"I greatly appreciate your very swift response, and your organization's work in general. It's a terrific source of authoritative information."

– Lisa Beyer
Bloomberg News
August 27, 2018
April 2020

Arms Control Today April 2020

Edition Date: 
Wednesday, April 1, 2020
Cover Image: 

Pandemic Reveals Misplaced Priorities

April 2020
By Daryl G. Kimball

For decades, national security and health experts have warned of the risks of global threats that are simply too big for one country to handle, such as disease pandemics, climate change, and nuclear war. For many years, the response of our national and global leaders has fallen short.

Twenty years ago, John Steinbruner, then the chair of the Arms Control Association Board of Directors, warned in his book Principles of Global Security that globalization is generating “a new class of security problems in which dispersed processes pose dangers of large magnitude and incalculable probability.” He argued that policymakers “will have to shift from contingency reaction to anticipatory prevention” and “this will have to be done in global coalition.”

Unfortunately, U.S. spending priorities and modes of thinking about security have become increasingly defined in military terms. Congress provided a record $746 billion for national defense in fiscal year 2020. U.S. arms manufacturers dominate the global arms trade and help fuel regional conflicts that undermine human development. In recent years, the Trump administration’s nationalist “America First” foreign policy has made it even more difficult for the world’s leading nations to work together on the toughest global challenges.

Today, the novel coronavirus (COVID-19) pandemic, which threatens the lives and livelihoods of millions worldwide, has laid bare the terrible human cost of these misplaced policy choices.

As the scope and scale of the coronavirus threat began to reveal itself in January and February, the Trump administration focused on other matters. For example, the administration in February asked Congress for $44.5 billion in fiscal year 2021 for programs to maintain and upgrade the U.S. nuclear arsenal, a 19 percent increase above the previous year.

The U.S. government spends tens of billions of taxpayer dollars to maintain a massive nuclear arsenal capable of destroying the planet many times over. Meanwhile, it does not have a stockpile of masks large enough to protect front-line health care workers who are battling COVID-19 and is proposing to cut programs that help provide for early disease detection.

The U.S. stockpile of medical supplies includes 12 million medical-grade N95 masks and 30 million surgical masks, which is only about 1 percent of the 3.5 billion needed in a year to deal with a disease pandemic. At the price of $0.50 a mask, it would cost approximately $1.75 billion to build up the N95 stockpile and about $350 million a year to replace expired masks, according to a report published by The War Zone. That is less than the $3.2 billion increase above fiscal year 2020 levels that the Pentagon is seeking for its multiyear programs to sustain and rebuild the U.S. triad of nuclear-armed missiles, submarines, and bombers.

Meanwhile, the administration is proposing to slash by 37 percent the budget request for the Defense Department’s Biological Threat Reduction Program, which “seeks to facilitate detection and reporting of diseases caused by especially dangerous pathogens.” As a result of that program’s previously provided threat reduction training efforts, local officials in Thailand detected the first case of the novel coronavirus there, only days after its initial discovery in Wuhan, China.

Now is the time for Congress to radically scale back the existing plan to replace and upgrade the already excessive U.S. nuclear arsenal, particularly plans for new missiles and bombers, new nuclear warheads, and production infrastructure. This would save billions of taxpayer dollars that should be spent on addressing higher priority human and health security needs.

Making matters worse, the United States has become part of the problem rather than helping to find viable solutions to counter the most serious global threats.

While the Trump administration is seeking to expand U.S. nuclear capabilities at the expense of programs that address human security needs, it is turning its back on hard-won agreements that have effectively reduced the nuclear threat.

President Donald Trump’s unilateral withdrawal from the 2015 Iran nuclear deal with no viable plan to replace it creates the potential for a new nuclear crisis. Iran’s leaders have retaliated to the reimposition of U.S. sanctions by breaching key limits on their nuclear activities.

In addition, the post-Cold War progress toward reducing the role and number of nuclear weapons has stalled. To date, Trump has failed to take up Russia’s offer to extend the only remaining treaty that limits the world’s two largest nuclear arsenals, the 2010 New Strategic Arms Reduction Treaty. The global nonproliferation and disarmament regime, the best prophylactic against a nuclear pandemic, is under serious threat.

The unfolding COVID-19 outbreak will not only take away the lives of people, but it will change our personal lives, and it will very likely force changes in the international system. If we are to survive well into this century, there must be a profound shift in the way we deal with global security challenges and how we align our scientific, economic, diplomatic, and political resources to address the health, climate, and nuclear dangers that threaten us all.


For decades, national security and health experts have warned of the risks of global threats that are simply too big for one country to handle, such as disease pandemics, climate change, and nuclear war. For many years, the response of our national and global leaders has fallen short.

Russia’s View on Nuclear Arms Control: An Interview With Ambassador Anatoly Antonov

April 2020

Arms Control Today conducted a written interview in early March with Anatoly Antonov, Russian ambassador to the United States on issues including the current status of U.S.-Russian strategic security talks, the future of New START, talks on intermediate-range missile systems, engaging China in arms control, and President Vladimir Putin’s proposal for a summit of the leaders of the five permanent members of the UN Security Council.

Ambassador Anatoly Antonov, then director of the Russian Foreign Ministry's Department of Security and Disarmament Affairs, speaks at the closing plenary of the New START negotiations on Apr. 9, 2010, one day after the treaty was signed in Prague by the U.S. and Russian presidents. (Photo: Eric Bridiers/U.S. Mission, Geneva)Antonov was appointed ambassador to the United States in August 2017. For more than three decades, he has served in the Soviet Ministry of Foreign Affairs and its successor, the Russian Ministry of Foreign Affairs, where he has specialized in the control of nuclear, chemical, and biological weapons. Serving as the ministry’s director for security and disarmament, he headed Russia’s delegation to the 2009 negotiations on the New Strategic Arms Reduction Treaty (New START). He was appointed deputy minister of defense in 2011 and deputy minister of foreign affairs in 2016.

Arms Control Today: What issues were discussed in the recent U.S.-Russian strategic security talks in Vienna? When do the two sides plan to meet next? Does Russia find this dialogue on issues affecting strategic stability useful and, if so, why?

Amb. Anatoly Antonov: Russia and the United States are the largest nuclear weapons powers and permanent members of the UN Security Council. They bear a special responsibility for preserving world peace and security. That is why it is crucial to maintain the bilateral strategic stability dialogue at any given circumstance, regardless of political situation. It goes without saying that such engagement should be conducted on a regular basis.

While discussing security issues, one must keep in mind that any conversation, no matter how substantial it might be, should focus on achieving tangible results. Reaching agreements on reducing tensions and mutually acceptable arms control solutions could help meet this goal. The primary task is to rebuild confidence in this area, attempt to preserve treaties that are still in effect, [and] mitigate crisis dynamic.

As for the consultations in January, our reaction can be described as “cautious optimism.” On the bright side is the fact that the meeting did take place, even though it exposed serious disagreements between our countries on a number of topics. Without going into detail, I must note that on many occasions we heard our partners talking about a concept of conducting dialogue within the framework of the so-called great power competition. In our view, such a formula could hardly serve as a foundation for building constructive cooperation on security issues between nuclear powers.

Nonetheless, Russian and American negotiators managed to discuss factors that significantly impact strategic stability (even though our partners somehow prefer the term “strategic security”). In our perspective, they include, above all, deployment of global missile defense, implementation of the “prompt strike” concept, threat of placement of weapons in outer space and designation of space as a “war-fighting domain,” quantitative and qualitative imbalances in conventional arms in Europe, development and deployment of low-yield nuclear warheads, and adoption of new doctrines that lead to lowering the threshold of using nuclear weapons.

In our view, another positive outcome of the renewed Russian-U.S. dialogue on strategic stability was the agreement reached in Vienna on conducting expert group discussions on specific topics, which we have to go over and agree on.

ACT: Do you agree or disagree with the idea that there is ample time to decide whether to extend the New Strategic Arms Reduction Treaty (New START)? From Moscow’s view, when must the presidents of the United States and Russia formally agree on extension of New START to ensure completion of the necessary processes before its expiration date? Is it Russia’s view that the treaty can only be extended once, or can it be extended multiple times totaling up to five years if the two parties decide to pursue that approach?

Is it possible for the Duma to provisionally recognize a joint decision by the two presidents to extend the treaty in order to allow a decision on extension closer to the expiration date?

Antonov: As you have correctly noted, Russian President Vladimir Putin clearly spelled out our stance on New START. On December 5, 2019, he declared our country’s readiness to immediately and unconditionally extend the treaty. Later last year, we officially suggested that Russia and the United States should review the entire set of corresponding issues including the term of the treaty’s possible extension (up to five years).

A Russian defense official shows Russia's 9M729 cruise missile at a facility outside Moscow on Jan. 23, 2019. Russian Ambassador to the United States Anatoly Antonov disputes the U.S. accusation that the missile violated the INF Treaty. (Photo: Vasily Maximov/AFP/Getty Images)However, we have yet to get a response. Trump administration representatives keep claiming that “there is still time” since the extension of the treaty in their view can be formalized in a matter of days. These statements are made despite our repeated clarifications that New START’s extension is not a “mere technicality,” but a rather extensive process that requires the Russian side to undertake a series of domestic legislative procedures. I would like to reiterate that as past similar review processes show, it may take several months to complete the New START extension.

Therefore, it is surprising that the U.S. Department of State refused to conduct consultations proposed by the Russian side on legal aspects of potential extension of the treaty. In response, we hear mixed comments (for instance, during the briefing of a “senior State Department official” on March 9, 2020) on the nature of interaction between the executive and legislative branches in Russia.

As for your last question, I would rather not contemplate in a conditional tense. I wish to emphasize: Russia stands ready to reach an agreement on New START’s extension even this very day. However, our goodwill is not enough. It requires U.S. consent, which we have not received yet. Should Washington agree, we will immediately begin implementation of the corresponding domestic procedures.

We hope that the United States will finalize its stance on New START in the nearest future since there is not much time left before the treaty expires in February 2021.

ACT: For nearly a year, the United States has insisted that China be involved in trilateral nuclear arms control negotiations with Russia and the United States. Chinese officials have said, however, that given the disparities between their arsenal and those of the United States and Russia, they are not interested in trilateral arms control talks at this time. Russia has said that if the U.S. side can persuade China to participate, then other nuclear-armed states such as France and the United Kingdom should be involved.

In Russia’s view, which nuclear arms issues and which types of weapons should be part of any bilateral or multilateral follow-on negotiation to New START? Would Russia be willing to engage in negotiations designed to limit or reduce stockpiles of nonstrategic nuclear weapons as well as strategic nuclear weapons? When, in Russia’s view, should any such New START follow-on talks begin?

Antonov: I would like to remind you that our stance on this issue dates back to 2010. We have said more than once that, with the signing of New START, any possibilities for further reduction and limitation of strategic offensive arms on a bilateral basis are virtually exhausted and that further progress in this area will require involvement of other states with military nuclear capabilities. However, we do not understand why some of our U.S. colleagues talk exclusively about China. Let’s also involve NATO members possessing nuclear weapons, Great Britain and France. In fact, that is what the special representative of the president for nuclear nonproliferation, Ambassador Jeffrey Eberhardt, suggested in his March interview with your journal, when he said, “we have to move beyond bilateral discussions between ourselves and Russia and bring in other countries.”

