May 2026
By Matthew Bunn

U.S. President Donald Trump’s April 1 address to the nation on the war on Iran offered a torrent of lies and no solutions.¹ Far from freeing the United States from “the specter of nuclear blackmail,” it appears that this war will leave a greatly weakened but embittered and harder-lined Iranian regime that is more determined than ever to build a nuclear deterrent. Although on-again, off-again talks are still underway, as of late April, Iran still retains much of the material and equipment needed to do so. That includes enough weapons-usable highly enriched uranium (HEU) for more than 10 nuclear bombs.

The issue is not just the HEU, however. After decades of effort by thousands of people, Iran has an array of material, equipment, and expertise that, although damaged, cannot be fully bombed away. The war will leave Iran’s government less able to defend itself but the danger of Iranian nuclear weapons will remain.

It is worth remembering how we got to this point. Under the 2015 Joint Comprehensive Plan of Action (JCPOA), better known as the Iran nuclear deal, Tehran agreed to stringent limits on its uranium enrichment, exported 97 percent of its enriched uranium to Russia, and accepted inspections going far beyond usual International Atomic Energy Agency (IAEA) safeguards.² But Trump pulled out of the deal in his first term, freeing Iran to produce HEU, develop more advanced centrifuges, and abandon the deal’s inspection provisions.

If Trump had not pulled out of the JCPOA, Iran’s stockpile of HEU—its shortest path to the bomb—would not exist. Trump’s claim that if he had not pulled out, Iran would now have nuclear weapons is simply a lie. (If, however, the United States and its partners had wasted the time the deal offered and no further accords had been reached, the sunset clauses on the limits Iran had accepted would be starting to kick in.) Moreover, before this war, Iran offered to blend down its HEU so it could not be used in weapons, pause all enrichment for several years, and not stockpile enriched uranium—an offer Trump rejected, without exploring whether it could be improved, in favor of war.³

To be fair, there is no doubt that the military attacks during Trump’s second term have been a major setback for Iran’s nuclear program. Iran’s major non-reactor nuclear facilities were seriously damaged in the airstrikes Israel and the United States carried out in June 2025; those attacks also killed several prominent scientists and engineers associated with Iran’s nuclear program. None of its uranium enrichment or conversion facilities appear to have been operable after those strikes.⁴ The current war, which began February 28, appears to have done some modest additional damage to Iran’s nuclear efforts, including hitting the entrance to the Natanz enrichment facility with bunker-buster bombs, further ensuring it could not readily be returned to operation.

Nevertheless, although this war will have lasting and far-reaching effects on everything from global geopolitics to oil prices to the U.S. reputation in the world, there is little evidence it will prevent a future Iranian nuclear weapon. Iran’s nuclear program is far from “obliterated.” Indeed, by dramatically increasing Iran’s resolve to acquire a nuclear deterrent and leaving key material and some equipment and expertise in place, I believe that the war’s net effect will be to increase, rather than decrease, the probability that Iran will have nuclear weapons a decade from now.

Key issues in assessing this judgment include the fate of Iran’s HEU; its other stocks of enriched uranium; its centrifuges and components to make them; and its weaponization-related equipment and expertise. They also include Iran’s incentives to pursue nuclear weapons; the ability of intelligence agencies to detect a future move toward nuclear weapons in time to stop it; and the need, ultimately, to find a new path to agreed restraints and effective international monitoring.

The Dangers of Iran’s Surviving HEU

The most immediate concern is the roughly 441 kilograms of HEU enriched to 60 percent U-235 that Iran had accumulated before the summer 2025 strikes.⁵ This material was stored in the form of uranium hexafluoride (UF6), in canisters similar to scuba tanks. Most, perhaps all, of this material appears to have survived last summer’s strikes and this year’s war.⁶ This is the material that Iran could most readily use to produce nuclear weapons, and it might pose broader hazards as well.

First, Iran might choose to enrich this material to 90 percent or more U-235 to use it in the weapon design that Iran was working on in the secret nuclear weapons program it largely stopped in late 2003.⁷ With just 100 of the advanced centrifuges it has developed since Trump’s pullout from the nuclear deal in his first term, Iran could further enrich enough material for a bomb in just a few weeks. If Iran still has access to all of the 60 percent enriched HEU stock, it would be enough, once further enriched, for more than 10 nuclear weapons.

Second, even without further enrichment, the 60 percent enriched material, once converted to metal, could be used in a bomb. It could be put in a weapon of the same design, exploding with less power, or it could be used in a modified design using more material and explosives.⁸ Indeed, this material could be used to make several simple “gun-type” nuclear weapons comparable to the Hiroshima bomb, which could be made quickly, although Iran would likely have to rely on covert means of delivery as the bombs would be too heavy for its missiles.

Beyond Iran using this material itself, there are other concerns. Nobody knows who might take possession of it in the resulting chaos if Iran’s government collapses. Even without a government collapse, the current crisis might lead some lower-level people managing the HEU to decide they needed to get out of the country; they might see selling this material as their golden ticket, although transporting the material in its current form would be a challenge. If the material falls into the hands of terrorists, multiple studies by the U.S. and other governments have warned that a sophisticated extremist group with HEU might well be able to make a crude bomb. As with the Iranian government, however, they would first have to convert it to metal, which would be an additional challenge.⁹

Where is Iran’s HEU Now?

A critical question is what has happened to this dangerous material as a result of the U.S. and Israeli attacks. Logically, there are three possibilities: It was in canisters that were destroyed in the June 2025 bombings and is no longer available because it is scattered over the walls, floors, and rubble; it was in canisters at sites that were bombed but the canisters remained intact and could be recovered by digging down into the facilities; or, it is in canisters that were at facilities that were not destroyed and could be recovered.

Unfortunately, it appears that most of Iran’s HEU is in this third category, although possibly blocked by rubble at the entrances that would have to be cleared to access it. The IAEA has noted that before the 2025 strikes, much of the HEU, as well as Iran’s 20 percent enriched U-235, was stored in deep underground tunnels near the nuclear facility outside the city of Isfahan.¹⁰ In particular, IAEA Director-General Rafael Mariano Grossi has reported that more than 200 kilograms of it was stored there before the strikes. The IAEA has not been permitted to visit since then.¹¹

Some reports suggest that other portions of Iran’s HEU were stored in a deep underground facility near Natanz known as Pickaxe Mountain, and in the enrichment site at Fordow, one of the locations bombed in 2025.¹² It is at least possible that Iran removed the HEU from some or all of these facilities before those strikes, as vehicles were seen at these facilities in the days before military action began.

In late March 2026, a satellite photograph surfaced that was taken June 9, 2025, just days before last year’s strikes began, showing a flat-bed truck outside the entrance to the Isfahan tunnel complex carrying more than a dozen 1.1-meter diameter containers. The truck was accompanied by another truck with a mobile crane, perhaps for lifting the containers, and vehicles that might be security vehicles. No definitive judgments can be made from this single photograph: it does not reveal what was in the containers, or even whether the truck was going into the tunnel complex or backing out of it. One plausible explanation is that Iran moved even more HEU into the Isfahan tunnel complex as protection from the attacks it expected were coming. The containers could be HEU canisters with overpacks for safe transport.¹³

The tunnels at Isfahan are significantly deeper than the underground Fordow enrichment facility that was seriously damaged by U.S. “Massive Ordnance Penetrator” bombs in the 2025 strikes, and they have not been destroyed. Gen. Dan Caine, the chairman of the U.S. Joint Chiefs of Staff, has reportedly acknowledged that the Isfahan tunnels are too deep to destroy with current bunker-buster bombs.¹⁴ The tunnels at Pickaxe Mountain are also deep and the mountain is granite, making those tunnels at least as challenging a target.

Although the tunnel complex near Isfahan has not been destroyed, the entrance was bombed, raising the question of how difficult it would be to get at the containers in the tunnels. Trump has asserted that “it would take months” to dig down to get the “nuclear dust” in these tunnels; in one interview, he said the HEU was “so far underground, I don’t care about that.”¹⁵

Yet a wide range of public evidence suggests Trump is wrong about how hard this material would be to reach. First, in bilateral talks before the ongoing war, and in the Islamabad talks during the ceasefire, the Trump administration demanded that Iran export this material, and Iran, before the war, offered to blend it down—implying that both sides thought Iran could get access to it. Second, substantial press reporting indicates that planning for a possible operation to seize or destroy this HEU includes construction equipment to excavate the rubble, suggesting that this work could be done fairly quickly.16 Third, the IAEA has made clear that it sees this material as a major continuing concern; it has observed “regular vehicular activity” around the tunnel entrances since the bombings last summer, although Grossi has said that activity did not include major digging.17 Finally, there are some reports that the material is already accessible.¹⁸

What is Next for Iran’s HEU?

If airstrikes alone will not solve the problem of the HEU in these deep tunnels, there are a few other options. One is surveillance and deterrence. Trump seems to be leaning toward simply leaving the HEU where it is and watching to make sure Iran does not remove it from the tunnels. As he put it in his April 1 address, the United States has the HEU “under intense satellite surveillance.… If we see them make a move, even a move for it, [we] will hit them with missiles very hard again.”¹⁹ Leaving more than 10 bombs’ worth of HEU in Iran, however, while relying only on a return to war to prevent its use, is hardly an effective long-term solution.

Another route the Trump administration is trying to pursue is negotiating a deal with Tehran to remove the material. In the JCPOA, Iran agreed to export nearly all of its enriched uranium to Russia. In the talks before the ongoing war, and apparently again in the Islamabad talks, the Trump team demanded that Iran export the HEU to the United States; Iran rejected that idea but offered to blend it down to lower levels of enrichment in Iran. As of mid-April 2026, there were no signs that Iran was willing to accept U.S. demands for exporting this material.

A deal, if one could be reached, would be a far more practical and less dangerous approach than any military option. Over the years, HEU has been removed from many cooperative countries. One early example was Project Sapphire, in which U.S. teams worked with Kazakhstan in 1994 to fly 1,280 pounds (580 kilograms) of HEU to safe storage in Tennessee.²⁰ Programs such as the Global Threat Reduction Initiative and its successor, Material Management and Minimization, have removed tons of plutonium and HEU from scores of sites around the world. Indeed, more than half of all countries that once had HEU or plutonium separated from spent fuel on their soil have shipped it back to its country of origin, ensuring that terrorists will not be able to obtain weapons-usable nuclear material from the emptied locations.

Even a cooperative approach would not be easy. Whatever rubble may be blocking the entrances to the site or sites where the Iranian HEU now exists would have to be cleared and the HEU canisters would have to be assessed, measured, and carefully packaged. Then, they would have to be transported out of the tunnels, trucked to an available airstrip, and flown to a country willing to accept the HEU.

Without Iran’s cooperation, the remaining option for eliminating the HEU would be sending in a team of U.S. or Israeli soldiers and experts to deal with the material. U.S. special operations forces have long trained with federal scientists and experts to disable or secure adversaries’ nuclear weapons and material,²¹ but a military operation would not be easy. Removing the material would require taking all the steps of the cooperative approach, except it would be in the middle of a war zone. There are probably dozens of containers, collectively weighing tons, in multiple locations deep inside Iran—a country as big as France, Germany, and Spain combined. Troops would probably need to collect the material from several facilities (possibly including digging through rubble at the entrances), secure an airstrip near each, truck or helicopter the equipment and material to and from the strip, and defend against attacks on the preparations and shipments. Flying everything out of Iran in the face of remaining air defenses would be another hazardous operation.

U.S. and Israeli air power might be able to devastate Iranian forces near the HEU sites, clearing the way for the deployment of U.S. troops. Trump has said he would only authorize a ground operation if Iran was “so decimated that they wouldn’t be able to fight on the ground level.”²² However, there may well be extensive booby traps and other defenses in the tunnels (and possibly decoy canisters that would have to be distinguished from the real ones), and Iranian missiles and drones could threaten the troops on the surface. Mark Esper, a defense secretary in Trump’s first term, has warned that sending ground troops into Iran to address the HEU stockpile would take a large force and be “very perilous.”²³

A simpler but messier option than removal would be to blow up the containers, with explosives attached to each one. The UF6 would end up on the walls, floors, and rubble in the tunnels, making it difficult to ever recover and use. The tunnels would then be contaminated and unusable, and the U.S. team would need to be careful about its own safety.

