from a simple quadcopter to a large MQ-9 Reaper or naval vessel. Although drone swarm platforms may be virtually identical to a non-swarming drone, the platform may incorporate transmitters and receivers for intra-swarm communication. The drones might carry some combination of infrared, electromagnetic, or other sensors; grenades, bombs, guns, or other munitions; jammers, microwaves, or other directed weapons; even chemical or biological weapons. The control station allows operators to provide command and control for the drone swarm and may be handheld or have man-portable controls, a station at a base, or integrated into another platform, like a manned aircraft or truck. The swarm management system makes the swarm a swarm. That system includes the algorithms, software, and any specialised hardware needed to connect the drones together and allow them to operate as a collective.

States have two options: build or buy. In the short term, states will almost certainly have to build their own, because drone swarms are an emerging technology. As the technology matures and becomes more broadly available, buying a drone swarm becomes increasingly plausible.

Building a drone swarm requires the financial, technological, and production resources to build the drone platforms, develop the swarming behavioural algorithms, write the software and firmware code to integrate the swarming behaviours into the drone platforms, equip the drones with whatever payloads they might use, and test and evaluate the result. Of course, not all needs to be done from scratch. The actor may already have drone platforms capable of autonomous functions that can be readily modified to accommodate swarming, as well as access to basic swarming algorithms through open-source research. However, as drone swarms are a novel technology, developing the system may require basic scientific research through university laboratories. It is possible that a state that does not have sufficient capability (or better capability exists elsewhere) can acquire it from another state by funding research, as suggested by June 2023 reports that Iran worked with British scientists to investigate using lasers for intra-swarm communication.^[37]

Battlefield recovery, theft, or illicit transfer of swarming drones, components thereof, or technical data could help. The value of drone swarms in generating attritable mass also creates a proliferation risk. A drone swarm might have the numbers to endure and overwhelm adversary defences, but the adversary can recover fallen drones to understand how they operate and develop their own. For example, in December 2011, Iran captured an American RQ-170 Sentinel drone by manipulating the RQ-170’s GPS, after which Iran proceeded to build its own version.^[38] Growth in drone forensics as a discipline may enable proliferation through battlefield recovery in expanding the know-how to acquire critical information about how recovered drones function.^[39] Alternatively, states can use cyberattacks or conventional espionage to steal technical information usable to design and build drone swarms. For example, for decades, China has stolen information on American military technology, including the F-22 and MQ-9 Reaper, and appears to have used that information to develop native weapon systems.^[40] However, once the information is acquired, a state still needs the technical capacity to use that information and build the new system.

Buying a drone swarm requires financing, and, more importantly, a willing seller. Only one state—Israel—possesses a drone swarm used in actual combat. If Israel is unwilling to sell that system, there is no other option for a battle-tested drone swarm. As drone swarm technology matures and more states develop, deploy, use, and produce drone swarms at scale, more options might open. However, states may still be unwilling to sell. In August 2020, a bipartisan group of US senators attempted to block the sale of the MQ-9 Reaper to Saudi Arabia and the United Arab Emirates due to concerns about the US getting too involved in the Saudi and Emirati war with Yemen and the potential to encourage broader proliferation.^[41] If drone swarms prove to be militarily significant, states may worry that sales will affect balance of power within regions. If the drone swarm involves aerial drones capable of carrying over 500 kilograms, the Missile Technology Control Regime would also apply.^[42] But even if the state cannot build a complete drone swarm, they may be able to buy commercial parts that can be used to help make one. For example, 40 out of 53 identified components in Iranian Shahed-136 drones are manufactured in Western countries, including multiple American companies, despite years of sanctions and tough export controls on Iran.^[43]

Commercial entities provide alternate sources to buy a drone swarm, but the military value may be limited. Companies are increasingly developing and making simple drone swarms available, such as Red Cat Holding’s 4-Ship.^[44] The 4-Ship allows operators to control six drones simultaneously: four carrying out operations with the other two to swap in as needed.^[45] Four drones might be a useful drone swarm for small unit reconnaissance and surveillance, but they are unarmed and far from the hundreds of drones that might make up a military swarm. The control systems and swarming algorithms in a commercial drone swarm will likely also need to be modified, if not wholly replaced, to accommodate military needs. A commercial drone swarm does not need the capability to release a bomb, engage adversaries, or respond to signal or GPS jamming or spoofing.^[46] Nonetheless, commercial drone swarms are still significant for the proliferation of military drone swarms, because the technical know-how they invest in, develop, maintain, and expand could be used for military purposes.

