June 2025
By Daryl G. Kimball

Nuclear and defense strategists have long understood that the developing and deploying of strategic missile interceptors is ineffective against determined nuclear-armed adversaries because it could lead them to build more numerous and sophisticated offensive missile systems—at a relatively lower cost and more quickly—to overwhelm and evade missile defenses.

Strategic missile defenses can never be effective and reliable enough to defend the entire United States, but they can still negate a portion of a potential missile attack and upset strategic stability. Unless limited by national policy or mutual agreement, as the 1972 Anti-Ballistic Missile (ABM) Treaty once did, ambitious U.S. missile “shield” proposals can spur adversaries such as China and Russia to enhance their strategic forces to maintain the ability to inflict a devastating nuclear retaliatory strike.

Consequently, even after the George W. Bush administration withdrew from the ABM Treaty in 2002 to rush deployment of a limited strategic interceptor program before the next election, U.S. policymakers have focused mostly on improving capabilities to address limited attacks from lesser North Korean missile threats rather than more substantial threats posed by the major nuclear powers.

Upending longstanding U.S. policy and ignoring strategic realities, U.S. President Donald Trump announced May 20 that he wants to spend at least $175 billion of taxpayer dollars for a crash strategic missile interceptor scheme dubbed “Golden Dome” to defend the continental United States against all missile threats, including those from China and Russia, “within three years.” Trump’s radical plan is technically complex, prohibitively expensive, and counterproductive.

Golden Dome, which is still in the conceptual phase, would patch together existing and possibly new ground- and sea-based interceptors and radars with the introduction of hundreds, if not thousands, of space-based sensors and missile interceptors. The system is designed to destroy incoming missiles by destroying them early in flight, targeting them midcourse, or seeking to hit them just before impact on their targets.

Members of Congress from both parties need to cut through Trump’s hyperbole and demand serious answers to basic questions about this initiative before writing a blank check to a few big defense contractors. What specific threat is it designed to defeat? What success rate is being sought? What is the plan for building the system? How will adversaries respond? What effect will it have on strategic stability and space security?

Clearly, Trump’s radical strategic missile defense scheme would be even more costly than advertised. Consider that more than $250 billion already has been allocated to Missile Defense Agency programs since 1985. The Congressional Budget Office recently estimated that a new space-based interceptor program would exceed $500 billion over the next 20 years. That program was originally scoped for the North Korean missile threat of 10 years ago, not today’s threat of the peer and near-peer states that Trump wants to defend against. Even Sen. Tim Sheehy, (R-Mont.), who announced plans to form a Golden Dome caucus, said: “It will likely cost in the trillions if and when Golden Dome is completed.”

No matter how much money is spent on exotic crash programs, however, an effective U.S. missile defense system will not be feasible anytime soon, which means that it will not affect strategic stability in the near term. As a result, Chinese and Russian military planners would be smart not to overreact by building up their strategic offensive nuclear forces.

If Trump’s strategic missile defense plan is eventually implemented, however, it would destabilize the already precarious nuclear balance of terror that exists among the major nuclear-armed powers. It would heighten the danger that space will become a war zone and push China and Russia to further improve and expand their strategic offensive capabilities.

Moscow is already developing new systems—such as an undersea torpedo, a hypersonic glide vehicle, and a nuclear-powered cruise missile—to ensure it can overcome any future U.S. missile defenses. It is also developing a capability for a nuclear-armed anti-satellite weapon, which underscores the vulnerabilities of a space-based interceptor network. Beijing, meanwhile, has already begun to respond to U.S. missile defense and conventional long-range strike capabilities by increasing its nuclear-armed ballistic missile force.

These factors make it even more vital that Russia and the United States engage immediately in strategic stability talks and negotiations to cap or reduce their massive strategic nuclear arsenals before the last remaining treaty, the New Strategic Arms Reduction Treaty, expires in 2026. China, too, needs to back up its complaints about U.S. unilateralism, strategic missile defense, and space-based interceptors by agreeing to participate in direct, high-level, strategic risk-reduction talks with the United States.