Blending the material to low-enriched uranium (LEU) so it could not support an explosive nuclear chain reaction is another option. Doing that on-site would also be difficult, requiring the delivery of equipment and tons of uranium for blending into an active war zone. The National Nuclear Security Administration has developed mobile equipment for packaging HEU in the past, although it has never been used to downblend hundreds of kilograms of UF6, let alone doing that in a war zone.²⁴ The result of such a blend-down would be many tons of LEU. Enrichment is a process that speeds up as it goes, so that when uranium has been enriched to 5 percent, two-thirds of the work of enriching to 90 percent is already done. That is why the JCPOA limited Iran to only 300 kilograms of LEU at a time. Leaving many times that amount in Iran would pose a serious risk that it could be re-enriched in the future.

Could Washington know for sure that it had secured all the HEU, whether by the cooperative or non-cooperative options? Before the June 2025 strikes, the nuclear material was all monitored by the IAEA. The agency knows how many canisters there were, and the serial numbers of each one, but no international inspectors have seen this material since then. What if U.S. forces found 70 percent of those canisters? Would they be confident the rest were deeply buried under rubble at such sites as Natanz and Fordow, and not held in some secret stockpile?

Iran’s Other Uranium

HEU is not the only issue. Before the June 2025 strikes, Iran also had 181 kilograms of uranium enriched to 20 percent U-235, more than six metric tons of 5 percent enriched LEU, more than two metric tons of 2 percent enriched LEU, and likely a large amount of uranium in UF6 that was not yet enriched.²⁵

Material such as 5 percent LEU cannot be used directly in a nuclear bomb, but as just noted, it would provide a two-thirds head start on enriching all the way to 90 percent. If further enriched, the stock of 20 percent enriched material would be enough for one to two nuclear weapons, whereas the stock of 5 percent would be enough for roughly a dozen.

What has happened to this material? The IAEA reports that the 20 percent enriched uranium also was stored in the deep tunnels at Isfahan, and thus it, too, is probably available to Iran, with some effort. The 5 percent, 2 percent, and unenriched material was stored in huge 30B containers, weighing nearly three metric tons when full. Many of these containers are thought to have been at the above-ground nuclear facility at Isfahan, where the UF6 is produced from uranium oxide, but some may have been at the bombed enrichment plants at Fordow and Natanz, and may be in either intact or ruptured containers buried in the rubble.²⁶ Publicly at least, little is known about how much of this material may have survived; U.S. and Israeli intelligence agencies presumably know more, but this stock again raises the question of how confident the U.S. and Israeli governments could be that there were no secret stockpiles.

Iran’s Centrifuges and Centrifuge Parts

The 2025 U.S. and Israeli attacks severely damaged Iran’s main enrichment sites at Natanz and Fordow. It is unlikely that any of the centrifuges at either site are still operable. Other strikes hit known facilities for assembling centrifuges and making key parts.

Nevertheless, there are serious questions about what centrifuge capabilities still remain. As Grossi has noted, at the time of the JCPOA, Iran was primarily relying on “a very primitive type of centrifuge.” Since Trump’s pullout from the nuclear deal, there have been no restraints on Iran’s development of more advanced machines, and it “has the most sophisticated, fast, and efficient” centrifuges, meaning that the enrichment needed to produce bomb material could be done faster, by fewer centrifuges, in an easier-to-hide facility.²⁷

One innovation of the JCPOA was that it established a system of monitoring Iran’s production of centrifuges, but that ended after Trump pulled out of the accord; Iran has been producing centrifuges and centrifuge parts for years without any monitoring. It would be surprising if Iran had not dispersed centrifuges and the parts for them to a wide range of locations; Grossi has warned that there may be “dozens of workshops” for this purpose.²⁸ It is hard to know whether U.S. and Israeli intelligence have been able to find and target all those places.

Pickaxe Mountain is a particular concern. Iran has described that facility as a plant for assembling centrifuges, yet reportedly it was not yet operational when the 2025 attacks began. There may well be stocks of centrifuges and the parts to assemble more there, and like the tunnel complex at Isfahan, it is too deep to be reached by U.S. bunker-buster bombs. More worrisome, the complex there is too big for just assembling centrifuges—big enough to assemble them, set up cascades to enrich uranium, have facilities to convert the uranium to metal, and then turn the metal into nuclear weapons.

Weaponization Equipment and Expertise

From the late 1990s until fall 2003, Iran had a serious covert program to design and build nuclear weapons and carry out a nuclear test. The country’s supreme leader, Ayatollah Ali Khamenei, ordered a halt to the visible weaponization parts of that program in late 2003, while enrichment efforts continued apace. Part of the dispute between Iran and the IAEA has had to do with what happened to materials and equipment from that past program.²⁹ Since Trump pulled out of the JCPOA, there have been new or renewed activities in Iran that Grossi has described as “concerning.”³⁰ Meanwhile, U.S. intelligence agencies dropped from their annual threat assessment that Iran had not resumed its weaponization effort.

During the 2025 strikes, several locations that the Israeli Defense Forces (IDF) asserted were associated with potential weaponization activities were destroyed, most likely along with some equipment related to weaponization. The necessary equipment, however, is not especially unique or difficult to rebuild.

Iran’s deeply buried facilities, including the tunnels at Isfahan and the facility at Pickaxe Mountain, remain an ongoing concern. They are largely immune to air attack, and one effect of the 2025 strikes was to educate Iran about how deep they had to dig to protect their facilities. It is difficult to know what equipment and materials may be stored there—or what secret locations intelligence agencies have not yet managed to locate.

The loss of the people who have been killed by Israeli strikes might inhibit an Iranian effort to rebuild a weapons program even more than the loss of equipment and facilities. After several assassinations in previous years, including Mohsen Fakhrizadeh, leader of the past nuclear weapon program, the IDF asserted that it killed some 20 weapon scientists and engineers in the 2025 strikes and has announced additional deaths during the 2026 war.³¹

Never before has a state undertaken such a large-scale killing campaign against the people participating in a nuclear program. The loss of these figures is undoubtedly a setback for any effort Iran might undertake to pursue nuclear weapons in the future. But how large a setback? The overall Iranian nuclear program has involved thousands of people, and the functions most related to weaponization likely had scores to hundreds of participants. Clearly, some people are more important to an overall program than others, so it may be that the people who have been killed were very central to the effort. Without detailed intelligence information, the overall impact is difficult to judge. Yet it seems very likely that with time, other experts will be able and willing to take the place of those who have been killed, should Iran decide to pursue nuclear weapons.

Nuclear-Weapon Incentives

Such a decision appears to be a real possibility. After this war, Iran is likely to have stronger incentives to build nuclear weapons than ever before. It has been attacked twice in the past year by powers with overwhelmingly greater conventional military power. The tools it relied on before this to deter attack—the regional groups it armed and supported, its missile fleet, and its nuclear latency—have utterly failed and are far weaker than they once were. Hard-line members of parliament have already submitted legislation requiring Iran to withdraw from the nuclear Nonproliferation Treaty (NPT), although it is not clear when that could be voted on, as parliament has not been gathering during the war. The NPT “has had no benefit for us,” according to Ebrahim Rezaei, a spokesman for the Iranian parliament’s National Security Commission.³²

Khamenei, the supreme leader who ordered a halt to the weaponization program in late 2003 and wrote a fatwa prohibiting nuclear weapons, is dead. Mojtaba Khamenei, his son and successor, is reportedly even harder-line and closer to the Islamic Revolutionary Guard Corps (IRGC) than his father.³³ The opening attacks of the 2026 war reportedly killed his father, mother, wife, teenage son, sister, and brother-in-law and seriously injured him.³⁴ That hardly seems likely to incline him toward a softer line toward the countries that attacked. He reportedly opposed his father’s fatwa against nuclear weapons and could readily reverse it. Below Khamenei, it appears that hard-liners in the IRGC, many of whom have long believed that Iran should build nuclear weapons, are increasingly in charge.

In short, it appears that although the June 2025 attacks and the 2026 war have reduced Iran’s capability to develop and build nuclear weapons, these attacks have built incentives for Iran to do so that are greater than ever before and put people in charge who are likely to be inclined to move in that direction.

The major problem Iranian bomb advocates face is the risk that such an effort would again be attacked before it could be completed. The U.S. and Israeli attacks of June 2025 and the 2026 war have revealed just how deeply Iran’s nuclear program was penetrated by Israeli and Western intelligence agencies, who were able to find and target a range of secret facilities and nuclear program participants. Iran has arrested hundreds of people in its effort to eliminate the spies, but clearly, any effort to launch a renewed nuclear weapon program at secret facilities would run a serious risk of being discovered and attacked.

Relying on Intelligence Success

The impressive military successes of the June 2025 strikes and the current war have been based on even more remarkable intelligence successes, which uncovered what Iran was doing in its nuclear program, where it was taking place, and, in many cases, who was doing it, down to the level of which bedroom particular people were sleeping in.

At the same time, the attacks have meant an end to any international inspection of Iran’s key nuclear facilities. Although Iran permitted safeguards inspectors to return to the nuclear power reactor at Bushehr and the Tehran Research Reactor, no one from the IAEA has seen any of Iran’s enriched uranium, centrifuges, or other key equipment since before the June 2025 attacks. The United States, Israel, and other countries concerned about Iran’s nuclear program have had to rely entirely on the continuing success of intelligence agencies in finding every secret nuclear effort.

So far, that success has been impressive, dating back many years, from the discovery of the secret Fordow uranium enrichment plant before the first centrifuge was installed in the 2000s to Israel’s theft of the vast archive of nuclear weapons documents from Iran. But how long will that success last? Some human and electronic sources have probably been destroyed or compromised during the recent attacks, and Iran’s government may have succeeded in uncovering others in its intensified counter-spy efforts.

The fact is that a nuclear weapons program need not be large. An enrichment plant capable of producing enough HEU for a bomb every year might be smaller than a typical supermarket and use a comparable amount of power. The facility for converting that uranium to metal and machining the metal into bomb components might be smaller still. The facility for assembling the nuclear weapons and fitting them to delivery vehicles would not need to be large either. Will intelligence agencies always be able to find these in time?

Consider, for example, the facility that Iran was building for manufacturing nuclear weapons decades ago, when its nuclear weapons program was at full throttle. When the program halted in 2003, Iran had dug out tunnels under a mountain for this facility and purchased the necessary equipment, although it was not yet installed. Neither U.S. nor Israeli intelligence had identified the construction as a nuclear facility. (To be fair, with no nuclear-related equipment installed, that would have been hard to do.)³⁵

The Need for a Deal

At this writing, the Islamabad talks seem to have left a vast gap between American and Iranian proposals. There appears to be a substantial chance that Trump will end the war against Iran with the HEU still in Iran, no agreed limits on Iranian enrichment, no international monitoring in place, and a harder-line regime in charge in Iran. Trump himself has raised the possibility that Iran would renew its nuclear weapons program, forcing a future U.S. president to attack Iran again.³⁶ Any negotiated agreement that led to that result would be justly described as a terrible deal, although Trump is trying to pitch this result as a victory.

Ultimately, Iran’s nuclear knowledge cannot be bombed away. With time and motivation, destroyed facilities can be rebuilt. Diplomacy provides the only long-term approach that can potentially provide lasting, stringent limits, effective international monitoring, and reduce the sense of threat that is the best argument for Iran’s nuclear weapon advocates. Provisions to accomplish those objectives were central to the JCPOA. As Grossi put it:

“[O]nce the military effort comes to an end, we will still inherit a number of major issues that have been at the center of all of this… a lot still has survived. They have the capabilities, they have the knowledge, they have the industrial ability to [pursue nuclear weapons]. This is why we need to go back to a negotiating table.”³⁷

Trump’s team is continuing to seek a deal with Iran on its nuclear program, the Strait of Hormuz, and other matters. But the president seems to believe that making extreme threats and sending negotiators with no nuclear expertise, verification expertise, Iran expertise, or sanctions expertise will be sufficient to reach an effective deal. That stands in stark contrast to the deep expertise the Obama administration’s negotiation team brought to the table—in talks that stretched over years, not hours.

Insisting, as the Trump team reportedly is, on Iran agreeing to zero uranium enrichment forever is neither necessary nor wise. A deal for zero would not buy very much extra security, as Iran could move from zero quickly, having demonstrated in the past that it can install 1,000 centrifuges—enough to launch a bomb program—in a month.  Moreover, Iran’s government has consistently rejected an enrichment ban for decades, framing it as colonial powers seeking to take away Iran’s God-given right to technological development.  Similarly, insisting that Iran export its HEU to the United States—the country that has just been attacking it—rather than to its uneasy partner Russia, is almost certainly non-negotiable. It is precisely because Russia might be willing to aid Iran if the United States reneged on a deal that a Russian option would have some small chance of being negotiable.