We are convinced that cooperation with third countries in developing possible new agreements in this area should be strictly consensus based and pose no threats to legitimate security interests of the parties. Beijing has clearly rejected the idea of being involved in the so-called trilateral agreements on nuclear arms control that you have mentioned. We believe that this “obsession” with the trilateral format can become a serious obstacle to the development of the Russian-U.S. strategic dialogue, in particular, in terms of preserving existing treaties and developing possible new bilateral agreements.

There is no doubt that the Russian-U.S. bilateral arms control agenda remains relevant. We are open to discussing within the strategic dialogue the issue of the newest and prospective weapons that do not fall under New START. However, the conversation on this topic should be conducted in a comprehensive manner, which takes into account interests of both sides.

At the same time, the possible extension of New START would give Russia and the United States an opportunity to discuss the prospects of bilateral and multilateral arms control regimes in the environment of strategic predictability.

ACT: Regarding your proposal to convene a heads-of-state meeting among the five permanent members of the UN Security Council, what specifically would be discussed at such a meeting, and what specific outcomes does President Putin think could be achieved and how?

Antonov: Currently we have been conducting preliminary discussion on a possible date and venue for the summit.

The goal of the summit, as stated by Russian President Putin, is to begin a substantial conversation on the fundamental principles of cooperation on the international arena in order to resolve the most pressing issues faced by the global community. A meeting of the leaders of the five permanent members of the Security Council is the most appropriate format for such a dialogue to commence.

We proceed from an understanding that the leaders will discuss the crisis situation in global stability and security, including the erosion of the UN-set foundations of the world order, regional conflicts, fight against international terrorism and transnational organized crime, challenges of migration, and destabilizing technologies. We will not be able to leave out disarmament and arms control issues. We hope that the summit will allow us to identify approaches to solving pressing strategic stability issues.

But it can only be achieved within an interested and mutually respectful dialogue that implies consideration of interests of all sides. Later, other countries can and must join these efforts since only collectively we may solve the global problems of humanity. The summit is our proposal to the international community to step away from confrontational thinking and get behind a productive agenda.

ACT: Would Russia’s proposal for talks on a moratorium on deploying intermediate-range missiles also prohibit Russian deployment of the 9M729 ground-launched cruise missile, which U.S. and NATO officials have charged as an Intermediate-Range Nuclear Forces Treaty (INF Treaty)-noncompliant system? Which geographic “environments” does the Russian proposal envision becoming nondeployment zones for these prohibited missiles? How would the parties to the agreement monitor and verify compliance or otherwise share information about the locations and numbers of the prohibited systems? Lastly, is Russia open to considering counterproposals to its initial concept, and with which countries does Russia seek to negotiate such a missile moratorium?

Antonov: Russian President Putin’s message to the heads of the leading countries, including the United States and other NATO members, dated September 18, 2019, states that our country made a voluntary commitment not to deploy ground-based intermediate- and shorter-range missiles in Europe and other regions so long as the United States refrains from doing so. On many occasions, we have called on other countries to support this initiative in order to prevent a new missile arms race, primarily on the European continent.

We believe that a multilateral moratorium in accordance with the Russian proposal will require additional verification measures, especially considering that launchers capable of firing intermediate-range land-based missiles are already deployed in Romania (Poland soon will follow suit). It was clearly proven during the test of a sea-based Tomahawk cruise missile fired from a ground-based Mk41 launcher conducted on August 18, 2019. Should our U.S. and European partners be interested, Russia is ready to work out corresponding technical aspects of the verification regime.

As for 9M729 missiles, the alleged “proof” amassed by the United States and NATO of our systems violating the INF Treaty (while it was in effect) has never been presented either to us or the international community.

Russia stands ready to discuss the issues of intermediate- and shorter-range ground-based missiles with all concerned countries. Our call to adhere to a moratorium, similar to the one already observed by our country, is addressed above all to Washington and its allies in Europe and the Asia-Pacific region.

ACT: Regarding the nuclear Nonproliferation Treaty (NPT), what are the main action steps on nuclear disarmament, previously agreed in the 2010 review conference outcome document, or perhaps new steps that Russia will encourage the 10th NPT review conference to support? What specific nuclear risk reduction measures is Russia ready to support in the context of the NPT review conference? [Editor: The 2020 NPT Review Conference will not meet as scheduled, see ACT news article, this issue.]

Antonov: Our stance and priorities in nuclear disarmament have been comprehensively described in the Russian working paper submitted to the second preparatory committee for the 10th NPT review conference. It stipulates a consensus-based incremental approach that implies consistent work on creating the right conditions that help the global community to continue down the path toward nuclear disarmament.

In this regard, we consider the forced development of the Treaty on the Prohibition of Nuclear Weapons (now open for signing) as wrongful. It fails to promote nuclear disarmament, undermines the NPT, and creates additional tensions between its participants. We believe that complete elimination of nuclear weapons is only possible within comprehensive and complete disarmament and under conditions of equal and indivisible security for all, including nuclear states, in accordance with the NPT.

A significant contribution to progress in nuclear disarmament would be made by extending New START and adopting a moratorium on the deployment of ground-based intermediate- and shorter-range missiles by the United States and its allies. An important role in efforts to limit and reduce nuclear weapons is played by the Comprehensive Test Ban Treaty (CTBT). Unfortunately, since the CTBT was opened for signature 20 years ago, the world has still been awaiting its entry into force.

As for nuclear risks, we are working on a joint statement with the other permanent members of the UN Security Council on the inadmissibility of a nuclear war (the United States has failed to respond to Russia’s proposal to do it in a bilateral format). This could in a way become a reconfirmation of the well-known Gorbachev-Reagan formula, this time in a multilateral format.

Russia’s ambassador to the United States discusses strategic security, New START, and other key topics.

‘Skynet’ Revisited: The Dangerous Allure of Nuclear Command Automation

April 2020
By Michael T. Klare

The U.S. Department of Defense’s proposed budget for fiscal year 2021 includes $28.9 billion for modernizing the U.S. nuclear weapons complex, twice the amount requested for the current fiscal year and a major signal of the Trump administration’s strategic priorities. Included in this request are billions of dollars for the procurement of new nuclear delivery systems, including the B-21 bomber, the Columbia-class ballistic missile submarine, and an advanced intercontinental ballistic missile (ICBM).

In the 1983 film "WarGames," a computer placed in charge of U.S. nuclear weapons begins a simulation that nearly leads to the launch of U.S. missiles. While no U.S. Defense Department plans resemble anything quite like that scenario, they do seek to develop essential building blocks for a highly automated command and control system that will progressively diminish the role of humans in making critical decisions over the use of nuclear weapons. (Photo: Hulton Archive/Getty Images)However, the largest program request in the modernization budget, totaling $7 billion, is not for any of those weapons but for modernizing the nation’s nuclear command, control, and communications (NC3) infrastructure, the electronic systems that inform national leaders of a possible enemy attack and enable the president to order the launch of U.S. bombers and missiles. The administration avows it will use this amount to replace outdated equipment with more modern and reliable systems and to protect against increasingly severe cyberattacks. Equally pressing, for Pentagon planners, is a drive to increase the automation of these systems, a goal that has certain attractions in terms of increased speed and accuracy, but also one that raises troubling questions about the role of machines in deciding humanity’s fate in a future nuclear showdown.

Science fiction filmmakers have long envisioned the possibility of machines acquiring the capacity to launch nuclear weapons on their own. The 1964 movie “Dr. Strangelove” presupposes that the Soviet Union has installed a “doomsday machine” primed to detonate automatically should the country come under attack by U.S. nuclear forces. When the U.S. leadership fails to halt such an attack by a rogue Air Force general, the doomsday scenario is set in motion. In 1983’s “WarGames,” a teenage hacker inadvertently ignites a nuclear crisis when he hacks into the (fictional) War Operation Plan Response (WOPR) supercomputer and the machine, believing it is simply playing a game, attempts to fight World War III on its own. Yet another vision of computers run amok was portrayed a year later in “The Terminator,” in which a superintelligent computer known as Skynet again controls U.S. nuclear weapons and elects to eliminate all humans by igniting a catastrophic nuclear war.

To be sure, none of the plans for NC3 automation now being considered by the Defense Department resemble anything quite like the WOPR or Skynet. Yet, these plans do involve developing essential building blocks for a highly automated command and control system that will progressively diminish the role of humans in making critical decisions over the use of nuclear weapons. Humans may be accorded the final authority to launch nuclear bombers and missiles, but assessments of enemy moves and intentions and the winnowing down of possible U.S. responses will largely be conducted by machines relying on artificial intelligence (AI).

The NC3 Modernization Drive

The total overhaul of the U.S. NC3 infrastructure was first proposed in the administration’s Nuclear Posture Review (NPR) report of February 2018. The existing system, the report stated, “is a legacy of the Cold War, last comprehensively updated almost three decades ago.” Although many of its individual components—early-warning satellites and radars, communications satellites and ground stations, missile launch facilities, national command centers—had been modernized over time, much of the interconnecting hardware and software has become obsolete, the Pentagon stated. The growing effectiveness of cyberattacks, moreover, was said to pose an ever-increasing threat to the safety and reliability of critical systems. To ensure that the president enjoyed timely warning of enemy attacks and was able to order appropriate responses, even under conditions of intense nuclear assault and cyberattack, the entire system would have to be rebuilt.

Given these highly demanding requirements, the NPR report called for overhauling the existing NC3 system and replacing many of its component parts with more modern, capable upgrades. Key objectives, it stated, include strengthening protection against cyber and space-based threats, enhancing integrated tactical warning and attack assessment, and advancing decision-support technology. These undertakings are ambitious and costly and go a long way toward explaining that $7 billion budget request for fiscal year 2021. Equally large sums are expected to be requested in the years ahead. In January 2019, the Congressional Budget Office projected that the cost of modernizing the U.S. NC3 system over the next 10 years will total $77 billion.1

The NPR report, as well as the budget documents submitted in 2020 in accordance with its dictates, do not identify increased automation as a specific objective of this overhaul. That is so, in part, because automation is already built into many of the systems incorporated into the existing NC3 system, such as launch-detection radars, and will remain integral to their replacements. At the same time, many proposed systems, such as decision-support technology—algorithms designed to assess enemy intentions and devise possible countermoves from which combat commanders can choose—are still in their infancy. Nevertheless, virtually every aspect of the NC3 upgrade is expected to benefit from advances in AI and machine learning.

The Allure of Automation

The quest to further automate key elements of the U.S. NC3 architecture is being driven largely by an altered perception of the global threat environment. Although the existing framework was always intended to provide decisionmakers with prompt warning of enemy nuclear attack and to operate even under conditions of nuclear war, the operational challenges faced by that system have grown more severe in recent years. Most notably, the decision-making system is threatened by the ever-increasing destructive capacity of conventional weapons, the growing sophistication of cyberattacks, and, as a result of those two, the growing speed of combat.

The existing NC3 architecture was designed in the previous century to detect enemy ICBM and bomber launches and provide decision-makers with enough time, as much as 30 minutes in the case of ICBM attacks, to assess the accuracy of launch warnings and still ponder appropriate responses. These systems did not always work as intended—the history of the Cold War is replete with false warnings of enemy attacks—but the cushion of time prevented a major catastrophe.2 The reasonably clear distinction between conventional and nuclear weapons, moreover, enabled military analysts to avoid confusing non-nuclear assaults with possibly nuclear ones.