Combatting Proliferation

States concerned about the proliferation of drone swarms can attempt to limit supply and reduce demand. Export controls on the transfer of drone swarm technology and the know-how necessary to create them may help slow the spread of drone swarms, while new international norms, backed up by international legal regimes, can help reduce demand. The effects of drone swarm proliferation can also be mitigated through concentrated and collaborative research on swarm countermeasures. If novel countermeasures neutralise the military value of drone swarms, then the effects on global stability may be limited.

Export controls can help reduce the transfer of military drone swarms, critical components, and the know-how to create either. Legitimate defence companies depend on trust relationships with their governments, and few would risk the loss of reputation, future contracts, and fees over a single deal. For drone swarms, the challenge is the small start-ups that may be unaware of their export compliance requirements, or lack the capacity to provide adequate due diligence.^[47] Small companies may also lack robust cybersecurity measures to guard against cyber theft. Universities and other research institutes are also a challenge because researchers may not appreciate the dual-use aspect of technology, may be repatriated to their home countries, or participate in academic exchanges with researchers in countries of concern.^[48] Governments should emphasise public-private collaborations to build awareness of compliance requirements and build necessary capacity.^[49] Global governments will likely struggle to detect illicit transfers without private-sector collaboration. As of 2020, 38 states already have armed drones, while 28 more have programmes in development.^[50] Turning those drones into drone swarms depends primarily on acquiring software that could be transferred on a thumb drive or the intangible know-how to develop the algorithms and code. Both will be difficult to detect.

International norms, augmented by binding international laws, can help reduce demand for drone swarms, and encourage global implementation of export controls. The reliability problems of drone swarms, especially armed, autonomous ones, may lead states to forego them entirely. No soldier wants a weapon that does not work. An errant attack that destroys a neutral or friendly target is a waste of munition, may hinder friendly movements, and may alert adversaries to friendly positions. Plus, states may be unwilling to use a weapon with significant law-of-war concerns. Although states may still develop and acquire unarmed drone swarms for intelligence gathering, they may refrain from massive armed drone swarms. New international norms, conventions, and laws restricting or banning autonomous weapons would also necessarily apply to autonomous drone swarms.^[51] Of course, states that perceive a strong military need for drone swarms are unlikely to forego armed drone swarms entirely. However, norms can still limit the speed of proliferation by reducing the availability of technical know-how and the number of states willing and able to transfer the technology.

Improved drone swarm defences would counter the effects of global proliferation. If states can reliably defeat drone swarms, effects on regional security balances are reduced. Although the outline of the drone swarm answer is clear, the details are not. The challenge with countering drone swarms is finding solutions that can disable, defeat, or destroy many drones in a cost-effective manner. High-powered microwaves like the Leonidas or THOR systems have some promise. High-powered microwaves have a very low cost per shot, demanding only electricity to fire, and can create effects over a broad area. However, the systems themselves are costly (Leonidas platforms cost about $16.5 million per platform), set-up takes a few hours, effective ranges are short, and simple countermeasures like microwave absorbing materials are in development.^[52] States concerned about drone swarm proliferation could participate in and encourage multilateral countermeasure development activities, invest in novel technologies and concepts, and export solutions to friendly and allied nations.

Conclusion

Certain states are rushing to build drone swarms; global proliferation, however, will take time. This is especially the case when it comes to large, complex drone swarms that are still a nascent technology. States may also choose to forego building them altogether. Although drone swarms offer cheap mass, limited operator requirements, and distributed complexity across a broad range of military applications, they also have potential reliability problems, require significant support infrastructure, and entail risk and opportunity costs. Some states also just do not like armed drones.

States that want armed drone swarms will need to build their own, although buying a drone swarm will become increasingly feasible as the technology matures. States concerned about the spread of drone swarms can aim to reduce the demand through new international norms and treaties. States can also limit the supply of drone swarms through robust, multilateral export controls, involving strong engagement with small companies and research institutions.