With Trump having pulled out of the JCPOA and twice launched military strikes while talks on new deals were underway, Iran now has very little reason to believe any promises that the United States makes, and a strong incentive to preserve its nuclear options. Nor does Washington have much reason to believe Iranian promises, despite the fact that Iran, unlike the United States, complied with the 2015 nuclear deal. The 2025 strikes and the 2026 war, although setting back Iran’s nuclear capabilities, may well have ruined the chances for negotiating an effective nuclear deal for many years to come.

ENDNOTES

1. Donald Trump, “Address to the Nation on Military Operations in Iran,” The American Presidency Project, University of California, Santa Barbara, April 1, 2026.

2. Gary Samore, ed., “The Iran Nuclear Deal: A Definitive Guide,” Harvard Kennedy School Belfer Center for Science and International Affairs, November 2017.

3. Kelsey Davenport, “U.S. Negotiators Were Ill-Prepared for Serious Nuclear Negotiations with Iran,” Arms Control Association, March 15, 2026.

4. David Albright, Sarah Burkhard, Spencer Faragasso, et al., “Comprehensive Updated Assessment of Iranian Nuclear Sites Five Months After the 12Day War,” Institute for Science and International Security, November 21, 2025.

5. For the quantity, see International Atomic Energy Agency, “NPT Safeguards Agreement with the Islamic Republic of Iran,” GOV/2026/8, February 27, 2026, p. 7.

6. Julien E. Barnes, Tyler Pager, Christian Triebert, Eric Schmitt, and Ronen Bergman, “Iran Could Retrieve Uranium at Site U.S. Bombed Last Year, Officials Say,” The New York Times, March 7, 2026.

7. Aaron Arnold, Matthew Bunn, Caitlin Chase, Steven E. Miller, Rolf Mowatt-Larssen, and William H. Tobey, “The Iran Nuclear Archive: Impressions and Implications,” Harvard Kennedy School Belfer Center for Science and International Affairs, April 2019.

8. Matt Kaplan, “Implosion Nuclear Weapons with 60%-Enriched Uranium,” Science & Global Security, Vol. 33, No. 1-3 (2025), pp. 89-101.

9. For an unclassified example, see Chapter 5, “The Non-State Adversary,” in Nuclear Proliferation and Safeguards, U.S. Congress Office of Technology Assessment, Washington, D.C., OTA, 1977.

10. IAEA, GOV/2026/8, p. 4.

11. Forrest Crellin, “Much of Iran’s near-bomb-grade uranium likely to be in Isfahan, IAEA’s Grossi says,” Reuters, March 9, 2026.

12. Julian Borger, “US weighs sending forces into Iran to secure nuclear stockpile, reports say,” The Guardian, March 19, 2026.

13. See, for example, François Diaz-Maurin, “Analysis: Iran likely transferred highly enriched uranium to Isfahan before the June strikes,” Bulletin of the Atomic Scientists, March 29. 2026.

14. Natasha Bertrand and Zachary Cohen, “US did not use bunker-buster bombs on one of Iran’s nuclear sites, top general tells lawmakers, citing depth of the target,” CNN, June 28, 2025.

15. Trump April 1 address transcript; Steve Holland, “US to leave Iran ‘pretty quickly’ and return if needed, Trump tells Reuters,” Reuters, April 1, 2026.

16. Ellen Nakashima, John Hudson, Alex Horton, and Karen DeYoung, “Risky commando plan to seize Iran’s uranium came at Trump’s request,” Washington Post, April 1, 2026.

17. The vehicular activity is mentioned in IAEA, “NPT Safeguards Agreement with the Islamic Republic of Iran,” p. 4. Grossi mentions the lack of evidence of efforts to recover and remove the material in Crellin, “Much of Iran’s Near-Bomb-Grade Uranium Likely to Be at Isfahan.”

18. Barnes, et al., “Iran Could Retrieve Uranium.”

19. Trump April 1 address transcript.

20. Togzhan Kassenova, “Project Sapphire: How to Keep 600 Kilograms of Kazakh Highly Enriched Uranium Safe,” War on the Rocks, April 1, 2022.

21. See, for example, Walter T. Ham IV, “Highly specialized US Army teams train to disable any potential enemies’ nuclear capabilities,” U.S. Army, January 25, 2024; Joseph Trevithick, “Loose Nukes In Iran Is A Scenario U.S. Special Operators Have Been Training For,” The War Zone, June 18, 2025.

22. Borger, “U.S. weighs sending forces.”

23. Ellen Mitchell, “Esper: Sending special forces to find enriched uranium in Iran would be ‘very perilous’ mission,” The Hill, March 10, 2026.

24. “The Mobile Packaging Program: An origin story,” National Nuclear Security Administration, November 16, 2022.

25. IAEA, GOV/2026/8, p. 7.

26. Interviews.

27. “Transcript: International Atomic Energy Agency Director-General Rafael Grossi on ‘Face the Nation with Margaret Brennan,’” CBS News, March 22, 2026.

28. Ibid.

29. Arnold et al., “The Iran Nuclear Archive.” For a useful summary of the IAEA’s investigations of these issues, see Kelsey Davenport, “IAEA Investigations of Iran’s Nuclear Activities,” Arms Control Association, February 2025.

30. Grossi on Face the Nation.

31. “Significance of the Targeted Nuclear Scientists in the 12-Day War,” Institute for Science and International Security, July 15, 2025.

32. Maziar Motamedi, “As war rages, Iranian politicians push for exit from nuclear weapons treaty,” Al Jazeera, March 28, 2026.

33. See, for example, Karim Sadjapour, “The Iranian Regime Doubles Down,” The Atlantic, March 10, 2026; Patrick Clawson and Farzin Nadimi, “What Kind of Supreme Leader Would Mojtaba Khamenei Be?” Washington Institute for Near East Policy, March 5, 2026.

34. Details of the deaths in that initial strike have not been made public, and accounts vary. This one is based on an interview with Iran’s Ambassador to Cyprus. See Helena Smith, “Mojtaba Khamenei was hurt in strike that killed his father, Iran’s Cyprus ambassador confirms,” The Guardian, March 11, 2026.

35. Arnold et al., “The Iran Nuclear Archive.”

36. David Sanger, “Trump says he halted the nuclear threat from Iran. Evidence suggests otherwise.” The New York Times, March 31, 2026.

37. Grossi on Face the Nation.

May 2026

Syria faces daunting challenges as it seeks to recover from its 2011-2024 civil war. Rebuilding its battered economy, reconstructing its devastated infrastructure, standing up a competent government and repairing trust within its national polity are all on the list. None of the trials is as unique, or sordid, as coping with the chemical weapons legacy of former President Bashar al-Assad, who was overthrown in 2024 and fled to Russia. Experts consider his government’s extensive use of chemical weapons the most significant challenge to the international nonproliferation regime in the 21st century. Last year, Syrian Foreign Minister Assad al-Shibani called the Assad regime’s chemical weapons abuses “one of the darkest chapters” in world history and promised the new government in Damascus would work to dismantle what is left, adhere to international norms and ensure accountability for past wrongs. Syria’s ambassador to the Organisation for the Prohibition of Chemical Weapons (OPCW), Mohamad Katoub, is the man charged with working with the organization on these goals, which are complicated by the fact that Assad kept the program top secret. Katoub discussed the Syria-OPCW mission April 16 over Zoom with Arms Control Today publisher Daryl G. Kimball and Carol Giacomo, the journal’s chief editor. The interview has been edited for space and clarity.

Arms Control Today: You were appointed last year to head Syria’s mission to the OPCW in part because of your personal encounter with chemical weapons and the victims of chemical attacks during the Syrian civil war. Why is it important to complete the unfinished task of dealing with the Assad regime’s deadly chemical weapons arsenal?

Ambassador Mohamad Katoub: The approach of the Syrian government is based on three pillars. First is the safety of our people and the national security of Syria. Those remnants of chemical weapons are dangerous in case they’ll go to the wrong hands, and also, the Syrian people don’t have the required awareness or the notification mechanism to deal with such equipment. We have sincere concerns for any civilians who have to deal with those remnants without knowing what they are, especially because many of those [chemical weapons] sites are not known to us. The second pillar concerns Syria’s obligations as a state to international law. This is coming from our belief in international law and the Chemical Weapons Convention.

The third one is accountability and justice. Accountability and justice for Syria is peacebuilding. Without holding perpetrators accountable and bringing justice for the victims, the peace in Syria will keep being fragile. The Syrian people suffered from those chemical weapons, and part of justice is the destruction of the remnants of chemical weapons. Part of it is the guarantee of nonrecurrence of those crimes in the Syrian Arab Republic or in any other place. We have victims and their families who suffered from those chemical weapons, from the intimidation, from the propaganda, and from the denial. Many Syrians were subject to several layers of violations, not only to chemical weapons.

Look at Ghouta, as an example. The people of Ghouta were besieged for years, and they were subject[ed] to chemical weapons. Many of them were subject to forced disappearance, arrest, and were forced to leave their hometowns.… These multiple layers of violations mean that these victims need support, they need to see justice with their eyes. Part of accountability is to guarantee a nonrecurrence of the use of chemical weapons in our country, It’s much wider than the destruction of the chemical weapons. We saw ISIS developing and using chemical weapons in Syria without having a big infrastructure and advanced technology. Accountability is a measure of nonrecurrence.

ACT: How did chemical weapons (CW) use in Syria affect you personally?

Katoub: I used to work for medical aid organizations when I was in Syria, and also after I left Syria, I kept working in this sector for years. Part of the work was documenting chemical attacks. I witnessed victims of chemical attacks in my hometown, Douma; coming from the biggest attack in Ghouta; and also attacks before. People are aware mostly of attacks, which killed a big number of civilians, but this attack in Ghouta on August 21, 2013, was number 32 in Syria. There were at least 10 attacks in Ghouta before this one.

I named this the “unfortunate experience.” I had to learn to collect samples to document those attacks, to connect witnesses to the investigation bodies; I had to understand more about the conventions, the obligations of the state, how the OPCW works, how is the process of decision-making in the organization, and also how other mechanisms for accountability work in the world. I had to cooperate with different investigation bodies.

The Syria conflict is one of the most documented conflicts when it comes to CW violations. It started with the joint investigation team that was established by the UN secretary-general in 2013, and which was in Damascus the night of the Ghouta massacre, followed by investigations by the Commission of Inquiry, which was established by the UN Human Rights Council. The OPCW later established the fact-finding mission in Syria. The UN Security Council established the joint investigation mechanism, which was stopped by a veto in 2017; and then, the OPCW made a very brave and important step by establishing the investigation and identification team, [which was mandated for the first time to identify perpetrators of CW use in Syria].

I witnessed all this history—the development of investigations and working with witnesses. I remember in 2013, witnesses asking how we should document and collect samples and maintain a chain of custody and work with investigators. By 2017-2018, [they started being frustrated from being interviewed time after time without any consequences, without any results, and the question became instead of how, why we need to collect samples, why should we work on documenting those attacks. They started to lose hope. Now, it’s another opportunity. Those witnesses can speak freely now. The OPCW can have access to original documents now, can have access to witnesses where they can be interviewed in Damascus.

In February this year, we hosted an event here in the OPCW with the [Syrian] National Commission for Transitional Justice. One of the speakers was a survivor of chemical weapons; he lost his two daughters in the massacre in Ghouta. He was on the panel, speaking inside the OPCW for the first time and he said, “for me, this is a step toward justice, to speak here freely about my pain in the OPCW.” So yeah, those are the first steps. We still have a long journey.

ACT: In March at the UN, Syria announced the “Breath of Freedom” task force, a group of countries that is going to assist the Syrian government in getting rid of the remnants of the CW program established by the former government. Could you tell us about the goals of that initiative and what is known currently about the status of the CW arsenal.

Katoub: The Breath of Freedom task force is a technical group that’s working together for the purpose of elimination of the remnants of chemical weapons of the Assad era. The states [that] are members of the task force have been supporting the rights of the Syrian people in fighting chemical weapons and in demanding justice and accountability for the usage of chemical weapons. They believe in our goal.

We are facing different types of challenges when it comes to the destruction of chemical weapons. Some are related to the security situation because of the Israeli airstrikes or what’s happening now in the region or related to the remnants of conventional mines or any other remnants that might block access of inspection teams, and also to technical teams who want to do the [CW] destruction.

The second challenge is the secretive nature of the program. That means a lot of work [is needed] to collect and analyze information. We have very good progress in collecting information and sharing information with the [OPCW, which has] more technical capability than the Syrian national teams at this stage.