With the introduction of increasingly capable conventional weapons, however, the distinction between nuclear and non-nuclear weapons is being blurred. Many of the new conventionally armed (but possibly nuclear-capable) ballistic missiles now being developed by the major powers are capable of hypersonic speed (more than five times the speed of sound) and of flying more than 500 kilometers (the limit imposed by the now-defunct Intermediate-Range Nuclear Forces Treaty) and are intended for attacks on high-value enemy targets, such as air defense radars and command-and-control facilities. With flight durations of as little as five minutes, defensive NC3 systems have precious little time to determine whether incoming missiles are armed with nuclear or conventional missiles and to select an appropriate response, possibly including the early use of nuclear weapons.3

Cybercombat occurs at an even faster speed, potentially depriving nuclear commanders of critical information and communication links in a time of crisis and thereby precipitating inadvertent escalation.4 In the highly contested environment envisioned by the 2018 NPR report, decision-makers may be faced with an overload of inconclusive information and have mere minutes in which to grasp the essential reality and decide on humanity’s fate.

Under these circumstances, some analysts argue, increased NC3 automation will prove essential. They claim that increased reliance on AI can help with two of the existing system’s most acute challenges: information overload and time compression. With ever more sensors, (satellite monitors, ground radars, and surveillance aircraft) feeding intelligence into battle management systems, commanders are being inundated with information on enemy actions, preventing prompt and considered decision-making. At the same time, the use of more hypersonic missiles and advanced cyberweaponry has compressed the time in which such decisions must be made. AI could help overcome these challenges by sifting through the incoming data at lightning speed and highlighting the most important results and by distinguishing false warnings of nuclear attack from genuine ones.5

Automation could be even more useful, advocates claim, in helping commanders, up to and including the president, decide on nuclear and non-nuclear responses to confirmed indications of an enemy attack. With little time to act, human decision-makers could receive a menu of possible options devised by algorithms. “As the complexity of AI systems matures,” the Congressional Research Service noted in 2019, “AI algorithms may also be capable of providing commanders with a menu of viable courses of action based on real-time analysis of the battle-space, in turn enabling faster adaptation to complex events.”6

Some analysts have gone even further, suggesting that in conditions of extreme time compression, the machines could be empowered to select the optimal response and initiate the attack themselves. “Attack-time compression has placed America’s senior leadership in a situation where the existing NC3 system may not act rapidly enough,” Adam Lowther and Curtis McGiffin wrote in a commentary at War on the Rocks, a security-oriented website. “Thus, it may be necessary to develop a system based on [AI], with predetermined response decisions, that detects, decides, and directs strategic forces with such speed that the attack-time compression challenge does not place the United States in an impossible position.”7

That commentary provoked widespread alarm about the possible loss of human control over decisions of Lieutenant General Jack Shanahan, director of the Joint Artificial Intelligence Center, appears at a September 2019 conference at Georgetown University. He spoke of the need for improving artificial intelligence in the U.S. military, but cautioned, "there is one area where I pause, and it has to do with nuclear command and control." (Photo: Georgetown University)nuclear use. Even some military officials expressed concern over such notions. “You will find no stronger proponent of integration of AI capabilities writ large into the Department of Defense,” said Lieutenant General Jack Shanahan, director of the Joint Artificial Intelligence Center (JAIC), at a September 2019 conference at Georgetown University, “but there is one area where I pause, and it has to do with nuclear command and control.” Referring to the article’s assertion that an automated U.S. nuclear launch ability is needed, he said, “I read that. And my immediate answer is, ‘No. We do not.’”8

Shanahan indicated that his organization was moving to integrate AI technologies into a wide array of non-nuclear capabilities, including command-and-control functions. Indeed, JAIC and other military components are moving swiftly to develop automated command-and-control systems and to ready them for use by regular combat forces. Initially, these systems will be employed by conventional forces, but the Pentagon fully intends to merge them over time with their nuclear counterparts.

Multidomain Command and Control

The Pentagon’s principle mechanism for undertaking this vast enterprise is called the Joint All-Domain Command and Control (JADC2) project. Being overseen by the Air Force, it is a computer-driven system for collecting sensor data from myriad platforms, organizing that data into manageable accounts of enemy positions, and delivering those summaries at machine speed to all combat units engaged in an operation. “First and foremost, you gotta connect the joint team,” General David L. Goldfein, Air Force chief of staff, said in January. “We have to have access to common data so that we can operate at speed and bring all domain capabilities against an adversary.”9

The JADC2 enterprise is said to be a core element of the Pentagon’s emerging strategy for U.S. victory in the fast-paced wars of the future. Called All-Domain Operations, the new strategy assumes seamless coordination among all elements of the U.S. military. General John E. Hyten, vice chairman of the Joint Chiefs of Staff, explained that the strategy combines “space, cyber, deterrent [conventional and nuclear forces], transportation, electromagnetic spectrum operations, missile defense—all of these global capabilities together … to compete with a global competitor and at all levels of conflict.”10

In moving forward on this, the Pentagon’s initial emphasis is likely to be on “data fusion,” or the compression of multiple sensor inputs into concise summaries that can be rapidly communicated to and understood by commanders in the field. Over time, however, the JADC2 project is expected to incorporate AI-enabled decision-support systems, or algorithms intended to narrow down possible responses to enemy moves and advise those commanders on the optimal choice.

This all matters because the Defense Department has indicated that the JADC2 system, while primarily intended for use by non-nuclear forces, will eventually be integrated with the nuclear command and control system now being overhauled. In his interview, Hyten was asked if the emerging JADC2 architecture was going to inform development of the remodeled NC3. He responded, “Yes. The answer is yes. But it’s important to realize that JADC2 and NC3 are intertwined.” Hyten, who formerly served as commander of the U.S. Strategic Command, added that some NC3 elements will have to be separated from the JADC2 system “because of the unique nature of the nuclear business.” Nevertheless, “NC3 will operate in significant elements of JADC2,” and, as a result, “NC3 has to inform JADC2 and JADC2 has to inform NC3.”11

Stripped of jargon and acronyms, Hyten is saying that the automated systems now being assembled for the U.S. multidomain command-and-control enterprise will provide a model for the nation’s nuclear command-and-control system, or be incorporated into the system, or both. In a future crisis, moreover, data on conventional operations being overseen by the JADC2 system will automatically be fed into NC3 computerized intelligence-gathering systems, possibly altering their assessment of the nuclear threat and leading to a heightened level of alert and possibly a greater risk of inadvertent or precipitous nuclear weapons use.

Parallel Developments Elsewhere

While the United States is proceeding with plans to modernize and automate its nuclear command-and-control system, other nuclear-armed nations, especially China and Russia, are also moving in this direction. It is conceivable, then, that a time could come when machines on all sides will dictate the dynamics of a future nuclear crisis and possibly determine the onset and prosecution of a nuclear war.

Former Soviet Colonel Stanislav Petrov sits at home in 2004 in Moscow. Petrov helped avert a possible U.S.-Soviet nuclear exchange in 1983, when he doubted the validity of an electronic warning that a U.S. missile attack was underway.  (Photo: Scott Peterson/Getty Images)Russia’s pursuit of NC3 automation began during the Soviet era, when senior leaders, fearing a “decapitating” attack on the Soviet leadership as part of a preemptive U.S. first strike, ordered the development of a “dead hand” system intended to launch Soviet missiles even in the absence of instructions to do so from Moscow. If the system, known as Perimeter, detected a nuclear explosion and received no signals from Moscow (indicating a nuclear detonation there), it was programmed to inform nuclear launch officers who, in turn, were authorized to initiate retaliatory strikes without further instruction.12 According to Russian media accounts, the Perimeter system still exists and employs some form of AI.13

Today, Russia’s leadership appears to be pursuing a strategy similar to that being pursued in the United States, involving increased reliance on AI and automation in nuclear and conventional command and control. In 2014, Russia inaugurated the National Defense Control Center (NDCC) in Moscow, a centralized command post for assessing global threats and initiating whatever military action is deemed necessary, whether nuclear or non-nuclear. Like the U.S. JADC2 system, the NDCC is designed to collect information on enemy moves from multiple sources and provide senior officers with guidance on possible responses.14

China is also investing in AI-enabled data fusion and decision-support systems, although less is known about its efforts in this area. One comprehensive study of Chinese thinking on the application of AI to warfare reported that Chinese strategists believe that AI and other emerging technologies will transform the nature of future combat, and so, to achieve victory, the People’s Liberation Army must seek superiority in this realm.15

The Perils of Heedless Automation

There are many reasons to be wary of increasing the automation of nuclear command and control, especially when it comes to computer-assisted decision-making. Many of these technologies are still in their infancy and prone to malfunctions that cannot easily be anticipated. Algorithms that have developed through machine learning, a technique whereby computers are fed vast amounts of raw data and are “trained” to detect certain patterns, can become very good at certain tasks, such as facial recognition, but often contain built-in biases conveyed through the training data. These systems also are prone to unexplainable malfunctions and can be fooled, or “spoofed,” by skilled professionals. No matter how much is spent on cybersecurity, moreover, NC3 systems will always be vulnerable to hacking by sophisticated adversaries.16

AI-enabled systems also lack an ability to assess intent or context. For example, does a sudden enemy troop redeployment indicate an imminent enemy attack or just the normal rotation of forces? Human analysts can use their sense of the current political moment to help shape their assessment of such a situation, but machines lack that ability and may tend to assume the worst.

This aspect of human judgment arose in a famous Cold War incident. In September 1983, at a time of heightened tensions between the superpowers, a Soviet nuclear watch officer, Lieutenant Colonel Stanislav Petrov, received an electronic warning of a U.S. missile attack on Soviet territory. Unsure of the accuracy of the warning, he waited before informing his superiors of the strike and eventually told them he believed it was a computer error, as proved to be the case, thus averting a possible nuclear exchange. Machines are not capable of such doubts or hesitations.17

Another problem is the lack of real world data for use in training NC3 algorithms. Other than the two bombs dropped on Japan at the end of World War II, there has never been an actual nuclear war and therefore no genuine combat examples for use in devising reality-based attack responses. War games and simulations can be substituted for this purpose, but none of these can accurately predict how leaders will actually behave in a future nuclear showdown. Therefore, decision-support programs devised by these algorithms can never be fully trusted. “Automated decision-support systems … are only as good as the data they rely on. Building an automated decision-support tool to provide early warning of a preemptive nuclear attack is an inherently challenging problem because there is zero actual data of what would constitute reliable indicators of an imminent preemptive nuclear attack.”18

An equal danger is what analysts call “automation bias,” or the tendency for stressed-out decision-makers to trust the information and advice supplied by advanced computers rather than their own considered judgment. For example, a U.S. president, when informed of sensor data indicating an enemy nuclear attack and under pressure to make an immediate decision, might choose to accept the computer’s advice to initiate a retaliatory strike rather than consider possible alternatives, such as with Petrov’s courageous Cold War action. Given that AI data systems can be expected to gain ever more analytical capacity over the coming decades, “it is likely that humans making command decisions will treat the AI system’s suggestions as on a par with or better than those of human advisers,” a 2018 RAND study noted. “This potentially unjustified trust presents new risks that must be considered.”19

Compounding all these risks is the likelihood that China, Russia, and the United States will all install automated NC3 systems but without informing each other of the nature and status of these systems. Under these circumstances, it is possible to imagine a “flash war,” roughly akin to a “flash crash” on Wall Street, that is triggered by the interaction of competing corporate investment algorithms. In such a scenario, the data assessment systems of each country could misinterpret signs of adversary moves and conclude an attack is imminent, leading other computers to order preparatory moves for a retaliatory strike, in turn prompting the similar moves on the other side, until both commence a rapid escalatory cycle ending in nuclear catastrophe.20

Given these multiple risks, U.S. policymakers and their Chinese and Russian counterparts should be very leery of accelerating NC3 automation. Indeed, Shanahan acknowledged as much, saying nuclear command and control “is the ultimate human decision that needs to be made…. We have to be very careful,” especially given “the immaturity of technology today.”