If Trump, Russian strongman Vladimir Putin, and Chinese dictator Xi Jinping continue to bicker in public rather than negotiate mutually beneficial limits on strategic offensive and defensive weapons, the world faces a more dangerous future. 

June 2025 
By Igor Moric

Impressed by the effectiveness of the Israeli Iron Dome defense system against short- to medium-range rockets and drones, U.S. President Donald Trump issued an executive order January 27 directing ambitious plans for an “Iron Dome for America.” The initiative, since renamed Golden Dome, maintains the goal of defending the United States against ballistic, hypersonic, and advanced cruise missiles from peer, near-peer, and rogue adversaries. In an announcement May 20, Trump stated that $25 billion had been allocated initially in the defense budget to begin constructing a missile defense architecture comprising land-, sea-, and space-based interceptors and sensors, “forever ending” the missile threat to the homeland by the end of his term.

Although Trump was correct to warn recently about the destructive capability of nuclear weapons, the need to reduce nuclear arsenals, and the massive budgets being spent on them,¹ pursuing the fantasy of national missile defense will accomplish the opposite; it will incentivize China and Russia to double down on building up their nuclear arsenals, it will cost the United States hundreds of billions of dollars, and it  
will not work as intended.

An invincible shield that can safeguard the United States from nuclear war just as a “roof protects a family from rain” is a captivating idea that survives despite many past failures.² Use of metaphors such as shields or domes made of iron or gold, or bullets hitting bullets, combined with a lack of publicly available information and a limited general understanding on how missile defense works, perpetuate the illusion that with enough time and money, technology will advance to eventually enable protection against all external threats.

Over the years, this has spurred over-ambitious attempts at building missile defense systems, with the momentum collapsing once limitations become apparent and policymakers recognize that even attempting to establish such a capability adds an element of strategic instability because of increasing pressures for the other side to deploy additional missiles to counter it.

In 1983, U.S. President Ronald Reagan addressed the nation, asking whether people wanted to continue relying on the threat of instant U.S. retaliation to deter a Soviet attack. Troubled by the moral implications of mutually assured destruction and the vulnerability of strategic ballistic missiles, and driven by the technological optimism of his science advisers, Reagan proposed the Strategic Defense Initiative (SDI). This effort aimed to stabilize the relationship between offensive and defensive nuclear arms and ultimately free the world from the threat of nuclear weapons by rendering them impotent and obsolete.

Due to its futuristic vision of missile defense, the program was derisively nicknamed “Star Wars.” It included ambitious concepts such as space-based lasers and particle-beam weapons, and a network of ground- and space-based interceptor systems. Contrary to Trump’s claim in the executive order, SDI was not prematurely cancelled. It was an obstacle to arms control, encouraged the buildup of Soviet offensive nuclear capabilities, and ultimately could not meet its technical requirements even in a scaled-down version. The system could be evaded and penetrated, and the cost dynamics favored the attacker, who had a major advantage by being able to “overwhelm” the interceptors with relatively cheaper warheads or “outfox” them with decoys.³

Star Wars: A New Hope

After the September 11 attacks, concerns over missile threats from terrorist and rogue states gained political momentum in the United States. This provided the Bush administration with an opportunity to withdraw from the Anti-Ballistic Missile Treaty in 2002, removing all restraints on development of missile defense systems. Over the past two decades at a price of more than $200 billion, the incremental deployment of building blocks of national ballistic missile defense has continued unchecked, driven by steady technological creep, bureaucratic inertia and policy neglect, without much reflection on the cost or its strategic implications.⁴

Although not formally designated as such, the U.S. Missile Defense Agency (MDA), already has been constructing its own version of the dome, even before Trump’s executive order. Without specifying the origin or the extent of the threat, the MDA plans envision a layered network to “defend the United States, its deployed forces, allies, and friends from missile attacks in all phases of flight.”