Drone swarms are no longer fodder for science fiction; they are used on battlefields today. What remains to be seen is how globally ubiquitous they may become.

The first version of this brief appeared in the ORF-GP volume, Future Warfare and Critical Technologies: Evolving Tactics and Strategies, which can be accessed here: 

Endnotes

^[a] Drone swarms can consist of thousands of drones to overwhelm adversary defences.

^[b]  Although China, India, Israel, Russia, and a European consortium are also developing them.

^[1] Peter Bergen, Melissa Salyk-Virk, and David Sterman, “World of Drones,” New America, July 30, 2020, https://www.newamerica.org/international-security/reports/world-drones/introduction-how-we-became-a-world-of-drones/.

^[2] Bergen, Salyk-Virk, and Sterman, “World of Drones”

^[3] Bergen, Salyk-Virk, and Sterman, “World of Drones”

^[4] Dion Nissenbaum, Sune Engel Rasmussen, and Benoit Faucon, “With Iranian Help, Hamas Builds, ‘Made In Gaza’ Rockets and Drones To Target Israel,” Wall Street Journal, May 20, 2021, https://www.wsj.com/articles/with-iranian-help-hamas-builds-made-in-gaza-rockets-and-drones-to-target-israel-11621535346; “Roster Of Iran’s Drones,” United States Institute of Peace, March 2, 2023, https://iranprimer.usip.org/blog/2023/mar/02/roster-iran%E2%80%99s-drones; Håvard Haugstvedt and Jan Otto Jacobsen, “Taking Fourth-Generation Warfare To the Skies? An Empirical Exploration of Non-State Actors’ Use of Weaponized Unmanned Aerial Vehicles,” Perspectives on Terrorism 14, no. 5 (October 2020), https://www.jstor.org/stable/26940037?seq=5.

^[5] Dan Rassler, “Islamic State and Drones: Supply, Scale, and Future Threats,” Combatting Terrorism Center at West Point, July 2018, https://ctc.westpoint.edu/wp-content/uploads/2018/07/Islamic-State-and-Drones-Release-Version.pdf.

^[6] Zachary Kallenborn and Philipp C. Bleek, “Swarming Destruction: Drone Swarms and Chemical, Biological, Radiological, and Nuclear Weapons,” Nonproliferation Review 25, no. 5–6 (2019): 523–43.

^[7] David Hambling, “What are Drone Swarms and Why Does Every Military Suddenly Want One?,” Forbes, March 1, 2021, https://www.forbes.com/sites/davidhambling/2021/03/01/what-are-drone-swarms-and-why-does-everyone-suddenly-want-one/?sh=3248e4392f5c; Wiebe de Jager, “Dutch Drone Makers Develop Autonomous Drone Swarm for Defense Dept.,” DroneXL, July 22, 2021, https://dronexl.co/2021/07/22/dutch-drone-makers-autonomous-drone-swarm/; Bruce Crumley, “Korean Air Advances Use of Drone Swarms in Plane Inspections,” DroneDJ, June 12, 2023, https://dronedj.com/2023/06/12/korean-air-advances-use-of-drone-swarms-in-plane-inspections/; David Hambling, “Israel Used World’s First AI-Guided Combat Drone Swarm in Gaza Attacks,” New Scientist, June 30, 2021, https://www.newscientist.com/article/2282656-israel-used-worlds-first-ai-guided-combat-drone-swarm-in-gaza-attacks/.

^[8] Hambling, “Israel Used World’s First AI-Guided Combat Drone Swarm in Gaza Attacks”

^[9] Aaron Mehta, “Pentagon Launches 103 Unit Drone Swarm,” DefenseNews, January 10, 2017, https://www.defensenews.com/air/2017/01/10/pentagon-launches-103-unit-drone-swarm/.

^[10] “Elbit Systems Demonstrated Heterogeneous Swarm Capability to the Dutch RAS Concept Development & Experimentation Program,” Elbit Systems, November 15, 2021, https://elbitsystems.com/pr-new/elbit-systems-demonstrated-heterogeneous-swarm-capability-to-the-dutch-ras-concept-development-experimentation-program/.

^[11] Michael C. Horowitz, The Diffusion of Military Power: Causes and Consequences for International Politics (Oxfordshire: Princeton University Press, 2010), https://press.princeton.edu/books/paperback/9780691143965/the-diffusion-of-military-power.