And the third one is the gap in the national capacity in Syria because this is not our [government’s CW] program, this is the Assad regime’s program. We knew nothing about it before the liberation except what was available publicly, but we know that Syria was subject[ed] to over 200 attacks because of those chemical weapons.

To overcome all those challenges, we need technical support, technology, and expert teams. If you look at the task force members, each one has a unique experience when it comes to dealing with chemical weapons. The elimination of the program doesn’t just involve destruction, but it is a wider scope of work that includes identifying the sites, facilitating the convoys of the inspectors, removing any remnants of conventional weapons, securing the site from any risk, [and] being prepared for medical evacuation to respond to any urgent situation.

Syria has approval from the [OPCW] Executive Council to use on-site expedited destruction in case the materials are not transferrable, or if the site doesn’t have sustainable access. So, there are certain criteria that we are working with the [OPCW] Technical Secretariat to develop for on-site expedited destruction.

The concept about establishing the task force is to create a mechanism for coordination between different states, either within the group or out of this group, [that] want to support Syria in those efforts under the verification of the Technical Secretariat. At the same time, we are trying to differentiate between the mandate of the Technical Secretariat and the obligations of the state. Certain types of support cannot be provided directly by the Technical Secretariat, given the need to maintain a clear distinction between its verification mandate and other forms of engagement. The secretariat cannot simultaneously undertake verification activities and deliver support in destruction. Accordingly, these functions should be led by the Syrian side, with the technical support of partner states.

ACT: On the Breath of Freedom task force, which countries currently are you working with, along with the OPCW? It sounds like you’re seeking additional assistance from additional states beyond the initial group.

Katoub: It’s Canada, France, Germany, Qatar, Turkey, the United Kingdom, and the United States. All their contributions and support are valuable to us. Outside the group, we started conversations with some [other] states. We hope that we will get an agreement soon with Jordan.

ACT: How many sites have you been able to survey so far, and when do you think your government might be in a position to make the necessary declarations to the OPCW about what remains of the program?

Katoub: Syrian national teams identified 35 sites so far. The Technical Secretariat has a longer list of 100 sites. We are working to verify and triangulate information, as well improve the readiness of our teams, whenever the Technical Secretariat decides to share the whole list of 100 sites with us, to work with them on those sites one by one.

We were able to facilitate 26 visits to suspected sites, so many of them are still not evaluated. We have about at least six sites, which were not declared [to the OPCW] by the Assad regime, and we are working to verify the information, so hopefully we will be able to make an interim update of the declaration of the Syrian Arab Republic to the OPCW on the CW program. I’m trying to be very careful on the language here, so it will not be a final update of declaration, but an interim one.

We don’t believe that we’ll be able to have a 100-percent accurate declaration. We might keep finding other sites while digging into the information and documents that we’re finding.… We look at the experiences of other states, [which] are finding chemical weapons after World War II now, after 70 or 75 years.

The plan is to update the declaration. We have an optimistic plan that we will be able to do so this summer. Because of the situation in the Middle East, the Technical Secretariat team had to evacuate since March 17, and this means that the verification will go slower. But this doesn’t mean we will not be able to work. Our national teams have the capability and whatever administration work we can do remotely or with the teams here from The Hague, we are doing this with them.

But there are circumstances out of our hands. We have an institutional capacity gap in Syria. We have three ministries that were established from zero and are involved in the destruction of chemical weapons: defense, interior and emergency and disaster management. If we find materials that we can destroy within the capacity that we have, we’ll go forward. If not, we will consolidate those until either we have the capacity, or if this capacity or technology will be very expensive or will take too much time, then we are open to deliver those remnants to another country like what happened in 2014, when other countries were involved in the destruction of the arsenal of the Assad regime.

ACT: Have you been able to do any destruction of chemical weapons so far?

Katoub: No, so far. We are still in the reconnaissance, evaluation, and assessment phase. We still didn’t start verification. This is why I’m saying sometimes circumstances are against us. We have now the information about several sites. Next step would be to do verification, to share that information with the Technical Secretariat, [and] verify it with them, but they should either visit the sites or find alternative methods for remote verification. That happened in certain circumstances, especially if the materials or the sites are very sensitive and they can’t access the site for any reason. There are methods and tools for remote verification, like sealed cameras or others. We are considering those; nothing final yet. So, this is an ongoing conversation between us and the Technical Secretariat.

ACT: Are the sites that you’ve identified being safeguarded?

Katoub: Most of them are guarded and under the control of the Ministry of Defense. Guaranteeing that unauthorized access is blocked in situations like Syria is challenging. Since the first day after the liberation [after Assad’s ouster], the Syrian army started to secure the sites that we had information about. Still, it’s not easy to maintain the blockade of unauthorized access. This is why this mission is urgent. This is why we need to accelerate destruction and have all the resources required.

In the first year, we worked a lot on structuring the teams and pooling resources to support them. Now, we are in a phase where we started to have a clearer structure and we have resources more available to us. I can’t say we have 100-percent accurate timeline, but we have a plan with milestones, and we’ve started to progress. Until the end of the year, we should be able to verify all the information about the list of 100 sites with the Technical Secretariat. By then, we’ll be able to do the destruction within the resources and the capacities that 
we have.

This will ease the burden of safeguarding those sites, especially those that are abandoned and not under the use of the Syrian army. This is where the challenge comes from, because we need to provide surveillance for those sites. Whenever we know that there are materials of high risk or high priority, those are definitely guarded. The issue is when we have materials that should be declared and [are] in abundant sites. We need to guarantee the safety of those sites from any unauthorized access.… But whenever we have high-risk materials, yes of course, we have those safeguarded. We were able to consolidate some materials to a temporary consolidation facility, and we have one site with high-risk materials, and this is under surveillance, or under guard, from the Syrian army.

ACT: As you describe it, it’s a very complicated undertaking. Under the best circumstances, how long do you think it would take Syria and the international community to complete the dismantling and destruction of these weapons? How much will it cost?

Katoub: They are two important and technical questions. We hope that at the end of 2026 we will be able to verify the information and do evaluations for the whole list that the Technical Secretariat has. We are trying to accelerate sharing information with us about those sites, so we can verify the information with the Technical Secretariat and with the help of other states. After the evaluation, we’ll be more able to put a timeline.

This verification will help us in another thing, which is to determine what kind of technology we need for the destruction.… Destruction can happen anytime, whenever we have those materials secured, in a consolidation facility. Destruction is not the biggest problem. This can happen in Syria, or it can happen in another country.

On estimate of cost, we have a list of needs, and to be frank, unfortunately, in Syria we don’t have the experts to do evaluation or a budget estimate for those very technical needs. We know what technology we need to do all the steps before destruction—as I mentioned, destruction is based on what we find—but the list of needs includes materials and training for enabling operation, convoy command, security for the convoys, logistics, medical preparedness, site evaluation and security from mines and other aspects, chemical weapons management, protection gear, on-site destruction, and the destruction.…We still don’t have a budget for it, but.… we’ve started to receive offers from state members of the task force to fill those gaps. Some of those are in-kind, some are training, and others will need financial contributions, maybe not within the group itself, maybe coming from other members of the OPCW. But unfortunately, we don’t have a budget.

But I can mention that a number that everyone knew, the annual budget for the technical secretariat work in Syria, the estimation is about 12 million euros. That’s only the Technical Secretariat; I’m not talking about the national teams. Definitely, it will be a big budget to do all of this. We have, of course, an ultimate goal to build national capacity, not only the destruction, because after two, three, four years, Syria should be a stable country, hopefully.… and then, we should take care of the legacy of those chemical weapons. So, building the national capacity itself is a goal for us.

ACT: Could you turn to your government’s plans and strategy for holding those who you find accountable for these past crimes? I suppose that the Ministry of Justice is in the lead of that?

Katoub: In the national working group that’s working on the chemical dossier, beside the ministries of defense, emergency and disaster management, health, and foreign affairs, we have the Ministry of Justice, Ministry of Interior, and the National Commission for Transitional Justice. So, the Ministry of Interior is leading the investigation in that regard. They have some good progress on that.

We were able, in an unprecedented step last October, to have the Executive Council of the OPCW decide to give access to the Syrian national accountability mechanisms to the information that was collected by the organization. This is a big step for us. We have the political decision now by the council, but we still don’t have the mechanism because this is complicated. Also, the International, Impartial, and Independent Mechanism (IIIM) established by the UN General Assembly doesn’t share information with us. We’re working with them to explore ways to do that. I’m happy we have good cooperation with them, and I think we should be able to achieve more progress.

On parallel, there is a lot of work to be done to reform the legal system in Syria. This is a big challenge, because the same system was used during the Assad era to give impunity to perpetrators for years. The same system never adapted to the obligation of the state to international law and to the conventions that Syria is part of, like the chemical weapons convention. Even the terms in this system don’t include things like war crimes or crime against humanity or genocide. The terms of still need to be adjusted.

We also have the Commission for Transitional Justice working a lot on structuring their team and also building mechanisms [to conduct] investigations. We should have some good news by the end of the year, hopefully, on the progress on accountability. I mean not that big, because, unfortunately, most high-ranking officials involved in this program fled the country. The hope is to establish the tribunal within the transitional justice system in Syria to hold those perpetrators accountable and to have those international mechanisms cooperating with it—either international organizations or also other states who can help us in holding perpetrators accountable. There are mechanisms for that, and we are very open to cooperate. We are very open to cooperating with states who use the universal jurisdiction mandate in their countries to hold perpetrators accountable, or to open the door for victims to file cases against those who used chemical weapons. We started conversations with some of them.

There are several options on the table on accountability. There was a group that discussed [potential accountability] in Syria before the liberation, before the fall of the Assad regime. Twelve states discussed several potential models of accountability for the use of chemical weapons, not only in Syria, but worldwide. From the outcome of this discussion, we take that we want the accountability mechanism to be national. Those perpetrators should be held accountable in Syria, in front of the Syrian people but of course, in cooperation with international mechanisms. This is the hope, of course. This is a long journey, again.

The expectations from the Syrian people are too high when it comes to justice. The first week of April marked the anniversaries of two chemical attacks in Syria—the one in Khan Shaykhun on April 4 and the Douma attack on April 7. On both anniversaries, the people of the two towns commemorated the victims and hosted an event, and some government representatives were there. The questions [from the people] were not about reparations. The questions were not about what those towns deserve. The questions were about when to hold those perpetrators accountable. So, they expect their government to be able to do that, and this is why we should work on it. We expect support from other states. This is another task that we here in the mission are working on. We already established the mechanism to support chemical weapon destruction, and we are working on ways to either establish or to join alliances and partnership to step forward on accountability, and to support the efforts of investigation and accountability in Syria.

ACT: One other challenge that you mentioned, of course, are the victims. How many victims have there been of these chemical attacks? What are the government’s plans to assist them, whether medical assistance or some other kind. Is this something that other countries have shown an interest in helping with?

Katoub: This is very important, actually. The number that we have from documentation is over 1,500. This is for the victims who were killed by chemical weapons. But those who were affected, the number is at least 10 times that, so at least 15,000. If you look at each one of those attacks, especially those which targeted urban areas, the victims were subject to several layers of violations: forced displacement, sometimes forced disappearance, besiegement, or others. If there were to be reparation measures, this should be part of a wider program on reparation for all the victims, not only for chemical weapons. Favoring victims based on the kind of violation that they were subject[ed] to causes a lot of damage to the community and to the society. There should be a program for the reparation for all victims in Syria.

There are ways to measure the support provided to each victim based on their needs more than based on how much they were damaged by this attack or whatever kind of violation they were subject[ed] to. Syria is trying to avoid favoring victims based on the type of violation. The second thing is, Syria does not have many resources for reparation programs. There should be a fund established to support Syrian victims, and then to have contributions from states who are willing to support this because this is peacebuilding for the country. Reparation doesn’t mean always financial support. It means acknowledgement of the pain of those people, even if it’s some minor measures that the government applies, sometimes with the help of civil society, international organizations, international community. This is very important for the peace in Syria.

The Commission for Transitional Justice is working hard on reparation measures and evaluating what measures they should provide to victims of different types. At the [Syrian] mission here in The Hague, we are encouraging academic institutions and researchers to study the long-term impact of usage of chemical weapons on communities and on individuals. We are very happy to connect anyone who wants to research this with research centers and universities in Syria because they will need labs and hospitals and clinical capacity. I encourage states to fund such programs. This is very important, not only for Syria. This is the widest use of chemical weapons after the convention was adopted in 1993. This was a wound in our hearts, and now it’s a lesson to learn from.

