In the coming years, the United States’ nuclear arsenal is likely to change in two important ways. First, recent statements by Biden administration officials suggest the United States may expand its nuclear modernization plans. In a speech on August 1, 2024, former acting assistant secretary of defense for space policy Vipin Narang stated, “Absent a change in the nuclear trajectories of the PRC, Russia, and North Korea, we may reach a point where a change in the size or posture of our current deployed forces is necessary.” The reason for a potential expansion is the rapidly changing strategic landscape, and the growth of Russian and Chinese arsenals with no arms control agreements to constrain them or prevent arms racing. To maintain a credible deterrent for the new strategic landscape, the argument goes, the United States should revisit and potentially expand its nuclear modernization plans.
Updating nuclear command, control, and communication will be an essential piece of nuclear modernization plans. Nuclear command, control, and communication enables the president to decide when, where, and how to use nuclear weapons and communicate that decision to strategic forces. Nuclear command and control is often described as the “fourth leg” of the nuclear triad, which includes intercontinental ballistic missiles (ICBMs), nuclear-armed ballistic missile submarines (SSBNs), and nuclear-capable aircraft. But that description of a “fourth leg” isn’t quite right: nuclear command and control is more like the top of the three-legged stool holding all the other legs together. Without a credible nuclear command and control system, nuclear deterrence would be undermined.
The second change is technological: in a speech on October 28, 2024, the commander of the U.S. Strategic Command General Anthony Cotton outlined a new approach for incorporating artificial intelligence (AI) and other technologies to inform nuclear decisionmaking. At an event at CSIS on November 19, 2024, General Cotton expanded on this vision to explain how artificial intelligence could be useful in increasing efficiency in intelligence, surveillance, and reconnaissance, as one example. He said explicitly, “It’s not about . . . having a machine present forces to the president of the United States. It’s just the opposite of that.”
AI integration is now an unavoidable reality in many of the capabilities that feed into the nuclear command, control, and communication system of systems, particularly for threat identification and tracking. In practice, this could look like reliance on AI in threat tracking, information processing, and verification, with the important caveats that this could be certified and verified.
If Nuclear Command and Control Fails, Deterrence Fails
The overarching objective of nuclear command, control, and communication is to support presidential decisionmaking in a crisis. This requires providing accurate information about nuclear use and/or incoming threats, facilitating communication with advisers, and executing nuclear strikes. According to the 2022 U.S. Nuclear Posture Review, the main functions of nuclear command, control, and communication are, “detection, warning, and attack characterization; adaptive nuclear planning; decision-making conferencing; receiving and executing Presidential orders; and enabling the management and direction of forces.” While nuclear command, control, and communication is indeed a highly technical issue, ultimately it is about enabling the president to make the hardest decision of their presidency.
In the event a president was faced with a decision of whether or not to use nuclear weapons, they would likely confer with close military and political advisors and be presented with a series of preplanned targeting options but could also request other options through a rapid adaptive planning process. If the decision was made to use nuclear weapons, this would be communicated to the appropriate military commands, which would authenticate the orders. The nuclear command, control, and communication system would then generate an emergency action message, correlated to a specific option in a code book, which would be transmitted to military commanders and operators who would unlock codes for a weapon system with multiple officers authenticating the order before executing. From even this brief summary of the nuclear command, control, and communication process, one thing should be obvious: it relies heavily on electronic communications to get information to the president, relay the president’s decision, and authentic orders.
The United States’ current nuclear command, control, and communication system is a complex system of systems and largely a legacy of the Cold War. It includes ground-based phased array radars and overhead persistent infrared satellites (OPIR) to identify and track incoming threats, an Advanced Extremely High Frequency (AEHF) system to support crisis communication, along with a survivable and secure architecture, known as the “thin line,” which should be able to function in an electromagnetic pulse (EMP)–contaminated environment. The system also includes aircraft with the ability to serve as a nuclear command post, such as the E6-B, which is currently being upgraded.
Modernizing the U.S. nuclear arsenal will also require updating nuclear command, control, and communication. Due to political and budgetary constraints, the modernization programs for the three legs of the triad are not staggered but all occurring concurrently, with the rollout of new capabilities expected in the late 2020s and early 2030s. The replacement programs are already facing challenges. The Sentinel program, for example, is estimated to cost at least 81 percent more than anticipated with a delay of several years. One of the primary reasons for the cost overrun has been command and control, as the program requires replacing command centers and command and control facilities.
Nuclear command, control, and communication modernization is currently focused on updating and increasing resilience in current systems to include satellite, aircraft, and communication technology. But nuclear command, control, and communication modernization also includes plans to replace many legacy systems. For example, the FY 2025 budget request included $1.05 billion for research, development, testing, and evaluation of its Evolved Strategic SATCOM (ESS) program, which will replace AEHF in 2032, intending to be jam-resistant and survivable.