^[12] Michael A. Hunzeker, Dying to Learn: Wartime Lessons from the Western Front (Cornell University Press, 2021), https://www.cornellpress.cornell.edu/book/9781501758454/dying-to-learn/#bookTabs=1.

^[13] Dennis J. Blasko, “Technology Determines Tactics: The Relationship between Technology and Doctrine in Chinese Military Thinking,” Journal of Strategic Studies 34, no. 3 (June 17, 2011), https://www.tandfonline.com/doi/abs/10.1080/01402390.2011.574979.

^[14] Kelley Sayler, “A World of Proliferated Drones: A Technology Primary,” Center for a New American Security, June 2015, https://drones.cnas.org/wp-content/uploads/2016/03/CNAS-World-of-Drones_052115.pdf.

^[15] Sayler, “A World of Proliferated Drones: A Technology Primary”

^[16] Andrea Gilli and Mauro Gilli, “The Diffusion of Drone Warfare? Industrial, Organizational, and Infrastructural Constraints,” Security Studies 25, no. 1 (February 25, 2016), https://www.tandfonline.com/doi/abs/10.1080/09636412.2016.1134189.

^[17] MIT Beaver Works, “Project Perdix,” https://beaverworks.ll.mit.edu/CMS/bw/projectperdixcapstone.

^[18] Michael Horowitz, Sarah Kreps, and Matthew Fuhrman, “Separating Fact from Fiction in the Debate Over Drone Proliferation,” International Security 41, no. 2 (October 1, 2016), https://direct.mit.edu/isec/article/41/2/7/12140/Separating-Fact-from-Fiction-in-the-Debate-over.

^[19] Of course, the relative value of each advantage, risk, and constraint will vary depending on the state in question.

^[20] Loc V. Pham et al., “UAV Swarm Attack: Protection System Alternatives for Destroyers,” 2012.

^[21] Pham et al., “UAV Swarm Attack: Protection System Alternatives for Destroyers”

^[22] Max Hunder, “Ukraine Says it Shot Down 36 Drones in Overnight Russian Attacks,” Reuters, May 25, 2023, https://www.reuters.com/world/europe/ukraine-says-it-shot-down-36-drones-overnight-russian-attacks-2023-05-25/.

^[23] Kris Osborn, “Drone Swarms Could Be Too Fast to Handle. Is AI the Answer?” The National Interest, April 14, 2021, https://nationalinterest.org/blog/reboot/drone-swarms-could-be-too-fast-handle-ai-answer-182659.

^[24] Zachary Kallenborn, “Swarm Talk: Understanding Drone Typology,” Modern War Institute at West Point, December 10, 2021, https://mwi.usma.edu/swarm-talk-understanding-drone-typology/.

^[25] Rebecca Hersman et al., “Under the Nuclear Shadow: Situational Awareness Technology and Crisis Decisionmaking,” Center for Security and International Studies, March 2020, https://csis-website-prod.s3.amazonaws.com/s3fs-public/publication/200318_UnderNucearShadow_FullReport_WEB.pdf; Zachary Kallenborn, “If the Oceans Become Transparent,” U.S. Naval Institute, October 2019, https://www.usni.org/magazines/proceedings/2019/october/if-oceans-become-transparent.

^[26] Joseph Trevithick, “Massive Drone Swarm Over Strait Decisive in Taiwan Conflict Wargames,” The Drive, May 19, 2022, https://www.thedrive.com/the-war-zone/massive-drone-swarm-over-strait-decisive-in-taiwan-conflict-wargames; David Hambling, “The US Navy Wants Swarms of Thousands of Small Drones,” MIT Technology Review, October 24, 2022, https://www.technologyreview.com/2022/10/24/1062039/us-navy-swarms-of-thousands-of-small-drones/.

^[27] Erik Lin-Greenberg, “Wargame of Drones: Remotely Piloted Aircraft and Crisis Escalation,” Journal of Conflict Resolution 66, no. 10 (June 6, 2022), https://journals.sagepub.com/doi/pdf/10.1177/00220027221106960?casa_token=SznysPgZMM8AAAAA:oYnjFvQjiFYp0jZqFYYkGRlvAdmilOyVT_w4DYL6BKSGQv9dsVWLoW4Z1OB5e7B5sUgupBiXUzmZlw.