The first Trump administration’s 2019 Missile Defense Review maintained that the focus of U.S. missile defense was against “rogue states,” but the systems MDA has been developing are increasingly capable of degrading Russian and Chinese second-strike capabilities, causing concern in Moscow and Beijing. The recent order simply codified longstanding ambitions to reconfigure U.S. ballistic missile defense architecture to defend against large-scale attacks, enforcing Trump’s past public statements that the goal of missile defense is to “detect and destroy any missile launched against the United States, anytime, anywhere, and any place.”⁵

As with the proposed Golden Dome, MDA’s planned architecture is to include a few types of interceptors, supported by a combination of ground- and sea-based radars, hundreds of infrared sensors to detect launches, and communication satellites to distribute data to interceptors. In cooperation with the Space Development Agency, the roadmap calls for a continuously replenished satellite constellation, including a “birth-to-death” tracking and fire control sensor developed in the Hypersonic and Ballistic Tracking Space Sensor program and referenced in the recent EO.

Although ballistic missile defense systems have been a longtime interest of the United States and Trump, the release of the executive order was likely influenced by recent events. In the April and October 2024 attacks on Israel, most of the launched Iranian ballistic missiles—about half of which failed at launch or in flight—were intercepted by defense systems operated by Israel and its allies. These systems, as well as Patriot interceptors operated by Ukraine in defense against the Russian invasion, demonstrated effectiveness at guarding sites against slower threats traveling in the atmosphere, including drones and medium- and short-range ballistic missiles.

Such successes by theater missile defense systems are widely publicized, creating an inaccurate public perception about missile defense in general. Replicating the same performance on a national scale is a very different problem.⁶ The Israeli Iron Dome effectively defends an area the size of New Jersey; intercepting faster and higher flying nuclear armed long-range ballistic missiles over a country 450 times larger is far more complex. Once incoming reentry vehicles traveling at hypersonic speeds enter the atmosphere at an altitude of around 100 kilometers, U.S. terminal 

phase interceptor systems such as Terminal High Altitude Area Defense and Patriot have only a few minutes to reach the intercept point, limiting the area they can realistically defend.

The most efficient way to defend a wide area is to destroy ballistic missiles in their boost phase, within 3-5 minutes after launch, while their engines are still firing and before reentry vehicles have been separated from the missile payload. During this phase, the missile’s bright exhaust makes it easier to detect the launch from space. Due to geographical realities and time constraints, however, U.S. land and sea-based interceptor systems cannot be positioned close enough to engage launches from deep within Russian or Chinese territory in a timely fashion.

Enter space-based interceptors, a capability referenced in the order and previously explored in SDI. This would involve a mega-constellation of orbiting satellites in low Earth orbit (LEO) carrying interceptor missiles and potentially other non-kinetic weapons such as lasers.

Because LEO satellites are constantly orbiting Earth at high velocities, many would be needed to provide coverage over any specific region. For example, to cover a single launch of the North Korean Hwasong-18 intercontinental ballistic missile (ICBM) and defend most locations in the United States (excluding Alaska), it is estimated that the United States would require at least 1,600 orbiting interceptors.⁷ Defending against missiles launched from China or Russia would demand many more, while also somehow avoiding reaction over the orbital placement of tens of thousands of weapons overflying their territory.

Attempting to construct a constellation capable of protecting against attacks by peer adversaries would be prohibitively costly, even accounting for the falling costs of space-launches.⁸ Additionally, as satellite orbits decay over time, orbiting interceptors would need to be continuously replenished. A single layer would not provide reliable protection, because the attacker could target the vulnerable satellites with relatively cheap missiles or maneuverable satellites, punching a hole through which long-range missiles could pass. China and Russia already possess such advanced anti-satellite (ASAT) capabilities. They also could detonate a nuclear explosion in orbit, blinding sensors detecting launches and tracking missiles, and polluting the interceptor orbit with debris.

Therefore, to ensure effectiveness of the shield, the United States would have to deploy multiple layers of interceptors in different orbits and a fleet of maneuverable satellites to defend them, multiplying the overall cost of the system and rendering the whole concept absurd.

Realistically, MDA’s near-term focus likely will remain building a patchwork of their existing interceptor systems, while continuing to expand space-based sensor and communication layers. Most of the agency’s efforts focus on intercepting ballistic missiles during their exo-atmospheric midcourse phase, in which missiles travel unpowered through space for 20 to 30 minutes and reentry vehicles are released from the payload.