Exercising prudence in applying AI to nuclear command and control is the responsibility, first and foremost, of national leaders of the countries involved. In the United States, military officials have pledged to expose all new AI applications to rigorous testing, but Congress needs to play a more active role in overseeing these endeavors and questioning the merits of proposed innovations. At the same time, all three nuclear powers would benefit from mutual consultations over the perils inherent in increased NC3 automation and what technical steps might be taken to reduce the risk of inadvertent or accidental war. This could occur as an independent venture or in the context of the irregular strategic stability talks held by senior U.S. and Russian representatives.

There is no more fateful decision a leader can make than to order the use of nuclear weapons. Ideally, such a decision will never have to be made. But so long as nuclear weapons exist, humans, not machines, must exercise ultimate control over their use.



1. U.S. Congressional Budget Office, “Projected Costs of U.S. Nuclear Forces, 2019 to 2028,” January 2019, https://www.cbo.gov/system/files/2019-01/54914-NuclearForces.pdf.

2. For an account of nuclear control accidents during the Cold War, see Scott D. Sagan, The Limits of Safety: Organizations, Accidents, and Nuclear Weapons (Princeton: Princeton University Press, 1993).

3. Michael T. Klare, “An ‘Arms Race in Speed’: Hypersonic Weapons and the Changing Calculus of Battle,” Arms Control Today, June 2019.

4. Michael T. Klare, “Cyber Battles, Nuclear Outcomes? Dangerous New Pathways to Escalation,” Arms Control Today, November 2019.

5. Kelley M. Sayler, “Artificial Intelligence and National Security,” CRS Report, R45178, November 21, 2019, pp. 12–13, 28–29.

6. Ibid., p. 13.

7. Adam Lowther and Curtis McGiffin, “America Needs a ‘Dead Hand,’” War on the Rocks, August 16, 2019, https://warontherocks.com/2019/08/america-needs-a-dead-hand/.

8. Sydney J. Freedberg Jr., “No AI for Nuclear Command and Control: JAIC’s Shanahan,” Breaking Defense, September 25, 2019, https://breakingdefense.com/2019/09/no-ai-for-nuclear-command-control-jaics-shanahan/.

9. Theresa Hitchens, “New Joint Warfighting Plan Will Help Define ‘Top Priority’ JADC2: Hyten,” Breaking Defense, January 29, 2020, https://breakingdefense.com/2020/01/new-joint-warfighting-plan-will-help-define-top-priority-jadc2-hyten/.

10. Ibid.

11. Colin Clark, “Nuclear C3 Goes All Domain: Gen. Hyten,” Breaking Defense, February 20, 2010, https://breakingdefense.com/2020/02/nuclear-c3-goes-all-domain-gen-hyten/.

12. Michael C. Horowitz, Paul Scharre, and Alexander Velez-Green, “A Stable Nuclear Future? The Impact of Autonomous Systems and Artificial Intelligence,” December 2019, pp. 12–13, https://arxiv.org/ftp/arxiv/papers/1912/1912.05291.pdf.

13. See, for example, Edward Geist and Andrew J. Lohn, “How Might Artificial Intelligence Affect the Risk of Nuclear War?” RAND Corp., 2018, p. 10, https://www.rand.org/content/dam/rand/pubs/perspectives/PE200/PE296/RAND_PE296.pdf.

14. See Tyler Rogoway, “Look Inside Putin’s Massive New Military Command and Control Center,” Jalopnik, November 19, 2015, https://foxtrotalpha.jalopnik.com/look-inside-putins-massive-new-military-command-and-con-1743399678.

15. Elsa B. Kania, “Battlefield Singularity: Artificial Intelligence, Military Revolution, and China’s Future Military Power,” Center for a New American Security, November 2017, https://s3.amazonaws.com/files.cnas.org/documents/Battlefield-Singularity-November-2017.pdf?mtime=20171129235805.

16. See Sayler, “Artificial Intelligence and National Security,” pp. 29–33.

17. On the Petrov incident, see Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W.W. Norton, 2018), pp. 1–2.

18. Horowitz, Scharre, and Velez-Green, “Stable Nuclear Future?” p. 17.

19. Geist and Lohn, “How Might Artificial Intelligence Affect the Risk of Nuclear War?” p. 18.

20. Scharre, Army of None, pp. 199–230.

Michael T. Klare is a professor emeritus of peace and world security studies at Hampshire College and senior visiting fellow at the Arms Control Association. This article is the fourth in the “Arms Control Tomorrow” series, in which he considers disruptive emerging technologies and their implications for war-fighting and arms control.

The Trump administration’s nuclear plans raise troubling questions about the role of machines in deciding humanity’s fate in a future nuclear showdown.

Updating the CWC: How We Got Here and What Is Next

April 2020
By Stefano Costanzi and Gregory D. Koblentz

In late June 2018, Charlie Rowley and Dawn Sturgess, residents of Amesbury in the United Kingdom, discovered a discarded perfume bottle. Just 15 minutes after Sturgess sprayed her wrists to test the scent at home, she became ill and, nine days later, became the only fatality from a chemical weapons agent used in the UK. Rowley, exposed to a lesser degree, survived after multiple hospitalizations.

Flowers are left in memory of Dawn Sturgess at the Salisbury, UK, facility where she died after being exposed to a nerve agent in 2018. Not targeted, she was the only death from the use of the chemical weapon agent. (Photo: Matt Cardy/Getty Images)UK authorities determined that the chemical in the perfume bottle belonged to a class of nerve agents known as Novichoks, one of which had just been used in the attempted assassination of Sergei Skripal, a former Russian agent who defected to the UK, in the nearby town of Salisbury in March 2018. That attack also poisoned Skripal’s daughter and two police officers. The UK has accused Russia of being behind the plot and has charged two Russian intelligence officers with conducting the attack. In response, the UK and its allies expelled more than 100 Russian diplomats suspected of being spies and imposed sanctions on Russia and government officials who were involved in the assassination plot.

The appearance of the Novichok agent in two UK towns also triggered an update to the Chemical Weapons Convention (CWC), a 193-member treaty that bans the development, production, stockpiling, and use of chemical weapons. CWC parties agreed in November 2019 to add Novichok agents to the lists of chemicals subject to the treaty’s declaration requirements and verification regime. In particular, they agreed to add Novichoks, as well as another class of chemical warfare agents belonging to a group of chemicals known as carbamates, to Schedule 1, a list of chemicals with few or no known uses beyond serving as chemical warfare agents or their precursors. These chemicals are subject to the most stringent restrictions and declaration requirements. The path to this amendment, the first ever modification to the treaty, was difficult to navigate.

Novichoks and Carbamates

Novichok (“newcomer” in Russian) is an umbrella term that collectively includes several families of nerve agents developed by the Soviet Union during the Cold War under the scope of a chemical weapons program known as Foliant.1 The chemical structure of these organophosphorus nerve agents, also called A-series agents, was first publicly revealed in 2009 in the book State Secrets by Vil Mirzayanov, a former analytical chemist in the Soviet chemical weapons program turned whistleblower.2 The agents work by inhibiting the enzyme acetylcholinesterase, which is essential for the functioning of the nervous system, with deadly consequences.3 Russia has consistently denied researching or producing the agents or being involved in the attack on the Skripals.

Although Western governments were aware of the Foliant program and Novichok agents by the 1990s, they did not seek to include these agents in the CWC negotiations for fear of compromising sources and methods and the proliferation risk associated with publicizing a new class of nerve agents. As a result, when the CWC was opened for signature in 1993, its original Schedule 1 omitted the Novichok families.

Carbamates are an entirely different class of chemicals. These carbamates are not chemically related to Novichoks and, unlike traditional nerve agents, are not organophosphorus compounds. Yet, the biochemical bases of their toxicity, interfering with acetylcholinesterase, are similar to those of organophosphorus nerve agents such as sarin, VX, and Novichoks. Carbamates were researched as chemical warfare agents by the United States during the Cold War, but never developed into chemical weapons.4 Probably for this reason, carbamates were omitted from the CWC’s original Schedule 1.

The Convoluted Road to Amendment

As part of its prohibitions, the CWC bans the intentional use of any chemical, through its toxicity, to harm humans or animals. To support its implementation and its verification regime, the CWC contains in its Annex on Chemicals three schedules of chemical warfare agents and precursors for their synthesis.5 Schedule 1 is reserved for chemicals and their precursors that have primarily military uses. The other two schedules identify chemicals that can have weapons applications but also have considerable peaceful uses that are not prohibited by the CWC. Schedule 2 covers agents that can be used in smaller quantities, and Schedule 3 addresses chemicals that are widely used for legitimate commercial purposes. Each schedule is also composed of a part A, which lists chemical warfare agents, and a part B, which lists precursors for their synthesis.

In the aftermath of the Salisbury incident, the director-general of the CWC’s implementing agency, the Organisation for the Prohibition of Chemical Weapons (OPCW), took a first step. He solicited the Scientific Advisory Board (SAB), a subsidiary body that provides specialized advice on science and technology issues to the organization, with collecting information from public literature on new toxic chemicals that could act as nerve agents. The director-general also asked the treaty’s members to voluntarily share information with the SAB.6 Although the SAB report has not been released publicly, the SAB reportedly found no peaceful uses for Novichok agents.7

The Salisbury events also prompted the submission of two competing proposals to amend Schedule 1. The first came from Canada, the Netherlands, and the United States on October 25, 2018.8 This joint proposal sought to add to Schedule 1 two of the families of Novichok agents described by Mirzayanov, technically, a family of fluorophosphonates and a family of fluorophosphates, both with amidine branches attached to the core of the molecules. The agent reportedly employed in the Salisbury incident, A-234, belongs to the latter family.9

The second proposal, originally comprising five distinct elements, was submitted by Russia on November 30, 2018. Although the OPCW did not publicly disclose the details of this proposal, its content was revealed when the U.S. Department of Commerce Bureau of Industry and Security sought public comments on the potential impact on commercial activities of the proposed additions to Schedule 1 of the CWC.10 The first two elements of the Russian proposal contained a limited subset of the two families listed by the joint proposal. The other three elements contained chemicals not covered by the joint proposal. In particular, the third element comprised a single fluorophosphonate compound, which had also been described by Mirzayanov.11 Unlike the agents covered by the joint proposal, this chemical has a guanidine branch attached to the core of the molecule, rather than an amidine branch.

The fourth element comprised carbamates, which, as mentioned, are an entirely different class of chemicals. Specifically, this element included two families of carbamates that had been researched as chemical warfare agents by the United States during the Cold War but never developed into chemical weapons.12

Finally, the fifth element of the Russian proposal comprised a group of chemicals, technically alkylphosphorofluoridates with carbonimidic branches, that had been described by some sources as alternative structures for Novichok agents.13

The OPCW’s governing body, the Executive Council, recommended by consensus the adoption of the joint proposal on January 14, 2019. The council examined the Russian proposal on February 25, 2019, and did not recommend its adoption. According to a summary of the meeting, “[T]he rejection of the decision resulted from the states parties failing to reach a consensus on the fifth proposal, due to disagreements on whether the chemicals within the proposal were consistent with the guidelines in the Convention for Schedule 1A.”14 In April 2019, within the 90-day, treaty-specified deadline, Russia submitted its formal objection to the joint proposal, thus preventing the council’s recommendation from becoming final. As a result, the joint proposal and the Russian proposal were forwarded to the annual OPCW meeting of all states-parties for a final decision.