ACT: The Chemical Weapons Convention has near-universal support, yet some states in the Middle East have not yet joined. What message do you think your government’s commitment to completing the elimination of chemical weapons in Syria should be sending to your neighbors and others who have not yet joined this important agreement?

Katoub: Syria hopes to see all—not only neighboring countries—join this convention because Syria suffered and we know what chemical weapons use means. My memory about chemical weapon incidents that I witnessed is the fear in the eyes of the people. Terrible. Where I lived in Ghouta, during the hardest days of the siege, and I witnessed attacks, airstrikes, all kinds of weapons, barrel bombs, everything. But chemical weapons are different: You see the fear in the faces is different. because the community has the sense that nothing will protect them. Not a fortified shelter, nothing. So, chemical weapons should not be used under any circumstances in any place. If Syria sends a message to everyone, it’s that we’ve suffered enough from them. The world suffered enough from chemical weapons, and they should not be used again.

May 2026
By Frank N. von Hippel

Anyone concerned about today’s crumbling world order might be encouraged by a time four decades ago when an insane nuclear weapons buildup ended suddenly and was followed by a period of relaxed nuclear tensions and deep reductions in Russia and U.S. nuclear forces.

On January 15, 1986, less than a year after he had taken office and was confronted by a U.S. program to build up its nuclear warfighting capabilities with a new generation of more accurate long-range missiles,¹ Soviet leader Mikhail Gorbachev issued a proposal for “nuclear disarmament by the year 2000.” The Soviet Embassy in Washington published the proposal as a full-page ad in The New York Times.²

U.S. Secretary of State George Schultz sent it to President Ronald Reagan with a note: “While this proposal contains many serious problems, it will be universally considered to be a major step and will raise hopes that your vision of the elimination of nuclear weapons may be realizable.”³

Gorbachev’s target year of 2000 was 14 years away, and it turned out that he had only six years before the Soviet Union disintegrated and he lost power. During the time he had, however, Gorbachev helped set the Soviet Union and United States on a course to deep nuclear cuts from a combined peak of about 60,000 warheads in 1986 to a plateau of about 8,000 warheads in 2018, when the last U.S.-Russia nuclear reductions treaty, the New Strategic Arms Reduction Treaty (New START), limits were met. (The other seven nuclear-weapon states today have a combined estimated stockpile of about 1,400 warheads.)⁴

At a moment of acute nuclear angst, Gorbachev’s proposal paved the way for the October 1986 summit with Reagan in Reykjavik, Iceland, which produced no concrete outcome but energized bilateral nuclear arms negotiations and prepared the ground for the 1987 Intermediate-Range Nuclear Forces (INF) Treaty, the first to eliminate an entire class of weapons.

Gorbachev’s nuclear disarmament proposal in the Times was organized in stages, starting with bilateral reductions by the Soviet Union and United States. It said that “the USSR and the USA will reduce by one half the nuclear weapons that can reach the other’s territory” followed by the “complete elimination of medium-range missiles of the USSR and the USA in the European zone, both ballistic and cruise missiles.” In addition, he proposed that the “USSR and the USA should from the very beginning agree to stop all nuclear explosions and call upon other states to join in” and, by 1995 “All nuclear powers will eliminate their tactical nuclear weapons, i.e., weapons having a range (or radius of action) of up to 1,000 kilometers.”

By the end of Gorbachev’s tenure, many of these actions were either accomplished or well underway.

Reducing Soviet and U.S. Strategic Arsenals

The U.S.-Soviet Strategic Arms Reduction Treaty (START) talks began in Geneva in May 1982 after Reagan, in a speech at his alma mater, Eureka College, called for “equal ceilings, at least a third below the current levels.”⁵ The negotiations stalled, however, after Reagan announced his Strategic Defense Initiative in March 1983.

Gorbachev’s January 1986 definition of strategic weapons extended beyond the traditional intercontinental bombers and missiles and submarine-launched ballistic missiles to include nuclear-armed, sea-launched cruise missiles (SLCMs), which also could “reach the other’s territory.” The Reagan and George H.W. Bush administrations refused to include SLCMs in START, however, agreeing only to an unverified commitment not to deploy more than 880 nuclear-armed SLCMs.⁶

Gorbachev initially conditioned any Soviet agreement to deep cuts on the United States committing to keep any orbiting space lasers and ballistic-missile interceptors, developed as part of Reagan’s Strategic Defense Initiative (SDI), from being tested in space for at least 10 years. Reagan refused, resulting in the failure of their Reykjavik summit. Gorbachev’s technical advisers, Evgeny Velikhov and Roald Sagdeev, argued that the U.S. systems being proposed could be neutralized with much less costly “asymmetric” countermeasures.⁷ After Reykjavik, their arguments were reinforced by Andrei Sakharov.

Sakharov spoke out at an international conference on nuclear disarmament in Moscow in February 1987, a year after Gorbachev made his nuclear disarmament proposal. At the time, Sakharov was freshly released from seven years of isolation in Gorky (now Nizhny Novgorod) after his public opposition to the Soviet invasion of Afghanistan in 1979. He dismissed the U.S. proposal for defending against ballistic missiles as “a Maginot Line in space,”⁸ a reference to the impressive French line of fortifications along the German border that Germany’s army simply circumvented at the beginning of World War II.

Jeremy Stone—the president of the Federation of American Scientists—and I met with Sakharov in his apartment the night before his presentation. Stone had traveled several times to Moscow in the late 1960s to promote what became the 1972 treaty limiting antiballistic missiles. At our meeting with Sakharov, Stone made the prescient argument that the SDI program would collapse of its own weight. After the Soviet Union disintegrated in late 1991, Matthew Evangelista provided me with a copy of the transcript of our discussion with Sakharov, which had been recorded by the KGB and translated for Gorbachev.⁹

START, signed in July 1991 by Gorbachev and Bush, imposed a limit of 6,000 strategic warheads each on Russia and United States by 2001, down from about 12,000 warheads held by the United States and 10,000 warheads by the Soviet Union in 1985.¹⁰ Two follow-on reduction treaties cut the number of deployed strategic warheads to about 1,700 each under New START. (The treaty’s warhead counting rules, which attribute only one warhead to each nuclear bomber, considered 1,550 the official warhead limit).¹¹ New START expired in February 2026.

To date, all the treaties reducing Soviet/Russian and U.S. strategic nuclear warheads have focused on deployed warheads. The United States also has kept a reserve of nondeployed warheads—the “hedge”—and extra carrying capacity on its strategic missiles. As a result, Washington could today double its deployed warheads relatively quickly.¹²

In 2023, the Commission on the Strategic Posture of the United States created by Congress recommended that the “U.S. strategic nuclear force posture should be modified … [to address] the larger number of targets due to the growing Chinese nuclear threat” and specifically, to “Prepare to upload some or all of the nation’s hedge warheads.”¹³

George H.W. Bush let Reagan’s Strategic Defense Initiative fade away but the Republicans in Congress kept ballistic missile defense alive and President Donald Trump has revived Reagan’s fantasy, renaming it the “Golden Dome for America.”¹⁴

Eliminating Certain Soviet and U.S. Missiles

In a speech at the National Press Club in November 1981, Reagan declared, “The United States is prepared to cancel its deployment of Pershing II [ballistic missiles] and ground-launched cruise missiles [to Western Europe] if the Soviets will dismantle their SS-20, SS-4, and SS-5 missiles” targeted on Western Europe.”¹⁵ During negotiations on this proposal, which led to the 1987 INF Treaty, the weapons to be eliminated expanded to include both nations’ land-based, intermediate-range missiles worldwide and to include additional shorter-range missiles down to 500-kilometer range.

A total of about 2,700 nuclear-armed INF missiles were destroyed.¹⁶ The shells of a dismantled SS-20 missile and a dismantled Pershing II were used to make the body of a dragon being killed by St. George in a sculpture presented by the Soviet Union to the UN in 1990 and displayed outside the UN headquarters in New York City.

In 2019, based on Russia’s testing of a cruise missile in the INF range, however, the first Trump administration, eager to match China’s buildup of land-based missiles targeted on U.S. bases and allies in the Far East, took the United States out of the INF Treaty and Russia immediately followed.¹⁷

Ending Nuclear Testing

Gorbachev declared a unilateral Soviet moratorium on nuclear testing August 6, 1985, that lasted through 1986. In July 1986, his adviser, Velikhov, dramatically demonstrated Soviet openness to in-country verification by arranging for a U.S. environmental group, the Natural Resources Defense Council, to set up seismic stations around the test site in Kazakhstan.¹⁸ In October 1991, two years after massive public protests following a bad vent of radioactivity from an underground nuclear testing ended in Kazakhstan,¹⁹ Gorbachev reinstated his unilateral test moratorium. After the disintegration of the Soviet Union at the end of that year, Russian President Boris Yeltsin took over responsibility for Soviet nuclear weapons and continued the moratorium.

In September 1992, in response to the Soviet/Russian moratorium, the U.S. Congress decided to end U.S. nuclear testing. It passed legislation allowing up to 15 tests for safety or reliability reasons between October 1993 and September 1996, but the Clinton administration decided that those tests were unnecessary. The Comprehensive Test Ban Treaty was negotiated in Geneva and opened for signature in 1996.²⁰ The treaty has not come into legal force because the United States and six other nuclear-weapon states plus Egypt and Iran have not ratified it.²¹ Only North Korea has conducted confirmed tests since 1998, but the first Trump administration accused Russia, and the second Trump administration accused China, of conducting clandestine low-yield tests.

Eliminating Tactical Nuclear Weapons

Almost all U.S., and a large fraction of Soviet, tactical nuclear weapons were withdrawn from deployment in 1991-1992, as the Soviet Union disintegrated. Concerned about the security of Soviet nuclear weapons, Bush initiated these reductions, which removed over 3,000 U.S. nuclear artillery shells,  short-range nuclear missiles, and nuclear bombs from NATO allies while leaving hundreds of nuclear bombs for delivery by U.S. and allied fighter-bombers.²² Gorbachev reciprocated by denuclearizing Soviet ground forces but kept about 3,500 tactical warheads—since reduced to about 1,500—for Russia’s Navy, Air Force, and Air and Missile Defense Forces.²³

With the advent of precision-guided conventional munitions capable of destroying tanks as effectively as nuclear weapons, the United States has reduced the number of U.S. nuclear bombs for U.S. and allied (Belgian, Dutch, German, and Italian) fighter-bombers in Europe further, from about 500 in 2005²⁴ to about 100 today.²⁵ Otherwise, there are no U.S. tactical nuclear weapons deployed elsewhere outside the United States. In 2018, however, the first Trump administration launched the development of a new nuclear-armed cruise missile to be carried by U.S. attack submarines.²⁶

Gorbachev deserves a substantial amount of the credit for the deep cuts of U.S. and Soviet/Russian nuclear weapons since 1986. So do the Western European publics that rose up against the introduction of U.S. nuclear-armed, ground-launched cruise missiles and Pershing II intermediate-range ballistic missiles, as well as the U.S. popular movement to “freeze” the nuclear arms race that rallied against the Reagan administration’s initial proposal to develop more accurate offensive missiles for nuclear warfighting.²⁷

Reagan, too, deserves credit for responding to that public uprising, including by declaring in a radio address in April 1982 that “a nuclear war cannot be won and must never be fought,”²⁸ and becoming serious about reducing nuclear arms despite his hope that defenses could make “nuclear weapons impotent and obsolete.”²⁹

The lesson from this successful downsizing of the U.S. and Soviet/Russian nuclear arsenals is that nuclear arms control cannot be left to the professional militaries. If war comes, their duty is to destroy the enemy with the tools they have been given—and we have given them thousands of nuclear explosives.  Millions of people could perish as a result.

Militaries try to put such consequences out of mind as “collateral damage,” which can be grossly underestimated. It is the duty of the public and political leaders to draw the line on what weapons are acceptable. In the 1980s, significant segments of the U.S. and European publics declared that nuclear weapons are unacceptable. Gorbachev and Reagan heard that and responded. But they left the job incomplete.

Our civilization could still be destroyed by the 10,000 warheads that remain in military service today, when the current U.S. and Russian leaders once again see war as legitimate. There are many issues in the current political chaos that need to be dealt with, but to continue leaving the nuclear issue to the professionals is to continue “gambling with Armageddon.”³⁰

ENDNOTES

1. Harold Feiveson and Frank von Hippel, “The Freeze and the Counterforce Race,” Physics Today, January 1983, pp. 36, 38, 40, 42, 44, 46-49.