MDA’s midcourse defense systems include the legacy Ground-Based Midcourse Defense (GMD), which consists of 44 “hit-to-kill” interceptors stationed in Alaska and California, the developing Next-Generation Interceptor (NGI) system with 64 interceptors that is to replace GMD, and the land- and ship-based Aegis Ballistic Missile Defense system, which uses shorter-range SM-3 IIA missiles with kinetic, non-explosive warheads.

Scripted tests performed under “controlled conditions” demonstrate that GMD interceptors have a single-hit kill effectiveness of just 56 percent, with all successful interceptions against ICBMs occurring during daytime. Aegis interceptors have only been tested successfully once against a short-range ICBM, also under optimal conditions where the launch times and the trajectory for the target missile were known in advance.

Because they were never tested in realistic conditions, the real-world reliability of GMD and Aegis systems against sophisticated threats remains unknown.

For example, none of the tests accounted for the possibility of jamming the interceptors or considered cyber or kinetic vulnerabilities of the defense. The target missiles were not allowed attempts to evade interception or carry more than one warhead. In practice, attacking ICBMs would carry multiple independently targetable reentry vehicles, each able to target a different location within some area. The missiles also could be launched with unexpected trajectories or from submarines near the U.S. coast, further limiting the system’s effectiveness.

In addition, as reentry vehicles containing warheads are released from the missile payload, the attacker would employ countermeasures designed to make interception more difficult. For example, some warheads could be replaced with cheap, lightweight decoys that can simulate the radar signature and trajectory of warheads. In LEO there is not much air resistance, and warheads and decoys move virtually the same, making discrimination challenging. The defender needs to identify decoys and ignore, or waste interceptors to destroy them.⁹

Given that there are many imaginative ways to confuse the defender, and the defender may only learn about the method used at the moment of the attack, the United States may end up launching some of its half-billion-dollar interceptors at chaff wires, decoys mimicking the signature of reentry vehicles, balloons, or bomblets, quickly exhausting the limited supply of interceptors without fully eliminating the threat.¹⁰

The Many Costs of the Golden Dome

In 1985, Reagan’s arms control adviser, Paul Nitze, outlined three criteria that a missile defense system must meet to be successful: It must work, it must be survivable, and it must remain cost effective at the margin. Most analyses of ballistic missile defense (BMD) focus on the first two requirements and examine the technological feasibility of the proposed systems. The 2011 report by the Federation of American Scientists, the 2012 report by the National Academy of Sciences, the 2016 report by the Union of Concerned Scientists, and the recently released report by the American Physical Society all indicate that at the moment, U.S. BMD systems have near zero capability against a large-scale Chinese or Russian nuclear attack, and only a very limited capability against a smaller-scale North Korean attack.¹¹

Shielded from public oversight, however, the U.S. government appears to have decided that the nearly half-trillion dollars spent so far on ballistic missile defense, and the $200 billion that was projected to be spent in 2020-2029, are not enough and that this time is different.¹² Despite decades of failures, the U.S. national BMD system, with enough funding and effort, may yet evolve into a reliable shield against nuclear ballistic missile attacks.

In a recent analysis,¹³ we therefore forwent questioning MDA’s techno-optimistic claims, and imagined that the technology required for ballistic missile defense with 90 percent overall system effectiveness had been reached, matching the goal of the GMD and mirroring the claimed effectiveness of the Israeli Iron Dome, in spite of vastly more demanding requirements.¹⁴ It also was assumed that the systems could not be easily jammed or destroyed by an adversary, and rather than asking how, we asked how much?

By accepting that all things work as advertised and focusing on missile defense costs, this analysis derived how many dollars the United States would need to spend to deploy land and ship-based defenses for each dollar China or Russia spend to field a major scale nuclear attack with their land-based ballistic missiles.

First, based on archival data and by adjusting for inflation and purchasing power, the cost of a hypothetical ICBM with adequate range to strike the United States from China or Russia was estimated at $42 million, including missile maintenance and construction of a silo launch facility. The analysis assumed that the attacker could deliver between 500 and 6,000 warheads, reflecting estimates of China’s expanding arsenal and Russia’s deployed and retired but not yet dismantled weapons.