UK Defense Ministry official Annabel Goldie speaks to OPCW's Conference of States Parties on Nov. 25, 2019. She called for the adoption of a joint proposal to amend the CWC's list of most lethal chemical agents. (Photo: OPCW)Before that Conference of the States Parties was set to meet in The Hague on November 25–29, 2019, both sides engaged in quiet diplomacy behind the scenes. On April 30, 2019, Russia proposed quadripartite consultations with the sponsors of the joint proposal to merge the two proposals into a “compromise” document. The joint sponsors refused this offer and insisted that the conference vote on the proposals separately. In light of this stance and the Technical Secretariat’s determination that the chemicals included in the fifth element of the Russian proposal did not meet the criteria for inclusion in Schedule 1, Russia recognized that its effort to include the fifth element of its proposal would not gain support among other states-parties. In September, Russia modified its proposal to drop the fifth element. This move enabled the conference to approve both proposals by consensus.15 In light of the growing polarization within the OPCW due to Russian support for Syria over its continued use of chemical weapons and Russia’s own use of chemical weapons in Salisbury, the adoption of such a historic change to the CWC by consensus was a victory for diplomacy and international cooperation.

Four New Entries

The chemicals that will be added to CWC Schedule 1 are listed below.16 All of the new entries will be in part A of Schedule 1.17

Entry 13 (CWC 1A13). A large family of Novichok agents comprised of fluorophosphonates with amidine branches, this is the first Novichok family listed in the joint proposal, which has a wider scope than the first element of the Russian proposal.

Entry 14 (CWC 1A14). A large family of Novichok agents comprised of fluorophosphates with amidine branches, this is the second Novichok family listed in the joint proposal, which has a wider scope than the second element of the Russian proposal. Notably, this is the Novichok family to which the A-234 agent reportedly employed in the Salisbury incident belongs.

Entry 15 (CWC 1A15). A single additional Novichok agent not covered by the joint proposal, namely a fluorophosphonate compound with a guanidine branch, this is the third element of the Russian proposal.

Entry 16 (CWC 1A16). Two families of carbamate agents, namely quaternaries and bisquaternaries of dimethylcarbamoyloxypyridines, this is the fourth element of the Russian proposal.

The amendment will enter into force on June 7, 2020, 180 days after the UN secretary-general received notification of the decision reached by the Conference of the State Parties. The addition of these nerve agents to Schedule 1 has the effect of subjecting them to the stringent CWC verification regime and declaration requirements. In particular, in accordance with Part VI of the CWC Verification Annex, states-parties that are operating, or intend to operate, a single small-scale facility for the production of the newly scheduled chemicals will have to provide the OPCW Secretariat a detailed description of the facility and its location. Similarly, states-parties will have to notify the secretariat of the presence of the newly scheduled chemicals in “other facilities for protective purposes…and other facilities for research, medical or pharmaceutical purposes.” These facilities, if already existing, will be promptly inspected by the OPCW, in line with the CWC verification regime. Moreover, the projected activities to be conducted at the reported facilities will have to be declared in advance to the OPCW on an annual basis.18 An exception are laboratories that synthesize an aggregate quantity of Schedule 1 chemicals of less than 100 grams per year for research, medical, or pharmaceutical purposes, which are not subject to the declaration and verification obligations.

Treaty members that have dismantled facilities that have engaged in work with the newly added Schedule 1 chemicals will not have to declare those facilities. This applies, for instance, to the Shikhany branch of the State Scientific Research Institute of Organic Chemistry and Technology, which was the Soviet Union’s primary chemical weapons research and development center. According to former Soviet chemical weapons scientists, Novichoks were originally developed there beginning in the 1970s.19 In April 2018, only a month after the UK accused Russia of being responsible for the Salisbury attack, Russia announced that it was going to demolish this chemical weapons center.20 The decision was ostensibly made in the wake of Russia’s completion of the destruction of its declared chemical weapons stockpile under OPCW supervision. Yet, if the chemical weapons facility at Shikhany is demolished by the time the amendment goes into effect in June 2020, then it will also relieve Russia of having to declare the facility under the new verification requirements.

Where to Go Now

The addition of these Novichok and carbamate agents to CWC Schedule 1 is very welcome, as they are chemicals endowed with very high toxicity, which has been reported to be comparable to or exceeding that of the potent nerve agent VX.21 Yet, recent computational studies are not in agreement with the high toxicity values reported by Mirzayanov.22

To further strengthen the CWC, its members should add more chemicals to the CWC schedules for the Novichok and carbamate classes. In particular, it is advisable to further amend the CWC schedules to cover the entire family of chemicals to which CWC 1A15 belongs. This is in line with the approach taken for entries CWC 1A13 and CWC 1A14, both of which contain a large family of chemicals, and will avoid dangerous loopholes that could allow proliferators to develop analogs of CWC 1A15 not covered by Schedule 1. It is also advisable to add to the CWC schedules another family of Novichok agents described by Mirzayanov, which still remains uncovered by the CWC schedules.23 The missing Novichok family is similar to the one to which CWC 1A15 belongs. They are both organophosphorus compounds with guanidine branches, but although the CWC 1A15 family is a family of phosphonates, the missing family is a family of phosphates. This parallels the relationship between the newly added entries CWC 1A13 and CWC 1A14, both of which describe organophosphorus compounds with amidine branches, phosphonates in the first case and phosphates in the second case.

Concerning carbamate nerve agents, a wealth of compounds whose structures do not fit within the two families described by the new Schedule 1 entry CWC 1A16 have been reported. Some of them are allegedly as potent as VX or more so.24 Hence, it is advisable to conduct a thorough survey of the known carbamate nerve agents and add the missing families to the CWC schedules.

To support the CWC verification mandates, it will be important to update the OPCW Central Analytical Database, ensure that inspectors are trained and equipped to test for Novichoks and carbamates during on-site inspections, and prepare OPCW-designated laboratories to test for these molecules.

A Resilient Treaty

The addition of these chemicals to the CWC is a testament to the strength and resilience of the treaty. The CWC has unlimited duration and, as a consequence, must remain current and relevant as the years pass, a concept known as qualitative universality.25 Amendments to the CWC schedules, like the very first one just witnessed, are indeed key tools to ensure that the pact’s declaration requirements and verification regime remain current and adapt to the changing landscape of the chemical weapons threat.

The CWC’s ability to remain relevant as the years pass and situations change goes beyond the possibility of being amended. In fact, even prior to the addition of Novichoks and carbamates to Schedule 1, their use as chemical weapons was still prohibited by the treaty: according to the CWC, any weapon that exploits the toxicity of chemicals to intentionally kill or harm humans or animals is considered a chemical weapon. This principle, known as the general-purpose criterion, is another fundamental pillar of CWC resiliency. The CWC focuses on the prohibition of the purposes for which technologies may be applied, not the technology itself. This is clearly stated in Article II, which says all “toxic chemicals and their precursors, except where intended for purposes not prohibited” under the CWC are to be considered chemical weapons.

Since the time the CWC was negotiated and entered into force, the chemical weapons landscape has changed significantly.26 The recently adopted changes to Schedule 1 will play a key role in strengthening the ability of the CWC to continue to serve as the linchpin in the international regime for preventing the proliferation of chemical weapons.



1. Stefano Costanzi and Gregory D. Koblentz, “Controlling Novichoks After Salisbury: Revising the Chemical Weapons Convention Schedules,” The Nonproliferation Review, Vol. 26, Nos. 5–6 (2019).

2. Vil S. Mirzayanov, State Secrets: An Insider’s Chronicle of the Russian Chemical Weapons Program (Denver: Outskirts Press, 2008), pp. 142–149.

3. Stefano Costanzi, John-Hanson Machado, and Moriah Mitchell, “Nerve Agents: What They Are, How They Work, How to Counter Them,” ACS Chemical Neuroscience, May 16, 2018, pp. 873–885; Kloske Marcin and Zygfryd Witkiewicz, “Novichoks—The A Group of Organophosphorus Chemical Warfare Agents,” Chemosphere, Vol. 221 (April 2019), pp. 672–682.

4. Hank Ellison, Handbook of Chemical and Biological Warfare Agents, 2nd ed. (Boca Raton, FL: CRC Press, 2007), pp. 105–139.

5. Chemical weapons are weapons that kill or harm through the toxicity of chemicals. Chemical warfare agents are the toxic chemicals on which chemical weapons are based.

6. Organisation for the Prohibition of Chemical Weapons (OPCW) Technical Secretariat, “Note by the Director-General: Request for Information From States Parties on New Types of Nerve Agents,” S/1621/2018, May 2, 2018.

7. OPCW Scientific Advisory Board, “Report of the Scientific Advisory Board at Its Twenty-Eighth Session, 10–14 June 2019,” SAB-28/1, June 14, 2019, p. 17.

8. OPCW Executive Council, “Decision: Recommendation for a Change to Schedule 1 of the Annex on Chemicals to the Chemical Weapons Convention,” EC-M-62/DEC.1, January 14, 2019.

9. Mark Urban, The SKRIPAL Files: The Life and Near Death of a Russian Spy (London: Macmillan, 2018), p. 229.

10. Bureau of Industry and Security, U.S. Department of Commerce, “Impact of Proposed Additions to the ‘Annex on Chemicals’ to the Chemical Weapons Convention (CWC) on Legitimate Commercial Chemical, Biotechnology, and Pharmaceutical Activities Involving ‘Schedule 1’ Chemicals (Including Schedule 1 Chemicals Produced as Intermediates),” 84 Fed. Reg. 40389 (August 14, 2019).

11. Mirzayanov, State Secrets, p. 145.

12. Ellison, Handbook of Chemical and Biological Warfare Agents, pp. 105–139.

13. Ibid., pp. 37–42; Steven L. Hoenig, Compendium of Chemical Warfare Agents (New York: Springer, 2007), pp. 79–88.

14. OPCW Scientific Advisory Board, “Summary of the Third Meeting of the Scientific Advisory Board Temporary Working Group on Investigative Science and Technology,” SAB-28/WP.3, June 4, 2019, pp. 10–11.

15. OPCW, “Decision: Changes to Schedule 1 of the Annex on Chemicals to the Chemical Weapons Convention,” C-24/DEC.5, November 27, 2019; OPCW, “Decision: Technical Change to Schedule 1(A) of the Annex on Chemicals to the Chemical Weapons Convention,” C-24/DEC.4, November 27, 2019.

16. OPCW Technical Secretariat, “Note by the Technical Secretariat: Consolidated Text of Adopted Changes to Schedule 1 of the Annex on Chemicals to the Chemical Weapons Convention,” S/1820/2019, December 23, 2019.

17. For a structurally annotated table of the chemicals that will be added to CWC Schedule 1, see Costanzi Research, “A Historical Event: Chemicals Added to CWC Schedule 1,” n.d., https://costanziresearch.com/cw-nonproliferation/cw-control-lists/cwc-schedule-1-amendment/ (accessed March 18, 2020).