Two Peer Competition Challenges for Nuclear Command, Control, and Communication
Many of our assumptions about nuclear command and control remain rooted in the Cold War and are simply outdated and don’t apply to the new strategic landscape. For example, there is a widespread belief that the president will only have 10–30 minutes to make a decision about nuclear use; but that is perhaps the least likely scenario in an environment where nuclear risks include low-yield tactical nuclear use in a regional conflict with no direct threat to the U.S. homeland. And while much attention has been given to whether or not the United States should incorporate AI into nuclear command, control, and communication, these debates risk missing the bigger picture: AI will have a role to play in this new nuclear command, control, and communication system, but it should be tied to strategic objectives and the new threat environment.
The growth of China’s arsenal, along with ongoing expansion in Russian and North Korean nuclear capabilities in the absence of arms control has prompted the rethinking of nuclear modernization plans. For example, the 2023 bipartisan Strategic Posture Commission concluded “the nuclear force modernization POR [program of record] is absolutely essential, although not sufficient to meet the new threats posed by Russia and China.” Just as the administration evaluated plans for modernization of the triad in a two-peer environment, the United States should evaluate nuclear command, control, and communication modernization plans for this new environment.
There are at least three important trends in the emerging strategic environment with implications for nuclear command, control, and communication. First, the United States’ adversaries are primarily interested in nuclear weapons for coercive purposes to support their regional ambitions. This has been particularly notable in Russia’s attempts at nuclear bullying in Ukraine to include nuclear drills and repeated mention of Russia’s nuclear arsenal. Additionally, both Russia and China are expanding their arsenals with dual-capable intermediate-range systems. The regional focus on nuclear risks differs somewhat from Cold War concerns when there was a fear among U.S. political and military leaders about a disarming first strike, which came with short decisionmaking time in a use-it-or-lose-it scenario.
Adversaries’ reliance on nuclear weapons for regional coercion could have implications for nuclear command, control, and communication and presidential decisionmaking time. As previously mentioned, many experts continue to rely on Cold War assumptions that the president would only have 10–30 minutes to make a decision about nuclear use. This assumption underpins many of the criticisms of the current nuclear decisionmaking process, arguing that no single individual should be expected to make a reasonable decision in such a short period of time and this policy should be revisited, or the United States should increase reliance on AI to get the president the necessary information in the compressed timeline. But these arguments are based on a Cold War model that assumed the Soviet Union would be launching a major attack on the U.S. homeland, putting at risk the ICBM and air legs of the nuclear triad. While the United States cannot rule out the possibility of future threats to the U.S. homeland, this is not the most likely scenario now. A 2024 report by the Congressional Research Service noted that “although the prompt launch options may have dominated U.S. planning during the Cold War, they may no longer dominate U.S. nuclear war plans.”
But the regional focus of adversaries’ nuclear threats could complicate nuclear decisionmaking in other ways by thickening the fog of war, a second trend with implications for nuclear command, control, and communication. The United States’ adversaries are already colluding on a host of issues and should be expected to enable each other in a crisis, as North Korea and Iran have done in Russia’s war in Ukraine. A more dangerous scenario would be an adversary taking advantage of a crisis in another theater to pursue their own regional ambitions and threaten U.S. allies. Russia, Iran, and increasingly China rely on disinformation as part of a wider strategy to undermine U.S. alliances and sow plausible (or implausible) doubt about their intentions. All these factors add to greater complexity and uncertainty, with a deluge of information and potentially conflicting signals, which would be problematic for presidential decisionmaking.
A third implication of the new two-peer environment on nuclear command, control, and communication is new vulnerabilities, particularly in the space and cyber domains. A study by the Nuclear Threat Initiative found that nearly 90 percent of nuclear weapons systems will get new digital components as part of modernization to include nuclear command, control, and communication. Technology is an important trait of two peer competition. China has already launched cyberattacks on U.S. critical infrastructure, such as attacks on ports and power grids, and the FBI has warned that China has pre-positioned cyber assets with the “ability to physically wreak havoc on our critical infrastructure at a time of its choosing.” Additionally, as revealed earlier this year, Russia appears to be developing a capability to put a nuclear weapon in space, which would put at risk not only U.S. (and others’) military satellites but also commercial and communication assets used every day by the general public. What this means for nuclear command, control, and communication is that resilience will have to be a priority in the modernization process to include building in redundancies and ensuring a continuous means of communication and operations.
Nuclear Command, Control, and Communication Needs to Evolve Faster
The priority for nuclear command, control, and communication modernization, like legacy systems, should remain to facilitate presidential decisionmaking in a crisis, but these crises could look very different from crises of the past. Presidential decisionmaking could be complicated by an influx of information and confusion created by a multiple adversary environment, potentially involving multiple theaters. With these new challenges in mind, nuclear command, control, and communication modernization efforts should consider at least five options for responding to the new strategic landscape.