Next, the analysis estimated the costs of establishing and maintaining a U.S. BMD system built around the planned ground-based NGIs and ship-based Aegis interceptors. The assumed unit cost for the NGI interceptor was about $487 million, based on a report released by the Cost Assessment and Program Evaluation office.¹⁵ For the Aegis ship interceptors, the approximate cost was $60 million per unit, including the SM-3 Block IIA missile, and its maintenance, and accounting for the cost of critical subsystems of the Aegis Combat System: the VLS-41 launcher and AN/SPY-6 radar.

Finally, a scenario was modeled where the United States established a two-layered defense, made possible by geographically distributing Aegis ships near trajectories of Chinese or Russian missiles. By considering the size of the defense and the attack, in a two-layered defense the defender can minimize the number of interceptors by being able to shoot, look to see how many threats have been destroyed, and shoot again. The BMD system was assigned an overall effectiveness of 90 percent, while individual interceptor kill-effectiveness was set at either 50 percent (a more realistic, but still highly optimistic case) or 90 percent (a wildly optimistic case, but targeted efficiency).

The results show that in an ideal case for the defender, where interceptors reach 90 percent kill effectiveness and the system can perfectly reject all countermeasures, the United States needs to spend on average 8 times more to defend itself than the attacker needs to spend to build, equip, and launch its missiles. This scenario assumes that each attacking ICBM carries five warheads, it excludes the cost of the defender’s sensors and ignores past and future projected missile defense technology development expenditures.

Recognizing that U.S. missile defense systems have not yet been proven to have any capability to discriminate decoys or identify countermeasures, we accounted for this in a more realistic scenario by assuming the attacker’s ICBMs carry two decoys per warhead, or 15 targets per missile. Because it is speculated that the attacker could pack many more decoys in the payload and use other countermeasures, this is still a highly optimistic case. With interceptor kill effectiveness set at the demonstrated 50 percent, while still ignoring the costs of development and external sensors, the United States needs to spend an average 70 times more than the attacker, making defense economically difficult to sustain.

These findings show that even if technical challenges could be overcome and the efficiency of U.S. national ballistic missile defense against faster long-range ballistic missiles matched the efficiency of the Israeli Iron Dome against shorter-range Iranian missiles and drones, the cost-exchange dynamics still heavily favor the attacker.

By extension, if the United States builds an expensive missile defense system that may never function as imagined, its adversaries have every incentive to respond by building nuclear weapons and delivery vehicles that can overwhelm the defense at a much lower relative cost. The United States may be able to operate with such an economic disadvantage against North Korea, but unless nuclear weapon arsenals are reduced dramatically a successful nuclear offense-defense race against China or Russia is unlikely to become feasible.

As technology improves and production is expanded, the cost of U.S. interceptors could decrease and match the cost of an attacker’s ballistic missiles relying on legacy 1960 technology. Yet, to detect missile launches, track reentry vehicles, and guide interceptors, the defender still requires a costly network of radars and space-based sensors.

That the cost-exchange dynamics favor offense was recognized by U.S. military planners since the beginning of missile defense programs. Relatively cheap countermeasures can nullify the effectiveness of any BMD system and, as Defense Secretary Robert McNamara argued in 1962, “no amount of money could make possible an absolute defense against ICBMs.” Unfortunately, decision-making on U.S. ballistic defense is often steered by political motivations and economic interests of the military-industrial complex, making it difficult to constrain the development drives by cost or strategic considerations.

With not much to show for it, by the end of the decade, the combined U.S. expenditures on ballistic missile defense since the late 1950s could reach many times more than the cost of the Manhattan Project, a few times more than landing the first men on the Moon, and double what is estimated to land humans on Mars.