18. OPCW Technical Secretariat, “Note by the Technical Secretariat: Guidance for States Parties on Article VI Declaration Obligations and Inspections Following Entry Into Force of Changes to Schedule 1 of the Annex on Chemicals to the Chemical Weapons Convention,” S/1821/2019, December 31, 2019.

19. Mirzayanov, State Secrets, pp. 142-145; Ilia Rozhdestvenskiy, “The Secret Folio: How the Nerve Agent ‘Novichok’ Was Created and Tested on Human Beings,” The Project, September 12, 2018, https://www.proekt.media/narrative/test-novichok-eng/.

20. “The Institute, Which Allegedly Developed the ‘Novice,’ Demolishes Its Buildings,” RIA Novosti, April 27, 2018, https://ria.ru/amp/20180427/1519528336.html (in Russian).

21. Mirzayanov, State Secrets, pp. 142–149; Ellison, Handbook of Chemical and Biological Warfare Agents, pp. 105–139.

22. Hanusha Bhakhoa, Lydia Rhyman, and Ponnadurai Ramasami, “Theoretical Study of the Molecular Aspect of the Suspected Novichok Agent A234 of the Skripal Poisoning,” Royal Society Open Science, Vol. 6, No. 2 (2019): 181831; Lars Carlsen, “After Salisbury Nerve Agents Revisited,” Molecular Informatics, Vol. 38, Nos. 8–9 (2019): 1800106.

23. Costanzi and Koblentz, “Controlling Novichoks After Salisbury.”

24. Ellison, Handbook of Chemical and Biological Warfare Agents, pp. 105–139.

25. Jean Pascal Zanders, “The Chemical Weapons Convention and Universality: A Question of Quality Over Quantity?” Disarmament Forum, No. 4 (2002), p. 23.

26. See Rebecca K.C. Hersman, Suzanne Claeys, and Cyrus A. Jabbari, “Rigid Structures, Evolving Threat: Preventing the Proliferation and Use of Chemical Weapons,” Center for Strategic and International Studies, December 2019, https://csis-prod.s3.amazonaws.com/s3fs-public/publication/191218_Hersman_RigidStructures_WEB.pdf?LHDLRedZ2X3eEhGwYz7Ko8lXzSs1hEEO.

Stefano Costanzi is an associate professor of chemistry at American University. Gregory D. Koblentz is an associate professor and director of postgraduate programs in biodefense at George Mason University's Schar School of Policy and Government.

New families of chemical agents soon will be explicitly prohibited by the Chemical Weapons Convention.

Assessing U.S. Plans to Modernize Its Nuclear Weapons

April 2020
By Allison Bawden

Thank you for inviting me to discuss the Government Accountability Office (GAO) views on National Nuclear Security Administration (NNSA) plans for modernizing the nation’s nuclear security enterprise and aligning its efforts with the Defense Department to modernize delivery systems. These remarks should be viewed as helping the NNSA set itself up for success.

 (U.S. Senate Committee on Environment and Public Works/Flickr)The nuclear security enterprise is embarking on its most ambitious level of effort since the Cold War era. The NNSA is currently managing four weapons modernization programs, proposing a fifth, and undertaking infrastructure projects that affect every strategic material and component used in nuclear weapons.

Today, I will discuss the schedule risks presented by the integrated nature of NNSA and Defense Department nuclear modernization efforts, the budget and schedule estimates for implementing the overall program, and the importance of the NNSA setting priorities among its efforts in the event of budget shortfalls or cost or schedule overruns.

First, on the schedule risks, because NNSA modernization is highly integrated, any delay could have a significant cascading effect on the overall effort. Here are three scenarios. First, weapons programs depend on the completion of certain infrastructure projects. For example, the W87-1 program will require all new components, including plutonium pits. The construction schedule for pit facilities is aggressive, and a delay could have an impact on the schedule for the weapons programs it supports.

Also, because the NNSA uses the same production infrastructure for each weapons program and capacity is limited, each program schedule can impact the next. In addition, NNSA weapons program schedules must remain aligned with the schedules for the Defense Department’s new delivery systems to ensure essential testing is completed at critical times. This is especially true for the W80-4 warhead and the Air Force’s long-range standoff missile, as well as the W87-1 warhead and the Air Force’s intercontinental ballistic missile replacement. The current schedules have little margin for delay.

Second, on budget and schedule estimates, in the past, the GAO has been critical of the NNSA’s performance on a number of weapons modernization programs and major construction projects. We identified poor planning and overly optimistic assumptions about performance that contributed to cost and overruns, schedule delays, and program and project cancellations. The NNSA has made improvements to management controls for these efforts, especially around cost and schedule estimating, and it is increasingly paying attention to program and project management capacity.

As the NNSA undertakes an increased scope of work, it is essential that its overall plans reflect realistic cost and schedule estimates, rather than best-case estimates. For example, although the NNSA has not yet fully developed its schedule for constructing pit facilities, its own analysis of alternatives suggests current dates will be difficult to achieve.

Finally, on setting priorities, the president’s fiscal year 2021 budget request includes a 25 percent increase for the NNSA modernization program and anticipates sustaining this increased funding level for at least the next five years. In 2017 the GAO reviewed the NNSA’s long-term plans for its modernization program. At the time, we found that the NNSA planned to defer work to a period beyond its five-year programming window. We concluded that these deferrals created a significant bow wave of funding needs in future years to undertake the simultaneous weapons programs and construction projects it planned. The requested budget increase for next year suggests this bow wave has arrived.

Requesting a funding increase is one way to address the bow wave and maintain the overall scope of planned modernization efforts. However, actual funding in future years could fall short of budget estimates, and programs or projects could and have encountered cost overruns or schedule delays.

The GAO recommended in 2017 that to increase the credibility of its modernization plans, the NNSA should develop a portfolio approach as a way to manage these risks. Such an approach would present options that could be exercised if budget or schedule risks materialized. This would include identifying programs for which starts could be deferred. Any such plan would need to be put forward in collaboration with the Defense Department. The NNSA’s most recent long-term plan includes an assessment of whether its budget requests fall within the range of its program cost estimates. However, it has not yet adopted a portfolio approach to setting its priorities, should cost or schedule risks materialize. NNSA planning could further benefit from this approach in light of the 2018 Nuclear Posture Review and additional programs it anticipates.

Adapted from testimony by Allison Bawden, director of the U.S. Government Accountability Office’s natural resources and environment team, delivered to the House Armed Services strategic forces subcommittee, March 3, 2020.

An oversight official raises concerns about the U.S. ability to modernize its nuclear forces on time and on budget.

NPT Review Conference Postponed

April 2020
By Daryl G. Kimball

The global coronavirus disease (COVID-19) pandemic has forced a postponement of the 10th review conference of the nuclear Nonproliferation Treaty (NPT), possibly until early 2021. Originally scheduled to be held at UN headquarters in New York from April 27 until May 22, the conference typically involves hundreds of representatives from most of the 191 states-parties to the treaty, as well as nongovernmental organizations and meeting support personnel. The conference caps off a five-year cycle of meetings through which states-parties review implementation and compliance with the treaty and seek agreement on action steps to overcome new challenges and to fulfill core goals and objectives.

Gustavo Zlauvinen of Argentina, president-designate of the 2020 NPT Review Conference, addresses the UN Security Council in February. (Photo: Evan Schneider/UN)In a March 26 interview with Arms Control Today, conference president-designate Gustavo Zlauvinen of Argentina said that after extensive consultations, he circulated a proposal on March 25 to NPT regional groups to postpone the review conference “until such time as conditions permit, but not later than April 2021.”

Zlauvinen said his communication explained that because the pandemic makes it impractical to hold meetings at the present time, the only way for states-parties to express their agreement on a decision to postpone the review conference is to do so in writing, and he requested responses by March 27.

All other related preparatory meetings and consultations, including plans for NPT workshops in Jordan, Mexico, and Thailand, have been postponed. A U.S.-hosted third working group meeting for the Creating an Environment for Nuclear Disarmament initiative, planned for April 8–9 in Washington, has also been postponed.

In March, as the scale and scope of the pandemic grew, Zlauvinen said he began discussing conference options with senior UN officials and NPT states parties, including the UN secretary-general on March 4. On March 13, he circulated a proposal to delegations calling for the conference to be “suspended” until a later date after a short procedural meeting on April 27 to elect a conference officers and possibly to agree on a program of work, if conditions allowed.

Given the importance of the NPT, Zlauvinen said he initially wanted to hold the meeting on April 27 to provide a strong “symbolic” start to the review conference and to “maintain the integrity” of the review process. The conference had originally been scheduled to begin the same day.

As the COVID-19 outbreak worsened in New York, however, the prospects of an in-person procedural meeting on April 27 became untenable, Zlauvinen said.

Another option explored was to use video conferencing to hold the April 27 meeting, he said, but that arrangement could have led to technical difficulties for some states, including potential problems regarding procedures and voting. He also said that there were differing views among states-parties about selecting a date for resuming the conference, with some seeking rescheduling in early 2021, others in the late-fall of 2020, while others proposed an earlier date.

According to a Xinhua news report in March, the Non-Aligned Movement of 120 member states decided on March 12 to recommend postponing the review conference to April and May of next year, saying it is too important to hold on a smaller scale at an earlier date.

Taking these factors into account, Zlauvinen came to the conclusion that he should recommend to states-parties that the meeting should be postponed until such time as it is possible for “the review conference to undertake its important work.” The UN publicly confirmed the postponement on March 27.

When the review conference will be held remains unclear. In addition to the uncertainties about the duration of the pandemic, Zlauvinen said that another complicating factor is that the UN conference facilities in New York are almost completely booked with other official events, including rescheduled events, for the next several months.

Preparations for the review conference have experienced other unexpected turns. Last year, the preparatory meeting for the 2020 review conference tapped veteran Argentine diplomat Rafael Grossi as president-designate for the meeting. In late 2019, however, he was selected to become the new director-general of the International Atomic Energy Agency (IAEA).

In his place, Argentina put forward Zlauvinen, who is a deputy foreign minister and has a long track record on nonproliferation matters. He served as the IAEA representative to the United Nations in New York for nearly eight years and was an alternate head of the IAEA’s NPT delegation from 2002 to 2009.

The NPT, which marked its 50th anniversary of entry into force on March 5, obligates the five nuclear-weapon states recognized by the treaty not to help non-nuclear-weapon states develop or acquire nuclear weapons, it obligates the non-nuclear-weapon states to forswear the pursuit of such weapons, it acknowledges the “inalienable right” of states-parties to the peaceful use of nuclear energy under safeguards, and the treaty commits states-parties to “pursue negotiations in good faith on effective measures relating to cessation of the nuclear arms race at an early date and to nuclear disarmament, and on a treaty on general and complete disarmament under strict and effective international control.”

The challenges facing the treaty have grown in recent years, especially since 2015 when NPT states-parties failed to reach agreement on a final document at that year’s review conference. Many non-nuclear-weapon states have also argued that key benchmarks agreed to at the historic 1995 review and extension conference and the successful 2010 review have not been met by the nuclear-weapon states.

At a UN Security Council session on nuclear nonproliferation issues convened by Germany in New York in February, the High Representative for Disarmament Affairs Izumi Nakamitsu warned that “relationships between states, especially nuclear-weapon states, are fractured.”

“The specter of unconstrained nuclear competition looms over us for the first time since the 1970s. We are witnessing what has been termed a qualitative nuclear arms race, one not based on numbers but on faster, stealthier and more accurate weapons. Regional conflicts with a nuclear dimension are worsening, and proliferation challenges are not receding,” she said.