2. Mikhail Gorbachev, “Nuclear Disarmament by the Year 2000,” The New York Times, February 5, 1986, p. A13.

3. Secretary of State George Schultz, “Memorandum to President Reagan” from Washington on January 15, 1986, Foreign Relations of the United States, 1981-1988, Volume V, Soviet Union, March 1985-October 1986, U.S. Government Publishing Office, 2020, p. 772.

4. Bulletin of the Atomic Scientists, “Nuclear Notebook: Nuclear Arsenals of the World,” accessed April 21, 2026.

5. Ronald Reagan, “Address at Commencement Exercises at Eureka College, Eureka, Illinois,” Ronald Reagan Presidential Library & Museum, May 9, 1982.

6. U.S. State Department, “Declaration of the United States of America Regarding its Policy Concerning Nuclear Sea-Launched Cruise Missiles,” START Treaty Associated Documents, July 31, 1991, pp. 59-61.

7. Pavel Podvig, “Did Star Wars Help End the Cold War? Soviet Response to the SDI Program,” Science and Global Security, Vol. 25, No. 1 (2017), pp. 3-27,

8. Andrei Sakharov, Moscow and Beyond: 1986 to 1989, Alfred A. Knopf, 1991, p. 22.

9. Frank N. von Hippel, “Sakharov, Gorbachev, and nuclear reductions,” Physics Today, April 2017, pp. 48-54.

10. Robert S. Norris and Thomas B. Cochran, US-USSR/Russian Strategic Offensive Nuclear Forces, 1945-1996, Natural Resources Defense Council, January 1991, p. 12.

11. Hans M. Kristensen, Matt Korda, Eliana Johns & Mackenzie Knight-Boyle, “United States nuclear weapons, 2026,” Bulletin of the Atomic Scientists, Vol. 82, No. 2 (2026), pp. 119-150; “Russian nuclear weapons, 2025,” Bulletin of the Atomic Scientists, Vol. 81, No. 3 (2025), pp. 208-237.

12. Matt Korda, “If Arms Control Collapses, US And Russian Strategic Nuclear Arsenals Could Double In Size,” Federation of American Scientists, February 7, 2023.

13. Congressional Commission on the Strategic Posture of the United States, “America’s Strategic Posture,” Institute for Defense Analyses, 2023, pp. 47, 48.

14. Erica L. Green, “Trump Unveils Plans for ‘Golden Dome’ Missile Defense System,” The New York Times, May 20, 2025.

15. Ronald Reagan, “Remarks to Members of the National Press Club on Arms Reduction and Nuclear Weapons,” The American Presidency Project, University of California, Santa Barbara, November 18, 1981.

16. Daryl G. Kimball, “The Intermediate-Range Nuclear Forces (INF) Treaty at a Glance,” Arms Control Association, August 2019.

17. “In Tit-For-Tat Move, Putin Announces Russian Suspension Of INF Treaty,” Radio Free Europe/Radio Liberty, February 2, 2019.

18. William J. Broad, “Westerners Reach Soviet to Check Atom Site,” The New York Times, Sec. 1, p. 1,  July 6, 1986.

19. Togzhan Kassenova, “How Kazakhstan Fought Back Against Soviet Nuclear Tests,” Carnegie Endowment for International Peace, February 14, 2022.

20. Frank N. von Hippel, “The Decision to End U.S. Nuclear Testing, Arms Control Today, December 2019, pp. 14-19.

21. Comprehensive Test Ban Treaty Organization, “Status of Signature and Ratification, Annex 2 States,” accessed April 21, 2026. Ratification by the 44 Annex 2 states is required for the CTBT to come into force.

22. Hans M. Kristensen, “US Nuclear Weapons in Europe: A Review of Post-Cold War Policy, Force Levels, and War Planning,” Natural Resources Defense Council, February 2005,
p. 32.

23. Jaya Tiwari, “U.S. and Russian Tactical Nuclear Weapons: A Forgotten Threat,” (Physicians for Social Responsibility, Center for Global Health & Security and Health, 2001; “Russian nuclear weapons, 2025.”

24. Hans M. Kristensen, “U.S. Nuclear Weapons in Europe,” p. 9.

25. “United States nuclear weapons, 2026.”

26. Anya L. Fink, “Nuclear-Armed Sea-Launched Cruise Missile (SLCMN),” Congressional Research Service, January 8, 2026.

27. Lawrence S. Wittner, Toward Nuclear Abolition: A History of the World Nuclear Disarmament Movement: 1971-Present, Stanford University Press, August 2003.

28. Ronald Reagan, “Radio Address to the Nation on Nuclear Weapons,” Ronald Reagan Presidential Library & Museum, April 17, 1982.

29. Ronald Reagan, “Address to the Nation on Defense and National Security,” Ronald Reagan Presidential Library & Museum, March 23, 1983.

30. Martin J. Sherwin, Gambling with Armageddon: Nuclear Roulette from Hiroshima to the Cuban Missile Crisis (New York, Alfred A. Knopf, 2020).

May 2026
By Kelsey Davenport

U.S. President Donald Trump said he would not use nuclear weapons against Iran weeks after his threat to destroy Iranian civilization prompted concerns that he was considering nuclear strikes.

Trump’s comment ruling out nuclear weapons came two days after he extended a two-week cease-fire to give Iran more time to respond to a proposal to end the war that U.S. negotiators presented to an Iranian team during Pakistani-mediated talks April 11-12.

Iranian Foreign Minister Abbas Araghchi posted on the social media site X that he shared a “workable framework to permanently end the war on Iran” after an April 24-25 trip to Islamabad. It appears, however, that the two sides remain far apart on key issues, including nuclear restrictions, as Trump announced on his social media site, Truth Social, April 25 that his negotiators would not return to the Pakistani capital.

When asked by a reporter April 23 if he would use nuclear weapons, Trump said the United States “in a very conventional way decimated [Iran] without it.… A nuclear weapon should never be allowed to be used by anybody.”

In an April 7 statement on Truth Social, Trump triggered broad international criticism, including from Pope Leo XIV, when he threatened that a “whole civilization will die tonight” if Iran refused to reach a deal and open the Strait of Hormuz.

Later that same day, however, Trump backed away from the threat, saying on Truth Social that “Almost all of the various points of past contention have been agreed to between the United States and Iran.” He announced a two-week cease-fire to “allow the Agreement to be finalized and consummated.”

Trump’s optimism, however, proved premature. A U.S. negotiating team led by Vice President JD Vance traveled to Islamabad April 11 for talks with the Iranian team, which included Araghchi and Mohammad Bagher Qalibaf, a member of Iran’s parliament and lead negotiator, but the meetings ended without an agreement. Statements from officials on both sides suggested that key sticking points remained unresolved after negotiating sessions that spanned two days.

Before leaving Islamabad April 12, Vance told reporters that the United States needs to see “an affirmative commitment that [Iran] will not seek a nuclear weapon and they will not seek the tools that would enable them to quickly achieve a nuclear weapon.”

He said that the U.S. team was “quite flexible” but also noted that the negotiating team put “our final and best offer” on the table before leaving Islamabad and that “we’ll see if the Iranians accept it.”

He said the failure to reach a deal was “bad news for Iran much more than it’s bad news for the United States of America.”

Araghchi said the United States was to blame for the failure to reach an agreement. In an April 12 post on X, he said that the parties were “just inches away” from a memorandum of understanding when the Iranian team “encountered maximalism, shifting goalposts, and blockade.” He suggested that the United States has not learned from previous negotiations: “Good will begets good will. Enmity begets enmity.”

Two key sticking points in the talks appear to be the duration of a moratorium on Iranian nuclear activities and the future of Iran’s 440 kilograms of uranium enriched to 60 percent, a level just shy of the 90 percent U-235 considered weapons grade.

Trump said April 16 that Iran agreed to hand over its “nuclear dust,” presumably referring to the enriched uranium gas, to the United States. Iranian Foreign Ministry spokesman Esmail Baghaei denied Trump’s claim, saying on state-run TV April 17 that “Iran’s enriched uranium is not going to be transferred anywhere under any circumstances.” Araghchi had previously stated that the stockpile must remain in Iran, but said Iran would blend down the highly enriched uranium gas to lower levels.

On several occasions, Trump has threatened to deploy U.S. forces to remove the enriched uranium from Iran, an operation that would likely require thousands of troops on the ground, take several weeks, and require the transport of special equipment to secure the material. It is also unclear if the United States could locate all of the enriched uranium, some of which may be difficult to access given U.S. strikes on key nuclear facilities in June. (See ACT, July/August 2025.)

The two sides also disagreed over the duration of a suspension on certain nuclear activities, particularly enrichment. According to unnamed U.S. officials quoted in The New York Times April 13, Vance’s team asked for a 20-year “suspension,” but when asked about that report, Trump told the New York Post April 14 that he does not like the 20-year time limit. He suggested that he still wants a permanent ban, which was the U.S. position before the Feb. 28 strikes. Trump said he does not want Iran “to feel like they have a win.”

Iran responded to the 20-year suspension by offering a 5-year suspension, according to U.S. and Iranian officials cited in The New York Times April 13.

The issue of monitoring Iran’s nuclear program did not appear to be a key area of focus in the Islamabad talks. International Atomic Energy Agency Director-General Rafael Mariano Grossi has repeatedly emphasized the importance of resuming IAEA inspections. He told reporters April 15 that “Iran has a very ambitious, wide nuclear program so all of that will require the presence of IAEA inspectors.” Without inspections, “you will have an illusion of an agreement,” he said.

Originally, comments from Trump and Pakistani officials suggested that a second round of negotiations would be held April 21. Vance was planning to attend that round of negotiations, but Iran said it would not participate while the United States was violating the terms of the cease-fire.

Both sides levied accusations about cease-fire violations, specifically regarding the Strait of Hormuz, which Trump said would be reopened as part of the cease-fire. However, Araghchi, in an April 7 statement confirming Iran’s acceptance of the cease-fire, said that “safe passage through the Strait of Hormuz will be possible via coordination with Iran’s Armed Forces.” Since then, Iran has attacked and seized ships transiting the waterway that Tehran says are not complying with the Iranian terms.

The United States accused Iran of violating the terms the cease-fire and, in response, Trump announced a blockade on Iranian ships transiting the Strait. The U.S. Navy intercepted and seized an Iranian-flagged container ship, the Touska, that was enroute to Iran. In an April 22 post on X, U.S. Central Command said it turned around 31 ships, most of which were oil tankers traveling to or from Iran.

Iranian state-run media cited the U.S. blockade as a key reason why Iran decided not to return to Islamabad for talks April 21.

May 2026
By Xiaodon Liang

U.S. spending on nuclear weapons and delivery systems in the Pentagon budget will rise to $71.4 billion in fiscal year 2027 if Congress approves President Donald Trump’s $1.45 trillion departmental budget proposal.

That sum is a 15-percent jump over last year’s proposed Pentagon nuclear weapons budget of $62 billion. Both figures include proposed appropriations through the regular appropriations process and through budget reconciliation bills.

Trump is asking Congress to appropriate $1.1 trillion for the Department of Defense through the regular process, while banking on the Republican-controlled Congress to provide an additional $350 billion through reconciliation.

That request may prove a risky bet, with Senate Budget Committee chair, Sen. Lindsey Graham (R-S.C.), opting to exclude additional defense spending in a budget reconciliation bill for fiscal year 2026 that was released April 21.

Congress will still have a chance to pass an additional fiscal year 2027 budget reconciliation act later in the year. Any such effort would face daunting prospects with both chambers of Congress in play in the November midterm elections.

“Regular order appropriations are the right way to meet the scale and scope of the requirements of our military,” said Sen. Mitch McConnell (R-Ky.), chair of the Senate Appropriations Subcommittee on Defense, expressing skepticism of the reconciliation strategy in an April 3 press release.

His minority counterpart, Sen. Chris Coons (D-Del.), opposed the president’s top line in his press release on the White House’s budget, claiming that “this request defies the bounds of what is needed.” Coons added: “The president thought of the highest number he could fathom and then tweeted it.”

The $350 billion reconciliation portion of the defense request includes large sums for defense industrial base expansion, autonomous weapons, artificial intelligence investments, and munitions. Unlike last year’s reconciliation act, this year’s bill only includes small additions to nuclear weapons spending.

The increase in this year’s nuclear weapons budget is driven by modernization programs entering full-scale production.