^[28] Daniel Brunstetter and Megan Braun, “The Implications of Drones on the Just War Tradition,” Ethics & International Affairs 25, no. 3 (2011), https://www.cambridge.org/core/services/aop-cambridge-core/content/view/97ABF476B8494CC44A71E011DD8B7600/S0892679411000281a.pdf/the-implications-of-drones-on-the-just-war-tradition.pdf.

^[29] “Germany to Get Weaponized Drones for the First Time,” AFP, April 6, 2022, https://www.thedefensepost.com/2022/04/06/germany-weaponized-drones/.

^[30] “AI Image Recognition Fooled by Single Pixel Change,” BBC, November 3, 2017, https://www.bbc.com/news/technology-41845878.

^[31] Zachary Kallenborn, “Future Warfare Series No. 60: Are Drone Swarms Weapons of Mass Destruction?” United States Air Force Center for Strategic Deterrence Studies, June 29, 2020, https://media.defense.gov/2020/Jun/29/2002331131/-1/-1/0/60DRONESWARMS-MONOGRAPH.PDF; Zachary Kallenborn, “Meet the Future Weapon Of Mass Destruction, the Drone Swarm,” Bulletin of the Atomic Scientists, April 5, 2021, https://thebulletin.org/2021/04/meet-the-future-weapon-of-mass-destruction-the-drone-swarm/.

^[32] Zachary Kallenborn, “InfoSwarms: Drone Swarms and Information Warfare,” Parameters 52, no. 2 (2022), https://press.armywarcollege.edu/parameters/vol52/iss2/13/.

^[33] Paul Scharre, “Counter-Swarm: A Guide to Defeating Robotic Swarms,” War on the Rocks, March 31, 2015, https://warontherocks.com/2015/03/counter-swarm-a-guide-to-defeating-robotic-swarms/.

^[34] Arthur Holland Michel, “The Black Box, Unlocked: Predictability and Understandability in Military AI,” United Nations Institute for Disarmament Research, 2020, https://unidir.org/sites/default/files/2020-09/BlackBoxUnlocked.pdf

^[35] David Hambling, “Robot Motherships to Launch Drone Swarms from Sea, Underwater, Air, and Near-Space,” Forbes, February 5, 2021, https://www.forbes.com/sites/davidhambling/2021/02/05/robot-motherships-to-launch-drone-swarms-from-sea-underwater-air-and-near-space/?sh=57cda4d4215c; Zachary Kallenborn, “368. The Swarm Mother,” Mad Scientist Laboratory, November 22, 2021, https://madsciblog.tradoc.army.mil/368-the-swarm-mother/; Joe Saballa, “China Unveils ‘Mothership’ to Launch Drone Swarms,” Defense Post, May 31, 2022, https://www.thedefensepost.com/2022/05/31/china-mothership-drone-swarms/.

^[36] Jack Watling and Nick Reynolds, “Meatgrinder: Russian Tactics in the Second Year of its Invasion of Ukraine,” RUSI, May 19, 2023, https://static.rusi.org/403-SR-Russian-Tactics-web-final.pdf.

^[37] David Rose and Felix Pope, “Britain’s ‘Swarming Drone’ Research Shared with Iran,” The Jewish Chronicle, June 15, 2023, https://www.thejc.com/news/news/britains-swarming-drone-research-shared-with-iran-5P709ETPi4A6ciem12exx.

^[38] Scott Peterson, “Exclusive: Iran Hijacked US Drone, Says Iranian Engineer,” Christian Science Monitor, December 15, 2011, https://www.csmonitor.com/World/Middle-East/2011/1215/Exclusive-Iran-hijacked-US-drone-says-Iranian-engineer; Tyler Rogoway, “Iran’s RQ-170  Clone Crashes Suspiciously on 10^th Anniversary Of the Real One Falling into its Hands,” The Drive, December 6, 2021, https://www.thedrive.com/the-war-zone/43390/irans-rq-170-clone-suspiciously-crashes-on-10th-anniversary-of-real-one-falling-into-its-hands.

^[39] “Framework For Responding to a Drone Incident: For First Responders and Digital Forensics Practitioners,” INTERPOL, January 2020, https://www.interpol.int/content/download/15298/file/DFL_DroneIncident_Final_EN.pdf.