“I hope the review conference can serve as a springboard for thinking on how to address the nuclear weapons challenges of our time,” Nakamitsu told the council.

“Obviously the NPT is not on the top of the priority list for our governments right now. The pandemic is, and rightly so. Once we resolve the issue of the postponement and once travel restrictions are lifted, I plan to continue the series of regional seminars on the three pillars of the NPT that my predecessor, Ambassador Grossi, started, as well as my personal consultations with states-parties in Geneva, Vienna, in New York, and in other capitals,” Zlauvinen said.

“Once we can move back into the important discussions on all of the issues relating to the NPT, I hope the states-parties can think like a community and find common solutions to common challenges,” Zlauvinen added. “The coronavirus is an invisible enemy.... Nuclear weapons are also a kind of invisible threat that most nonexperts don’t think about in their everyday lives. These are both global challenges that require global solutions.”


COVID-19 Delays Security Meetings, Treaty Inspections

A lone worker walks through the rotunda of the Vienna International Centre on March 19. The center normally houses more than 2,200 IAEA employees who have been directed to work at home during the public health crisis caused by the novel coronavirus. (Photo: Dean Calma/IAEA)


Efforts to slow the spread of the novel coronavirus disease (COVID-19), have led to a suspension of arms control and international security meetings at the UN and other institutions in Geneva, New York, Vienna, and beyond.

Most prominently, this year’s review conference for the nuclear Nonproliferation Treaty (NPT) at UN headquarters in New York has been postponed.

In Vienna, the International Atomic Energy Agency (IAEA) has delayed or deferred most events through the end of May, including the eighth review conference of the Nuclear Safety Convention. The IAEA is maintaining its nuclear safeguards operations, including in Iran. The IAEA and Comprehensive Test Ban Treaty Organization are operating remotely, following instructions of Austrian authorities.

Mandatory restrictions in Switzerland have locked down the nation, including a ban on gatherings of more than five people until mid-April. As a result, the Palais des Nations has minimized its on-site staff and cancelled upcoming events. Among them are two plenary sessions of the Conference on Disarmament, as well as a working group and second informal meeting in preparation for the sixth conference of parties to the Arms Trade Treaty.

Also delayed is the third round of consultations on a political declaration to address the humanitarian harm arising from the use of explosive weapons in populated areas.

Other meetings affected by COVID-19 include the Group of Seven summit, once scheduled to be hosted by the United States, which will now be held via video conference June 10–12.

New START Inspections Suspended

According to diplomatic and congressional sources, the United States and Russia agreed in late March to suspend on-site inspections under the 2010 New Strategic Arms Reduction Treaty until May 1. They also agreed to postpone the Bilateral Consultative Commission (BCC) meeting scheduled for March in Geneva, due to the pandemic. U.S. and Russian officials agreed to keep their full allotments of inspectors and intend to conduct two BCC meetings this year, as called for by the treaty. Information exchanges and notifications that would normally have been shared at the BCC meeting will be conducted through regular diplomatic channels instead.—LEA SCHAAD and DARYL G. KIMBALL

Reflecting public health concerns, the nuclear Nonproliferation Treaty review conference, scheduled to begin Apr. 27, has been postponed.

Iran Boosts Enriched Uranium Stockpile

April 2020
By Julia Masterson

Iran has accumulated approximately 1,021 kilograms of low-enriched uranium (LEU), the International Atomic Energy Agency (IAEA) detailed in a report released March 3. The report notes that Iran is continuing to expand its uranium-enrichment program and is now accumulating enriched material from all 1,044 first-generation IR-1 centrifuge machines at its Fordow facility and from 5,060 IR-1s and a limited number of advanced model machines at its Natanz facility.

IAEA Director-General Rafael Grossi speaks to the media on March 9. He told the agency's Board of Governors the same day that Iran has continued to allow the IAEA to conduct its activities defined by the 2015 nuclear deal. (Photo: Dean Calma/IAEA)Under the 2015 nuclear deal, known formally as the Joint Comprehensive Plan of Action (JCPOA), Iran’s LEU stockpile is capped at 300 kilograms of uranium hexafluoride gas enriched to 3.67 percent uranium-235 and is limited to output from 5,060 IR-1 centrifuges at Natanz. Three hundred kilograms of uranium hexafluoride gas equates to about 202 kilograms of uranium by weight.

Iran’s enriched uranium stockpile is now about five times larger than the JCPOA limit. The stockpile’s size shortens Iran’s breakout time, the time it would take to produce enough nuclear material for one weapon, if Tehran were to choose to pursue nuclear weapons development. How quickly Iran could produce enough fissile material for a weapon depends on several factors, including the number and type of operating centrifuges. When the JCPOA was fully implemented, Iran’s breakout time was estimated to be 12 months.

Iran’s growing uranium stockpile should not necessarily be perceived as a sprint to the bomb. Iran continues to comply with the IAEA on-site verification and monitoring activities that are designed specifically to detect higher levels of enrichment and a diversion of materials for weapons purposes, said IAEA Director-General Rafael Grossi at his agency’s Board of Governors meeting on March 9.

In addition to its comprehensive safeguards agreement with the IAEA, required of all non-nuclear-weapon states party to the nuclear Nonproliferation Treaty, Iran is obligated under the JCPOA to adhere to an additional protocol to its safeguards agreement that allows inspectors increased access and tools to verify the peaceful nature of Iran’s nuclear program. The JCPOA grants the IAEA a long-term presence in Iran and allows the agency to continuously monitor Iran’s uranium enrichment, among other things.

“The agency continues to verify the nondiversion of nuclear materials declared by Iran under its safeguards agreement,” Grossi said, adding that the agency is undertaking investigations into Iran’s undeclared nuclear activities.

Grossi also said that despite Iran’s Jan. 5 announcement that it would no longer be bound by the deal’s operational restrictions, “to date, the agency has not observed any changes to Iran’s implementation of its nuclear-related commitments under the JCPOA.”

Iran continues to comply fully with its JCPOA-related safeguards and monitoring commitments and to adhere to the deal’s prohibition on plutonium reprocessing, according to the March 3 report. Iran also has not installed any additional IR-1 centrifuges at the Natanz facility and has not taken steps to pursue construction of the Arak heavy-water reactor under its original design, in keeping with the deal’s requirements that Iran install no more than 5,060 IR-1 centrifuges at Natanz and convert the Arak reactor to produce less plutonium.

Tehran’s Jan. 5 announcement marked its fifth breach of the nuclear deal since it first began reducing its compliance with the agreement in May 2019. Nevertheless, Iran’s violations likely are not indicative of imminent nuclear weapons development but are rather an attempt by Tehran to pressure the remaining parties to the JCPOA to offer sanctions relief promised under the deal.

In response to Iran’s escalatory measures, the European members of the nuclear deal (France, Germany, and the United Kingdom) triggered the deal’s dispute resolution mechanism Jan. 14 in an effort to facilitate dialogue necessary to address Iran’s noncompliance and salvage the agreement. The dispute resolution mechanism, laid out in the text of the JCPOA, provides for a 15-day period of discussions within the governing Joint Commission, which comprises the deal’s remaining members. The time-bound period for discussions within the Joint Commission can be extended by consensus vote.

EU foreign affairs chief Josep Borrell, who chairs the Joint Commission, announced on Jan. 24 that “there is agreement that more time is needed due to the complexity of the issues involved. The timeline is therefore extended.”

The members of the Joint Commission met for the first time Feb. 26 in Vienna. The meeting was attended by representatives from China, France, Germany, Iran, Russia, the United Kingdom, and the European Union.

In his statement following that session, Borrell said that “serious concerns were expressed regarding the implementation of Iran’s nuclear commitments under the agreement,” but added that “participants also acknowledged that the reimposition of U.S. sanctions did not allow Iran to reap the full benefits arising from sanctions lifting.”

Borrell remarked that a series of expert-level discussions had taken place in recent weeks regarding Iran’s violations of the JCPOA and Washington’s reimposition of sanctions following the U.S. withdrawal from the deal in May 2018. Although he did not clarify whether the Joint Commission period of the dispute resolution mechanism would be further extended, Borrell noted that expert-level discussions would continue to move forward.

“All participants [in the Feb. 26 meeting] reaffirmed the importance of preserving the agreement, recalling that it is a key element of the global nuclear nonproliferation architecture,” Borrell said.


As Iran stores more nuclear material than allowed by the 2015 nuclear deal, it continues to allow IAEA monitoring of its nuclear activities.

IAEA Urges Iran to Cooperate

April 2020
By Kelsey Davenport

Tehran is refusing to cooperate with an International Atomic Energy Agency (IAEA) investigation into possible undeclared nuclear materials and activities in Iran, saying that the agency’s evidence is biased.

Amb. Jackie Wolcott, U.S. representative to the IAEA, attends an agency meeting in July 2019. She raised "very serious concerns" about Iran's compliance with its IAEA nuclear safeguards agreement. (Photo: Dean Calma/IAEA)IAEA Director-General Rafael Grossi laid out the agency’s attempts since January 2019 to get information from Tehran about the possible storage and use of nuclear materials at three locations in Iran in a March 3 report to the agency’s Board of Governors.

In December, Grossi said that Iran was not responding satisfactorily to IAEA questions and revealed in February that he may ask for support from the agency’s 35-member Board of Governors if Tehran continued to refuse to cooperate with IAEA requests.

Iran has “not engaged in substantive discussions” to clarify agency questions about possible use and storage of nuclear materials and has “not provide access to these locations,” Grossi said on March 9. He called on Tehran to “cooperate immediately and fully” with IAEA efforts.

As a state-party to the nuclear Nonproliferation Treaty (NPT), Iran is required to implement a safeguards agreement with the IAEA. A safeguards agreement includes a declaration detailing the country’s nuclear activities and the locations of nuclear materials. The IAEA is responsible for verifying that a country’s nuclear materials are accounted for and being used for peaceful purposes.

As part of its implementation of the 2015 nuclear deal, Iran also agreed to provisionally implement an additional protocol to its safeguards agreement. The additional protocol gives the IAEA access to additional information about a country’s nuclear program, provides expanded access for inspectors, and allows for greater use of environmental sampling to test for the presence of nuclear materials.

The March 3 report says that the IAEA requested access to two of the sites in January 2020 to take environmental samples, but Iran has not allowed inspectors to visit those locations. The agency also observed activities that appeared consistent with sanitization efforts at one of the sites, the report said.

Iran dismissed the allegations of concealment as based on false reports from countries hostile to Iran. “Any absurd claim made by any regime or individual should not be the basis of the agency’s questions,” Iranian Foreign Ministry spokesman Seyyed Abbas Mousavi said on March 11.

He may be referring to information that Israel stole from Iran in 2018 and later shared with the IAEA. Israeli Prime Minister Benjamin Netanyahu said that the documents provide details about Iran’s past nuclear weapons work and has urged the IAEA to follow up on the information.

The March 3 report did not reference material provided by any state, but noted that all safeguards-relevant information provided to the IAEA is subject to “an extensive and rigorous corroboration process.”

Based on the IAEA report and Iran’s communications with the agency, it appears that the locations in question may be storing materials from Iran’s past nuclear weapons program and are not being used for ongoing or recent illicit nuclear activities.