Trump gave the Pentagon only 60 days to propose an architecture for the Golden Dome, but it took military contractors even less time to overpromise on something on which they ultimately will underdeliver. For example, Lockheed Martin called for a “Manhattan Project-scale” effort to deliver the first Golden Dome within a year. L3Harris praised Trump’s bold vision for Golden Dome, urging a dramatic acceleration to adopt commercial products to a “catastrophic threat.” SpaceX, led by Elon Musk, who also heads administration cost-cutting efforts, reportedly has teamed up with defense startups Anduril and Palantir to compete for the Golden Dome contract.¹⁶

Reason for the Need

When announcing SDI, Reagan called on the scientific community to develop BMD as a safer, morally preferable alternative to relying on nuclear deterrence based on mutual destruction. In 2019, Trump called for a program that can shield every U.S. city, and has since warned that there is no greater danger than the menace of nuclear weapons, signaling the need to strengthen U.S. security without relying on the threat of nuclear retaliation.¹⁷

Yet, as long as nuclear weapons exist, fully eliminating the danger of nuclear war is not a plausible choice. Even if Golden Dome is fully assembled and achieves 90 percent interception rate, 10 percent of incoming warheads would still get through. In the event of a major-scale attack, at least 60 Chinese and 150 Russian warheads would still strike targets in the United States.

The resulting fallout would inflict catastrophic damage across the country and trigger a U.S. nuclear response, regardless of the adversary’s targeting plans. Although there is no evidence a conflict between nuclear-weapon states would just fizzle out, there are plenty of indicators to suggest otherwise.

These include the recently declassified 1983 “Proud Prophet” wargame, which demonstrated how a limited nuclear war in Europe likely would escalate rapidly and catastrophically. Air Force Gen. John Hyten, commander of U.S. Strategic Command, described their nuclear wargaming exercises in 2017: “They end the same way every time. They end badly, with global nuclear war.” Finally, Trump himself warned that any nuclear weapon use would mean “probably oblivion.”¹⁸

National ballistic missile defense is technologically unfeasible, prohibitively costly, and bad for deterrence. Although it may be efficient in “mopping-up” missiles from an adversary’s limited retaliatory nuclear strike (in the unlikely case a hypothetical U.S. nuclear first strike manages to take out most of the other capabilities), this scenario also presumes that adversaries would passively observe the massive undertaking required to establish an effective national BMD system.

A defense with 90 percent effectiveness would allow only 10 out of 100 warheads to go through, but a strike consisting of 1,000 warheads would leak 100. While in a static case a missile defense system could enhance deterrence by reducing the adversary’s confidence that an attack will be successful, if it is not perfect, it also creates incentives for a nuclear arms race. Unless constrained by resources, adversaries will attempt to regain lost confidence by increasing the size of their nuclear arsenals or by modernizing to improve delivery vehicles.

In addition to building up its nuclear stockpile and adding hundreds of new ICBM silos, China has tested systems specifically aimed at evading missile defense, including by reviving the idea of placing nuclear weapons in fractional orbit. Meanwhile, Russia is modernizing its ICBMs and developing “exotic” nuclear systems to bypass missile defenses.¹⁹ Russia is also testing concepts for nuclear anti-satellite weapons, and has recently warned that any attempts to deploy U.S. space-based interceptors would further fuel the militarization of space.²⁰

Although U.S. missile defense efforts are not the only reason, they provide a pretext for Chinese and Russian military expansion and refusal to discuss nuclear disarmament and arms control. U.S. ballistic missile defense also is not deterring North Korea’s nuclear buildup, with Pyongyang rapidly closing its technological gap with U.S. peer competitors.

Justified by pseudo-rational military purposes and guarded from accountability by the protection of national security, the ideas of national ballistic missile defense periodically resurface to defend the United States against threats of its own making, driven by promises of impending technological breakthroughs and a sunk cost fallacy.

As with SDI, public perception and political commitments to the idea of what Golden Dome should be are conflicting with what it can be—limited by cost, physical and technological realities constraining its effectiveness, and strategic implications of its deployment.

Before deciding to pursue national ballistic missile defense, Trump’s administration should consider: Will the deployment of Golden Dome truly enhance U.S. national security interests, or will it increase the risks of nuclear escalation and fuel arms racing? Can such a system, with all its limitations, deter a Russian or Chinese attack? Should deterrence fail, can it meaningfully limit the consequences of nuclear war?