^[40] Ellen Loanes, “China Steals US Designs for New Weapons, and it’s Getting Away with ‘The Greatest Intellectual Property Theft in Human History’,” Business Insider, September 24, 2019, https://www.businessinsider.com/esper-warning-china-intellectual-property-theft-greatest-in-history-2019-9.

^[41] “US Senators Push to Block Drone Sales to Saudi Arabia, UAE,” Middle East Eye, August 7, 2020, https://www.middleeasteye.net/news/us-uae-saudi-arabia-senators-drone-sales; Edward Wong, “Trump Administration is Bypassing Arms Control Pact to Sell Large Armed Drones,” New York Times, July 24, 2020, https://www.nytimes.com/2020/07/24/us/politics/trump-arms-sales-drones.html.

^[42] Missile Technology Control Regime, https://mtcr.info/.

^[43] Natasha Bertrand, “CNN Exclusive: A Single Iranian Attack Drone Found to Contain Parts from More Than a Dozen US Companies,” CNN, January 4, 2023, https://www.cnn.com/2023/01/04/politics/iranian-drone-parts-13-us-companies-ukraine-russia/index.html.

^[44] Loz Blain, “You Can Now Buy a Co-Ordinated Multi-Drone Swarm in a Box,” New Atlas, June 2, 2022, https://newatlas.com/drones/red-cat-drone-swarm-product/.

^[45] Blain, “You Can Now Buy a Co-Ordinated Multi-Drone Swarm in a Box”

^[46] Kallenborn, “InfoSwarms: Drone Swarms and Information Warfare”

^[47] Rudi du Bois, Julia Bell, and Dries Bertrand, “To Share Or Not To Share?: The Challenge Of Controlled Technologies in Research and Development (R&D),” Strategic Trade Review 9, no. 10 (Winter / Spring 2023), https://strategictraderesearch.org/wp-content/uploads/2023/02/Rudi-du-Bois-Export-Controls-and-Research-and-Development.pdf.

^[48] Andrea Viski, “Advanced Conventional Weapons and Emerging Technologies: Recognizing and Preempting Proliferation Threats,” Strategic Trade Review, February 2022, https://strategictraderesearch.org/wp-content/uploads/2022/02/Advanced-Conventional-Weapons-and-Emerging-Technologies.pdf.

^[49] Viski, “Advanced Conventional Weapons and Emerging Technologies: Recognizing and Preempting Proliferation Threats”

^[50] Bergen, Salyk-Virk, and Sterman, “World of Drones”

^[51] Zachary Kallenborn, “A Partial Ban on Autonomous Weapons Would Make Everyone Safer,” Foreign Policy, October 14, 2020, https://foreignpolicy.com/2020/10/14/ai-drones-swarms-killer-robots-partial-ban-on-autonomous-weapons-would-make-everyone-safer/; Dharvi Vaid, “Human Rights Watch Seeks Treaty Banning ‘Killer Robots,” DW, August 11, 2020, https://www.dw.com/en/human-rights-watch-seeks-treaty-banning-killer-robots/a-54521323.

^[52] Ashley Roque, “US Army Selects Epirus’ Leonidas for High-power Microwave Initiative,” Breaking Defense, January 23, 2023, https://breakingdefense.com/2023/01/us-army-selects-epirus-leonidas-for-high-power-microwave-initiative/#:~:text=WASHINGTON%20%E2%80%94%20Epirus%2C%20a%20technology%20company,drones%2C%20the%20firm%20announced%20today; Oliver Parken, “THOR Microwave Anti-Drone System Downs Swarms in Test,” The Drive, May 19, 2023, https://www.thedrive.com/the-war-zone/thor-microwave-anti-drone-system-downs-swarms-in-test; Zachary Kallenborn and Marcel Plichta, “Breaking the Drone Shield: Countering Drone Defenses,” Joint Force Quarterly.

• International Affairs
• Defence and Security
• China
• India
• USA and Canada
• Middle East
• Armed drones
• China
• combat
• drone swarm technology
• drone swarms
• exports
• global proliferation
• Hamas
• Hezbollah
• Houthi rebels
• international norms
• Iran
• Islamic State
• Israel
• military value
• national export controls
• US

The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.