The Columbia-class submarine would cost taxpayers $16.1 billion under the president’s proposal, spread across shipbuilding and research and development accounts. This year’s budget request, which provides full funding for the fourth boat of the class, is slightly more than two-thirds higher than the $9.6 billion enacted last year.

The Navy projects that the 12-boat program will have a lifetime shipbuilding cost of $146.4 billion. That is nearly 16 percent higher than the $126.4 billion estimate in the fiscal 2025 budget request.

The Navy may buy three more Columbia boats, Robert Kadlec, the assistant secretary of defense for nuclear deterrence, chemical, and biological defense, policy and programs, told the House Armed Services Committee's Strategic Forces subcommittee April 23.

Speaking before the Senate Armed Services Committee March 26, the chief of U.S. Strategic Command, Adm. Richard Correll, said that “additional capacity and capability at sea, in terms of launchers, is very beneficial from my perspective.”

The Navy still expects to receive the first Columbia-class boat by the end of 2028, Vice Adm. Robert Gaucher, the service’s direct reporting portfolio manager for submarines, said April 20 at an industry conference.

A life-extension program for the Trident II submarine-launched ballistic missile—the second such program for the missile—will exit an R&D phase and move fully into procurement in fiscal 2027, leading to an increase in costs from $2.6 billion enacted last year to $3.9 billion. Costs for R&D work on the Mark 7 aeroshell for the W93 warhead would rise from $619 million to $841 million.

The Navy did not request additional funding for the nuclear-capable, sea-launched cruise missile this year as it continues to spend down last year’s reconciliation money. The service anticipates asking Congress for $6.1 billion more, however, in R&D funding for the missile through fiscal 2031.

The B-21 bomber program costs are also increasing as low-rate initial production continues, with spending rising from $5.6 billion in enacted funds last year to $6.1 billion under the president’s proposal.

Although the prior year’s reconciliation act provided an additional $4.5 billion to expand bomber production capacity, the Air Force has yet to announce an increase in the service’s planned fleet, which currently targets 100 planes.

In written testimony March 17 to the House Armed Services Committee, Correll confirmed that U.S. Strategic Command is continuing to advocate for more bombers but also that “assessments to determine the final procurement quantity” were ongoing. (See ACT, May 2025.)

His counterpart at Indo-Pacific Command, Adm. Samuel Paparo, called for 200 B-21s at an April 21 Senate Armed Services Committee hearing.

Kadlec said Apr. 23 that the Air Force will buy more of the Long-Range Standoff Weapon, a nuclear-tipped, air-launched cruise missile that will arm the B-21 and B-52H strategic bombers. Procurement costs for the weapon, which is undergoing integration flight tests with the B-52H, will total $11.8 billion over the program’s lifetime, budget documents suggest. Two years ago, those projected costs were $9.8 billion.

The budget request calls for spending $1.5 billion on the missile in fiscal 2027, compared with about $790 million enacted last year.

Ongoing design work for the air-delivered nuclear delivery system, an early-stage program for a nuclear bunker-buster weapon, would cost $91.6 million in fiscal 2027. (See ACT, March 2026.)

R&D and procurement spending on the Sentinel intercontinental ballistic missile (ICBM) will fall slightly, from $5.0 billion appropriated for fiscal 2026 to $4.6 billion proposed for fiscal 2027.

The Air Force anticipates attaining Milestone B approval for the Sentinel program by the end of 2026, following significant cost overruns and a subsequent program restructuring. (See ACT, September 2024.)

The Air Force and lead contractor Northrop Grumman broke ground in February on a test silo that will validate a new modular silo launcher design, U.S. Strategic Command announced in a Feb. 17 press release.

Although military construction costs for the Sentinel program would rise under the budget request, no money is earmarked for large-scale silo construction. Instead, roughly $1 billion would go toward construction of support buildings, land acquisition, and a utility corridor. Congress appropriated $110 million for similar construction last year.

Gen. Dale White, the Air Force’s direct reporting portfolio manager for critical major weapon systems, said the Pentagon will produce a new cost estimate for the restructured Sentinel program this summer.

The budget zeroes out two polar-orbit missile warning satellites, while funding for two geosynchronous satellites grows from $1.4 billion to $1.5 billion.

Spending on the program, known as the evolved strategic satellite communications program, to replace the legacy advanced extremely high frequency satellite-based communications network would also nearly double from $1.1 billion to $2.0 billion.

Other NC3 programs include the E-4 Survivable Airborne Operations Center, which would rise from $1.8 billion to $2.2 billion, and replacement of the E-6B Mercury—the link between headquarters and Navy ballistic missile submarines—with the E-130J, which would increase from $1.2 billion to $1.9 billion.

Proposed spending on missile defense this year is divided between $67.9 billion for the Missile Defense Agency and the armed services and a defense-wide special Golden Dome fund that is allocated $17.9 billion. In total, this $85.8 billion proposed budget is more than 25 percent above the corresponding $68.3 billion request in fiscal 2026.

Substantial reconciliation funding would go toward replenishing stocks of theater medium- and intermediate-range missile interceptors, expended during the U.S.-Israeli war against Iran that began in February.

The budget proposes spending $4.2 billion on procurement of the SM-3 Block IIA missile, which was tested against an ICBM-range target in December 2020, up from $445 million. That sum would buy 114 interceptors, compared with 12 last year. But of those 114, 88 would be contingent on Congress delivering reconciliation funds.

Similarly, the Army would spend $11.4 billion to purchase 857 Terminal High Altitude Area Defense (THAAD) interceptors under the budget proposal. Of those interceptors, 830 would be paid for through reconciliation. Budget documents indicate the Army will continue buying interceptors at a cost of about $6 billion a year through fiscal 2031, equivalent to more than 450 interceptors annually.

These large purchases would replenish and deepen Pentagon stockpiles. According to an April 21 analysis from the Center for Strategic and International Studies, the Navy has expended between 130 and 250 of its SM-3 interceptors—including older Block IB missiles—out of a prewar inventory of 410. Likewise, the Army has used between 190 to 290 of its prewar inventory of 360 THAAD interceptors.

The Washington Post reported April 8 that the White House still intends to send Congress an additional separate request for $80-100 billion for Iran war costs.

Ongoing Missile Defense Agency R&D programs would also receive a boost under the new budget. Work on the Ground-Based Midcourse Defense system would rise from $764 million to $1.4 billion. R&D on the glide-phase interceptor would increase from $185 million to $214 million and on ballistic missile defense sensors, it would quadruple from $183 million to $865 million.

The Pentagon did not provide line-item details on the Golden Dome fund—almost entirely a reconciliation-funded proposal—choosing instead to name several priorities, including space-based interceptors, a space-based hypersonic and ballistic missile tracking satellite network, a missile defense underlayer, non-kinetic options, and integration efforts.

The Space Force announced April 24 it had entered into agreements with 12 companies for work on space-based interceptors over the past half year. Those companies stand to earn up to $3.2 billion in awards through the competitive arrangement to develop a proliferated constellation of boost-phase, midcourse, and glide-phase interceptors. (See ACT, November 2025.)

The Pentagon’s Cooperative Threat Reduction program will endure a 22-percent cut to $221 million under the new budget, with its biological threat reduction program accounting for most of the reduction. The program’s budget request suggests it will seek to offload the cost of biological threat programs in Europe to regional allies and partners.

May 2026
By Michael Klare

Artificial intelligence has played a major role in selecting targets for attack during Operation Epic Fury, the U.S. air and missile campaign against Iran that began Feb. 28. Senior U.S. officials said that the Pentagon relied on an AI-powered data-fusion and decision-support program, the Maven Smart System, to identify top-priority targets and help choose the weapons used in attacking them.

According to an April 8 White House accounting, the U.S. military struck more than 13,000 targets in Iran during the first 38 days of the war, including more than 2,000 command-and-control targets, 1,500 air defense targets, and 1,450 industrial base targets. The White House claimed that all were legitimate military targets, but The New York Times and other news media reported the visual destruction of civilian facilities. This led critics to claim that overreliance on AI resulted in targeting errors and unnecessary civilian casualties.

As employed in Iran, the Maven Smart System, produced by Palantir Technologies Inc., collected information on enemy positions from radar signals, satellite and drone imagery, electronic communications, and other sources and combined it into a “common operating picture” of the battlefield, Cameron Stanley, the Pentagon’s chief digital and AI officer, told Palantir’s AIPCon 9 conference March 12. He said that Maven was also used to identify the friendly unit best positioned to strike any given target and to generate courses of action for U.S. strike units to follow, encompassing attack vectors, munitions to be employed, and other pertinent data.

“Instead of having eight or nine systems” for commanders to consult when making combat decisions, Maven fuses everything “into a single visualization tool” for use in decision-making, Stanley explained. Once so equipped, commanders can identify a target and then select a strike package to engage it, simply by clicking on a screen, he said. “So, we’ve gone from identifying the target to now coming up with a course of action to now actioning that target, all from one system. This is revolutionary.”

Stanley said that Maven has become increasingly sophisticated over time. When initiated in 2017 by the Pentagon’s Algorithmic Warfare Cross-Functional Team (later the Joint Artificial Intelligence Center), Project Maven was designed to use computer vision and machine learning to sort through drone footage of Middle Eastern battlegrounds and identify potential militant hideouts for possible attack. In the years since then, Palantir has steadily improved the technology, now renamed the Maven Smart System, enabling it to collect and collate data from multiple sources and to recommend possible combat moves.

In late 2024, Anthropic’s Claude AI operating system was merged with the Maven technology to provide military users with enhanced targeting options. Commanders can now use the combined system to generate target lists by assorted criteria, such as radar station, missile battery, communications node, and senior commander, and rank them by strategic importance. Once a target has been attacked, the system can review damage assessment reports and automatically produce new target lists—all in a matter of minutes.

Anthropic has now been barred from providing services to the U.S. military because of its refusal to work on autonomous weapons systems and domestic surveillance operations; other AI companies, including OpenAI, have been recruited to assume Anthropic’s role.

For U.S. military officials, Maven’s speed in selecting and reselecting targets represents a distinct combat advantage, allowing U.S. forces to disable Iranian combat capabilities swiftly and incessantly, preventing their reconstitution.

“Our war fighters are leveraging a variety of advanced AI tools,” said Adm. Brad Cooper, commander of U.S. Central Command, in a March 11 video briefing. “These systems help us sift through vast amounts of data in seconds so our leaders can cut through the noise and make smarter decisions faster than the enemy can react,” Cooper said.

But this very speed in target selection is what worries many observers, given the risk that sites will be chosen for attack without adequate human oversight. Although human officers supposedly review every target before a strike order is issued, there is a growing danger that “humans may rely too much on the system,” and fail to double-check its recommendations, Nilza Amaral, head of research at Chatham House’s Global Governance and Security Centre, told The National, an Abu Dhabi-based newspaper.

With humans granted ever-diminishing time in which to review AI-derived targeting decisions, the risk of error naturally increases, many analysts say. Whether AI played a role in the Feb. 28 U.S. cruise missile strike on the Shajareh Tayyebeh girls' elementary school in Minab, Iran, resulting in the deaths of over 170 people, most of them children, is unclear. According to a preliminary assessment by Central Command, U.S. intelligence maps failed to indicate that the school facility—once part of a military base—had long ago been converted to civilian use, and had been added to an AI-generated target list without adequate human supervision. The New York Times reported March 11 that "officials said the error was unlikely to have been the result of new technology" but rather “human error in wartime.”

Although human error—a failure to update military intelligence maps—has been deemed the most likely cause of the Shajareh Tayyebeh missile strike, many observers warn that increased reliance on AI-powered targeting systems, a phenomenon known as “automation bias,” will result in further tragedies of this sort.

“There’s a concern that targeting [approval] could end up just being a mere formality because of the automation bias, where people are just relying on what the machine is telling them,” Amaral said.

May 2026
By Xiaodon Liang

The Department of Energy’s National Nuclear Security Administration is taking steps to design and produce new nuclear weapons faster, according to recent legislation and budget justification documents for fiscal year 2027.

The semi-autonomous agency responsible for U.S. nuclear weapons requested from Congress a total of $32.8 billion for this coming fiscal year in President Donald Trump’s budget proposal, released April 3. The vast majority of the spending, $27.4 billion, would be devoted to nuclear weapons activities, up from the $24.2 billion appropriated by Congress for fiscal 2026.

Expanded work on new nuclear weapons would be paid for in part by a proposed cut to the department’s environmental management division, which is responsible for cleaning up past weapons production, from its fiscal 2026 enacted budget of $8.6 billion to $8.2 billion.