First, while many of the criticisms of sole authority overstate the time constraints, they do raise valid questions and concerns about the president’s familiarity with nuclear issues. A crisis should not be the first time a president has to consider the risks of nuclear use. Historically, some presidents have been extremely well-versed in nuclear issues and have taken a personal interest in understanding the risks associated with nuclear weapons. As Ankit Panda argued in a recent Project Atom study, deterrence strategies and nuclear postures should consider, “how current and future presidents may weigh response options.” Part of nuclear command, control, and communication modernization, therefore, should also include review and revision of the presidential briefings on nuclear issues at the start of a term, along with exercises for presidential advisors who would be called upon in the event of a nuclear crisis.
Second, like in the Cold War, in the new strategic environment, the president will be particularly hungry for concrete information and answers to complicated questions. Questions the president might ask include: Was the attack authorized or unauthorized? What is happening in the conventional conflict? What are our allies saying? How do I know the information you are giving me is accurate? For this reason, the intelligence community has an important role to play in collecting information before a crisis and confirming it during one. The Director of National Intelligence should not only increase the collection and analysis of adversaries’ nuclear strategies and employment doctrines, as recommended by the Strategic Posture Commission, but also be part of nuclear command, control, and communication modernization discussions to understand what questions the president might be asking in a crisis and to ensure intelligence can be relayed with confidence.
Third, to facilitate the transmission of all this information, some of which might be processed using AI, communication channels will need increased bandwidth. Modern presidents are used to 5G-speed communication, whether that be in their official duties or their personal lives. Even in a regional crisis, communication assets may be compromised and limit the ability to get accurate information to the president in a digestible format. John Harvey recommended, “To support consultations among allies, partners and potentially adversaries, in addition to senior military and advisors in complex conflict scenarios involving, say, combined offense and defense, nuclear and conventional operations . . . will require global, secure, high-quality voice, video and data transmissions that are resilient in stressed nuclear environments and go well beyond what was required for the Cold War mission.” Bandwidth should be a priority.
Fourth, nuclear command, control, and communication modernization will need to take a hybrid approach to AI integration. Experts such as Lowther and McGiffin call for increasing reliance on AI in nuclear command, control, and communication to include not only detecting an attack but also, potentially, preprogramming a response in the event of an incapacitated leadership. On the other hand, experts such as Saltini fear that AI models would be dangerous in nuclear command, control, and communication because of “their opaque nature, unpredictability, and susceptibility to cyber-attacks.” General Cotton’s comments suggest a more hybrid approach, explaining, “AI will enhance our decision-making capabilities. But we should never allow artificial intelligence to make those decisions for us.” Identifying the opportunities and risks of AI integration in NC3 is an ongoing exercise.
As a final consideration, given the rapid rate of technological change and adversaries’ advances in space and cyber threats, systems currently in prototype could be obsolete by the time they reach production. The current nuclear modernization efforts have not set an encouraging precedent for delivering programs on time and to budget, but the entire nuclear enterprise has been neglected for decades and cannot readily be “switched on” to innovate to keep up with the threat. As described by former STRATCOM commander General Hyten, “A lot of elements of the nuclear enterprise, we don’t have time—everything is just in time.” Nuclear command, control, and communication modernization should be survivable in a new technological environment, but it should also be flexible allowing for the new strategic environment.
Moving Towards Persistent, Flexible Modernization
The Department of Defense is already exploring new ways to be more adaptable. One example of this is ESS, which uses alternate acquisition pathways to encourage partnership with private sector innovation teams for different segments of development to allow for rapid prototyping and development. Another example is the Space Development Agency’s business model to harness commercial satellite development in delivering a new layer of low Earth orbit satellites, which will be operational in late 2024 and include 28 satellites for use by Indo-Pacific Command. At the same time, nuclear command, control, and communication modernization should not focus on future threats at the expense of current or legacy ones. For example, the United States cannot rule out the prospect of threats to the homeland over the next decade and therefore should continue to allow for presidential decisionmaking within a compressed timeline. The United States should not assume undue risk in the process of evolving faster.
The United States might need to confront an uncomfortable reality: to stay apace with the technological and threat environments, nuclear command, control, and communication modernization will not be an isolated endeavor, but rather a constant and persistent mission. For example, the Strategic Posture Commission called for prioritizing adaptive cyber defenses for nuclear command, control, and communication, along with strategic delivery platforms and warheads. This is not a one-and-done project. It will require constant innovation and evolution.
Heather Williams is director of the Project on Nuclear Issues and senior fellow with the Defense Studies Department at the Center for Strategic and International Studies in Washington, D.C.
This commentary was made possible by the support of Lockheed Martin Space.