In a Jan. 28 letter to the IAEA, Iran said it “does not consider itself obliged to respond to such allegations” because Tehran met its obligations under the 2015 nuclear deal to cooperate with the IAEA investigation into past nuclear activities.

The 2015 agreement, known as the Joint Comprehensive Plan of Action, required Iran to comply with the IAEA investigation into what was then known as the “possible military dimensions” of Iran’s nuclear program prior to receiving any sanctions relief.

The IAEA published a report in December 2015 concluding that Iran had a nuclear weapons program prior to 2003 and that some of the activities continued through 2009, but that there was no evidence of weaponization activities after 2009 or any credible indication that nuclear materials had been diverted for those programs.

Although the report closed the IAEA investigation into Iran’s past nuclear weapons activities, the agency is still required to investigate any evidence of undeclared nuclear activities.

Jackie Wolcott, U.S. representative to the IAEA, told the IAEA Board of Governors on March 11 that the IAEA report raises “very serious concerns regarding Iran’s compliance with its safeguards obligations” and noted that Iran “could be” violating its safeguards agreement.

She said that “any further delay, denial, or deception by Iran that inhibits” IAEA work “would require that the board appropriately escalate this issue.”


Tehran has stonewalled efforts to investigate allegations that it may be storing undeclared nuclear materials or information.

Trump Still Wants Multilateral Arms Control

April 2020
By Kingston Reif and Shannon Bugos

The United States is open to meeting with the other permanent members of the UN Security Council to discuss nuclear arms control and will soon put forward a trilateral arms control proposal with Russia and China, President Donald Trump said in February. China, however, has continued to express its opposition to trilateral talks with Russia and the United States, and it has not yet responded to U.S. overtures to begin a bilateral strategic security dialogue.

President Donald Trump takes questions from the media in the White House briefing room on Feb. 29. He said the five permanent members of the UN Security Council will likely discuss nuclear arms control this year. (Photo: Alex Wong/Getty Images)At the same time, the Trump administration continued to deflect questions about its stance on the New Strategic Arms Reduction Treaty (New START), which is due to expire in 2021.

Trump told reporters on Feb. 29 that Russia, China, the United Kingdom, and France “all want to now discuss arms control” and that the leaders of those countries will likely discuss the subject at the UN General Assembly in New York in September.

Russian President Vladimir Putin first proposed the idea of a summit of those nations earlier this year to discuss a broad range of security topics, including arms control. “We have discussed this with several of our colleagues and, as far as I know, have received a generally positive response to holding a meeting of the heads of state of the five permanent members of the UN Security Council,” Putin stated during a late January trip to Israel.

In a March 5 statement commemorating the 50th anniversary of the entry into force of the nuclear Nonproliferation Treaty, Trump said he “will be proposing a bold new trilateral arms control initiative with Russia and China to help avoid an expensive arms race and instead work together to build a better, safer, and more prosperous future for all.”

Trump, who first proposed a trilateral approach to arms control nearly a year ago, said such an approach is needed because, “[o]ver the next decade, China seeks to at least double the size of its nuclear stockpile while Russia is developing expensive and destabilizing new types of delivery systems.” (See ACT, May 2019.)

Neither the president nor other officials have provided a timeline for when the administration might release a proposal.

Nevertheless, “we’re optimistic that it will be possible to engage both with Russia and with China, and to bring those bilateral engagements forward into a trilateral engagement that will ultimately result in the kind of agreement that President Trump has tasked us with trying to come to,” a senior State Department official told reporters on March 9. “So we are cautiously optimistic and hope very much to be engaged with both, not just one, of those two parties in the very near future,” the official added.

Beijing, however, has continued to express no interest in a trilateral approach. “China has repeatedly reiterated that it has no intention of participating in the so-called trilateral arms control negotiations,” a Chinese Foreign Ministry spokesman said on March 6.

In addition to pursuing trilateral talks with Russia and China, Christopher Ford, U.S. assistant secretary of state for international security and nonproliferation, invited China last December to begin a bilateral strategic security dialogue.

China has yet to respond to the offer. A Chinese Foreign Ministry spokesman said on March 6 that “China is always open to bilateral exchange with the U.S. in the field of strategic security.”

The Trump administration’s continued pursuit of arms control discussions with Russia and China comes as the clock continues to tick toward the February 2021 expiration of New START. The United States and Russia can exercise the option to extend the agreement by up to five years by mutual agreement.

The administration has yet to make its decision regarding the future of the accord. The senior State Department official said the administration is evaluating a decision on extension in the context of whether prolonging New START aids or sets back the pursuit of limits on additional types of Russian nuclear weapons not covered by the treaty and bringing China into the arms control process.

“We are not forecasting what that answer is,” the official said.

The official would not comment on the administration’s specific goals for trilateral arms control or whether the administration would consider extending New START to buy additional time to engage China.

The official added that there remains plenty of time to extend the treaty before it expires and that an extension could be quickly accomplished with “nothing more than an exchange of diplomatic notes.”

Defense Secretary Mark Esper said on Feb. 19 that if the United States chooses to extend New START, then a new agreement should “capture the new Russian strategic weapons. I also believe that the Russians should bring underneath that treaty the nonstrategic nuclear weapons. And then, of course, the Chinese.”

Moscow has expressed its readiness to extend New START immediately and without any preconditions, but warned that time is running out and any proposal to include Russian weapons not covered by the treaty would require an amendment to the agreement. (See ACT, January/February 2020.)

Vladimir Leontyev, deputy director of the Russian Foreign Ministry’s Department for Non-Proliferation and Arms Control, said in February that the United States rejected a proposal from Russia to hold “a lawyers’ meeting to…work out an understanding of the technical aspect of the extension process.”

“No follow-up agreement is in sight, and it is clear that there is no chance of producing anything meaningful over the time left” before the treaty expires, Leontyev noted.

Russia has also maintained that it “will not try to convince China” to join trilateral talks, according to Russian Foreign Minister Sergey Lavrov. “If the Americans are quite sure that it makes no sense to take any further steps on the New START…without China, let them get down to business on this all on their own,” he said on Feb. 10.

A reported hindrance to the U.S. effort to negotiate a more comprehensive replacement for New START has been the Trump administration’s inability to find a lead negotiator for the undertaking. Politico reported on Feb. 12 that the administration has offered the role to several potential candidates but struggled to find someone to take it. The Guardian reported on March 4 that the White House had chosen Marshall Billingslea, the current assistant secretary for terrorist financing at the Treasury Department, for the role.

New START caps U.S. and Russian strategic nuclear arsenals at 1,550 deployed warheads, 700 deployed missiles and heavy bombers, and 800 deployed and nondeployed missile launchers and bombers each. It also put into place a rigorous inspections and verification regime, on which the U.S. military relies for knowledge about the Russian arsenal.


The extension of New START remains up in the air as the Trump administration pushes talks with China.

U.S. Questioning Open Skies Treaty

April 2020
By Kingston Reif and Shannon Bugos

The U.S. Defense Department is halting funding to replace the Pentagon’s aging aircraft used for flights under the Open Skies Treaty until the Trump administration decides the future of U.S. participation in the pact, Defense Secretary Mark Esper told Congress in March.

Navy Lt. Bethany Baker monitors the path of an OC-135B aircraft during a January 2016 flight over Haiti unrelated to the Open Skies Treaty. The Pentagon has elected not to upgrade the aircraft as the Trump administration assesses the U.S. role in the Treaty. (Photo: Perry Aston/U.S. Air Force)“Until we make a final decision on the path forward, I am not prepared to recapitalize aircraft,” Esper told the Senate Armed Services Committee during a March 4 hearing. “We still have the means to conduct the overflights,” he added, “[b]ut at this time, we’re holding until we get better direction.”

The Trump administration has yet to make a final decision on whether to remain a party to the treaty. The administration has reportedly warned allies that it could withdraw from the agreement if its concerns about Russian noncompliance and other shortcomings are not allayed. (See ACT, January/February 2020.)

“In due course, we will be getting together to do that, to decide the best path forward for our nation,” Esper said at a Feb. 20 press briefing.

Esper has not disclosed how he is advising President Donald Trump on the agreement, but told senators that he has “a lot of concerns about the treaty as it stands now.”

The United States maintains two Boeing OC-135B aircraft based at Offutt Air Force Base in Nebraska for treaty overflight missions. Congress appropriated $41.5 million last year to continue replacement efforts for these aircraft. The Air Force was planning to seek $76 million in fiscal year 2021 to continue the replacement process, but the final budget request published in February did not include any request for funding.

U.S. lawmakers criticized the administration’s actions. “By not recapitalizing the Open Skies aircraft, we are adding risk to our aircrews,” said Rep. Don Bacon (R-Neb.) in a statement to Defense News. Sens. Bob Menendez (D-N.J.), ranking member of the Senate Foreign Relations Committee, and Jack Reed (D-R.I.), ranking member of the Senate Armed Services Committee, sent a letter to Secretary of State Mike Pompeo on Feb. 28 opposing a potential withdrawal of the United States from the Open Skies Treaty.

“If this administration moves forward with a precipitous unilateral withdrawal from the treaty the United States will be less safe and secure,” they wrote.

But Sens. Richard Burr (R-N.C.), Tom Cotton (R-Ark.), and Ted Cruz (R-Texas) sent a letter to Trump on March 12 calling on him to withdraw from the agreement.

“It is well past time to withdraw,” they wrote, citing concern about the costs of the OC-135B replacement efforts and suggesting that the United States receives few benefits from the accord.

The senators also referenced Russia’s noncompliance with the treaty as grounds for withdrawal. The United States asserts that Russia is violating the agreement by restricting observation missions over Kaliningrad to flying no more than 500 kilometers and by establishing a 10-kilometer corridor along Russia’s border with the Georgian border-conflict regions of South Ossetia and Abkhazia.

The Defense Department has responded by restricting flights over the Pacific Fleet in Hawaii and the missile defense interceptor fields in Fort Greely, Alaska.

Meanwhile, Vayl Oxford, director of the U.S. Defense Threat Reduction Agency, commented on Feb. 11 before the House Armed Services Committee that the United States has had “a lot of consultation with our treaty partners” who view Open Skies as “very valuable.”

Oxford also said that “[i]f we fly all the missions currently planned this year, it’d be the busiest Open Skies season ever.”

James Gilmore, U.S. representative to the Organization for Security and Co-operation in Europe, on March 2 described Russian support for a recent mission that took place over Kaliningrad as “very cooperative.”

Gilmore also noted that Russia will no longer raise an “objection” for the United States and its allies to “fly over one of their major exercises,” as was reported.

“However, we don’t believe that that’s good enough,” he noted. “We think that we have to be holding the Russians strictly to account on the Open Skies Treaty.”

The widening impact of the novel coronavirus disease (COVID-19) pandemic is creating further complications. Citing the “serious public health risk posed by the spread of COVID-19,” on March 16, the chair of the Open Skies Consultative Commission, Belarus, requested a suspension of all Open Skies flights until April 26, according to a statement. The other parties agreed and have notified their suspension of flights, according to diplomatic sources.

Signed in 1992, the treaty permits each state-party to conduct short-notice, unarmed, observation flights over the others’ entire territories to collect data on military forces and activities. The treaty entered into force in January 2002 and currently has 34 states-parties, including the United States and Russia. Since 2002, there have been nearly 200 U.S. overflights of Russia and about 70 overflights conducted by Russia over the United States.

The Pentagon is no longer seeking to update its aircraft used to photograph Russian military sites.


Subscribe to RSS - April 2020