The NNSA will reorganize its stockpile, research, technology, and engineering portfolio, a major top-line component of the agency’s work, by creating a unified “rapid and advanced capabilities” program.

The new program, which consolidates existing efforts in two older programs, has the mission of “assessing, designing, prototyping, and demonstrating advanced systems, as well as accelerating the delivery of new nuclear capabilities to meet national security needs.”

The new program will have a budget of $499 million in fiscal 2027 under the president’s proposal, up from the $305 million appropriated for the constituent offices last year.

The reorganization follows Congress’ instruction in Section 3113 of the fiscal 2026 National Defense Authorization Act that the NNSA stand up a rapid capabilities program. (See ACT, January/February 2026.)

Beyond creating a new top-level program, the NNSA also has responded by setting up a separate “cross-program, cross-enterprise” nuclear deterrent rapid capabilities team to “rapidly develop advanced system prototypes and ensure readiness for accelerated acquisition and fielding,” according to the agency’s budget request.

Three novel weapons designs are hinted at within a new “futures program” budget line, which has a first-year price tag of $100 million. These include a “Nuclear Deterrent System-Air-Delivered,” which the Air Force has already begun work on and is likely a modern bunker-buster bomb. (See ACT, March 2026.)

The other two weapons in the futures program may match ongoing studies within the Rapid and Advanced Capabilities office: a concept assessment for the WXX program, and a “pre-phase” study of a future strategic sea-based warhead. Last year’s budget request described the WXX as part of the “next generation re-entry capabilities” program.

The April budget request also funds several existing nuclear weapons programs. The W80-4 life extension program would receive $1 billion in fiscal 2027, down from $1.3 billion. The NNSA expects to achieve in fiscal 2027 the first production unit of the warhead for the Long-Range Standoff Weapon despite the budget cut, claiming that a carry-over of past appropriations will provide adequate funding.

The variant for the naval-capable, sea-launched cruise missile, now renamed the W80-5, would not receive any funding through regular appropriations under the president’s budget plan, as the office spends down the $2 billion provided in last year’s budget reconciliation act.

Continuing design and development of the W87-1 warhead would cost taxpayers $913 million in fiscal 2027, up from $649 million last year as appropriated by Congress. Budget documents attribute some of the increase to a new weapon development cost report, suggesting that there have been recent revisions to the NNSA’s estimate of the program’s total lifetime price. A prior cost report from fiscal 2023 put the weapon, intended for the new Sentinel intercontinental ballistic missile, at $15.9 billion in future-years’ dollars.

The W93 submarine-launched ballistic missile warhead will move into development engineering in fiscal 2027, according to NNSA plans, with an expected price tag of $1.1 billion for next year, rising from $807 million enacted last year.

The B61-13 high-yield gravity bomb would be funded at $46 million, down from $49 million the year before.

Continuing work to overhaul NNSA plutonium pit production capabilities and related facilities would cost $4.9 billion under the request, up from $3.4 billion spent last year through regular appropriations and reconciliation funding. That price is also more than 20 percent higher than what the Biden administration expected to spend on plutonium modernization back when it was preparing the fiscal 2025 budget request.

Although the agency says it remains focused on achieving established plutonium pit production goals in this year’s budget request—30 pits per year at Los Alamos National Laboratories and 50 per year at the Savannah River Site’s future Plutonium Processing Facility—costs for plutonium pit production could rise again if that goal is adjusted. In a draft environmental impact statement released April 10 for its two-site pit production plans, spread across Los Alamos and the Savannah River Site, the agency evaluates an expanded proposal for production of up to 80 plutonium pits per year at Los Alamos and 125 at Savannah River.

Although the impact statement notes that production of 30 pits per year at Los Alamos and 50 a year at Savannah River remains national policy, this marks the second time the NNSA has suggested an expanded production rate at Los Alamos. (See ACT, April 2026.)­­

May 2026
By Kelsey Davenport

The head of U.S. Forces Korea confirmed that the United States is moving missile defense interceptors from South Korea to the Middle East. The decision is another sign that defending against Iranian missile strikes is straining the already-limited U.S. stockpile of missile defense interceptors.

General Xavier Bunson, commander of U.S. Forces Korea, said April 21 at a Senate Armed Services Committee hearing that the Terminal High Altitude Defense Systems (THAAD) remain in South Korea and are expected to stay there, but he acknowledged that the United States is shipping out some of the systems’ interceptor missiles. Bunson said the THAAD systems were moved around in preparation for transferring the interceptors and that the movement of the systems caused a “big kerfuffle on the peninsula.”

The United States first deployed THAAD to South Korea in 2017 to defend against North Korean ballistic missiles. A THAAD battery costs about $1 billion and typically includes a radar to detect and track missiles, six launchers, and eight interceptors per launcher. THAAD interceptors are designed to shoot down short- and medium-range ballistic missiles. In January, the Defense Department and Lockheed Martin announced an agreement to increase annual production from 96 interceptors to 400, but it will take seven years to reach that target, according to a Jan. 29 press release from Lockheed Martin.

In March, several media outlets reported that the United States was relocating missile defense systems from the Pacific region to the Middle East to defend against Iranian strikes. When asked about the reports, South Korean President Lee Jae Myung told journalists March 10 that U.S. Forces Korea “may dispatch some air defense systems abroad in accordance with its own military needs. He added: “While we have expressed opposition, the reality is that we cannot fully push through our position.” Lee  later said in a cabinet meeting that the transfer would not “seriously hinder our deterrence strategy against North Korea.”

The U.S. Defense Department already operates several THAAD batteries in the Middle East, at least one of which may have been damaged by an Iranian missile that penetrated U.S. defenses. The decision to move interceptors from South Korea suggests that Iran’s attacks put pressure on existing missile defenses and the available stockpile of interceptors in the region.

The New York Times reported April 26 that, “The Pentagon used more than 1,200 Patriot interceptor missiles in the war, at more than $4 million a pop, and more than 1,000 Precision Strike and ATACMS ground-based missiles than, leaving inventories worrisomely low,” according to internal Defense Department estimates and congressional officials.”

Bunson did acknowledge that the United States had moved radars out of South Korea before Operation Midnight Hammer, the U.S. strikes on Iran’s nuclear facilities in June 2025. He said some of what was transferred ahead of the June strikes “have not come back yet.”

U.S., Gulf Arab-state, and Israeli missile defense systems have proven successful against Iranian systems. Experts assess that 90-92 percent of Iranian missiles are intercepted. But Iran also launched maneuverable missiles and systems armed with cluster munitions, or multiple small bombs that are more challenging to intercept.

The United States also redeployed Patriot missile systems from the Indo-Pacific and Europe, according to reports from the Associated Press in March 2026.

The Patriot missile defense system is designed to intercept drones, cruise missiles, and short-range ballistic missiles. A Patriot system includes radar, launchers, and interceptor missiles.

After Iran fired ballistic missiles toward Turkey in early March, the United States moved two Patriot systems from Germany to Turkey, according to the Turkish Ministry of Defense. Other Patriot systems were repositioned toward the Middle East, the official quoted by Associated Press said. The redeployments have raised concerns in Europe about defending against Russian threats, according to European officials.

May 2026
By Kelsey Davenport

North Korea announced it is developing cluster munitions and tested several short-range missiles armed with that type of warhead in April. As North Korea is investing in new missile capabilities, the head of the International Atomic Energy Agency (IAEA) warned that Pyongyang’s expanding nuclear program could accelerate the production of nuclear warheads.

North Korea tested five tactical-range ballistic missiles, the Hwasong-11Ra, armed with a new “cluster bomb warhead and fragmentation mine warhead” April 19. The missiles flew about 90 miles before landing in the Sea of Japan, according to a statement from South Korea’s Joint Chiefs of Staff.

Cluster munitions are designed to scatter smaller submunitions or bomblets over a larger area. In 2010, the Convention on Cluster Munitions, which bans the production and use of cluster munitions, entered into force, but North Korea is not a party to the treaty. The April 19 launch appears to be North Korea’s second test of a missile armed with cluster munitions. North Korea said it tested a different short-range missile armed with a cluster munition warhead as part of a series of launches April 7-8.

The state-run Korean Central News Agency (KCNA) said April 20 that the purpose of the test was to “verify the characteristics and power” of the warheads. North Korean leader Kim Jong Un was present for the launch and “expressed great satisfaction” with the test. According to KCNA, Kim said the warheads have “weighty significance in military actions to boost the high-density striking capability to quell a specific target area as well as the high-precision striking capability.”

Kim’s comments and the short range of the missiles suggest that North Korea will deploy these systems near the border with South Korea.

Several days before the launch, IAEA Director-General Rafael Mariano Grossi visited South Korea and warned about the implications of North Korea’s advancing nuclear program. Grossi told reporters in an April 15 press briefing that the agency has observed a “rapid increase in the operation of the Yongbyon [5-megawatt] reactor,” which produces plutonium into spent fuel that is reprocessed for nuclear warheads. Grossi said that this light water reactor and the “activation of other facilities … all of them point to a very serious increase in the capabilities” of North Korea to produce nuclear weapons “which is estimated at a few dozen warheads.”

One of the other facilities Grossi was referring to might be a new enrichment plant at Yongbyon. He reported to the IAEA Board of Governors March 2 that the agency was monitoring construction of a new facility at the Yongbyon complex that had similar dimensions as the uranium enrichment facility at Kangson.

Prior to Grossi’s visit, South Korean Unification Minister Chung Dong-young told his country’s parliament that North Korea was operating an enrichment facility at Kusong. The Kusong site had not been officially confirmed as a nuclear facility and, according to South Korean media, the United States is restricting some intelligence-sharing with South Korea in response to Chung’s comments.

According to Yonhap News Agency, the United States started restricting access to certain satellite intelligence at the beginning of April. The official quoted in Yonhap said the restriction does not affect South Korea’s military readiness or information-sharing about North Korea’s military activities.

After the media reported on the U.S. decision to restrict certain intelligence, Chung said his comments about Kusong 
were based on publicly available information and was not an intelligence leak.

South Korean President Lee Jae Myung defended Chung in an April 20 post on the social media site X. Lee said the information about the site was “widely known” and that he would look into “why such an absurd thing is happening.”

May 2026
By Libby Flatoff

Iran has threatened to withdraw from the nuclear Nonproliferation Treaty (NPT) in response to U.S. and Israeli military strikes on its territory. The implementing legislation requiring withdrawal was introduced in Iran’s parliament just weeks before states-parties met in New York to review the treaty and discuss steps to strengthen it.

Malek Shariati, a member of the Iranian parliament, introduced the legislation titled “Support for the Nuclear Rights of the Iranian Nation” March 28. The Iranian news agency WANA reported that the proposal is built on three main pillars: withdrawal from the treaty, repeal of the law on implementing the 2015 nuclear deal known as the Joint Comprehensive Plan of Action, and support for forming new international agreements with partner countries to advance peaceful nuclear technologies.

Article X of the NPT allows a state-party to withdraw “if it decides that extraordinary events, related to the subject matter of this Treaty, have jeopardized the supreme interests of its country.” The state-party must give a three-month notice of intent to withdraw and submit a statement on the extraordinary events that led to the decision to withdraw.

In a March 27 post on the social media site X, Ebrahim Rezaei, spokesman for the National Security and Foreign Policy Commission of Iran’s parliament, said that remaining a part of the NPT would be meaningless as it “has had no benefit for us.”

According to Iranian media reports, the bill is to be fast tracked “with triple urgency.” However, since Israel and the United States began strikes against Iran Feb. 28, the parliament has not met in person and there is no clarity on when sessions may resume. If the legislation passes parliament, it will still need to be reviewed by the Guardian Council, which has the power to veto it.

Iranian Foreign Ministry spokesman Esmail Baghaei, in remarks March 28 to Iranian media, confirmed the legislation had been introduced and reaffirmed that “the Islamic Republic of Iran has never sought, and does not seek, nuclear weapons.” He added that, “regarding remaining a party to this treaty, and notwithstanding our clear position on the prohibition of all weapons of mass destruction, this is a matter of genuine discussion within public opinion and at the parliamentary level.”

WANA quoted Fada Hossein Maleki, a member of the national security commission, as saying that the strikes by the United States and Israel provide an “opportune moment” for Iran to begin reconsidering its international commitments.

Iran submitted five working papers for the 2026 NPT Review Conference after the war broke out, suggesting that it still plans to participate in the conference, which runs April 27 to May 22.

If Iran withdraws from the NPT, the bedrock of the international nonproliferation regime, it will be the second state to do so, after North Korea illegally withdrew without proper notice in 2003. (See ACT, January/February 2003.)