The United States also should further examine the technological feasibility of national ballistic missile defense before committing billions to building it. This is not a political question or one for deterrence theoreticians but for scientists and engineers. To start, Trump should direct MDA to set clear system requirements and conduct independently reviewed tests under real-world conditions, without limiting the types of techniques used to obtain favorable results.

In 2010, the JASON independent advisory group of experts produced a report on discrimination techniques for U.S. midcourse missile defense. JASON could be called upon again, given access to required classified data, and tasked with evaluating the credibility of planned interceptor and sensor architecture against Russian and Chinese ICBMs and submarine-launched ballistic missiles, and their delivery vehicles.

Alternatively, if Trump is serious about reducing unnecessary military spending, he could leverage administration experts with significant technical knowledge and without conflicts of interest, tasking them to evaluate the program objectively and ensure that its continuation aligns with goals of government efficiency and effective resource allocation.

ENDNOTES

1. “Remarks by President Trump at the World Economic Forum,” January 23, 2025.

2. Comparison of missile defense to a roof which protects family from rain comes from a President Reagan speech in 1986, “Remarks at the High School Commencement Exercises in Glassboro, New Jersey”, Ronald Reagan Presidential Library and Museum, June 19, 1986.

3. Hans A. Bethe, Richard L. Garwin, Kurt Gottfried, Henry W. Kendall, Carl Sagan, and Victor F. Weisskopf, “Why Star Wars Is Dangerous and Won’t Work,” The New York Review of Books, February 14, 1985.

4. Jaganath Sankaran, “The Delusions and Dangers of Missile Defense,” Arms Control Today, Vol. 53 (September 2023). Cost estimate from the U.S. Government Accountability Office, “Missile Defense: Addressing Cost Estimating and Reporting Shortfalls Could Improve Insight into Full Costs of Programs and Flight Tests,” February 2, 2022.

5. U.S. Department of Defense, “Trump Pledges to Protect America From Any Enemy Missile,” January 17, 2019.

6. Fred Kaplan, “Trump’s Order to Create a U.S. Version of Iron Dome Makes No Damn Sense,” Slate, January 29, 2025.

7. “Strategic Ballistic Missile Defense Challenges to Defending the U.S.,” American Physical Society, March 3, 2025, p. 30.

8. Congressional Budget Office, “Effects of Lower Launch Costs on Previous Estimates for Space-Based, Boost-Phase Missile Defense,” May 5, 2025.

9. The Restless Technophile, “ABM 101,” November 8, 2019.

10. For an overview of countermeasures see, Andrew M. Sessler et al., “Countermeasures: A Technical Evaluation of the Operational Effectiveness of the Planned US National Missile Defense System,” Union of Concerned Scientists, April 18, 2000; Union of Concerned Scientists video, “Missile Defense Countermeasures,” originally published in 2000 and uploaded to YouTube on December 19, 2011; and Richard L. Garwin, “Effectiveness of Proposed National Missile Defense Against ICBMs from North Korea,” March 17, 1999.

11. Reports in order: Yousaf Butt and Theodore Postol, “Upsetting the Reset: The Technical Basis of Russian Concern Over NATO Missile Defense,” Federation of American Scientists, September 2011; National Academies, “Making Sense of Ballistic Missile Defense,” 2012; Laura Grego, George N. Lewis, David Wright, “Shielded from Oversight,” June 23, 2016; “Strategic Ballistic Missile Defense Challenges to Defending the U.S.,” APS, March 3, 2025.

12. For cost estimates, see Frank N. von Hippel, “U.S. Expenditures on Ballistic Missile Defense Through Fiscal Year 2021,” October 26, 2021. For cost projections, see CBO, “Costs of Implementing Recommendations of the 2019 Missile Defense Review,” January 2021.

13. Igor Moric and Timur Kadyshev, “Forecasting Costs of U.S. Ballistic Missile Defense Against a Major Nuclear Strike,” Defence and Peace Economics, September 3, 2024.

14. Charles L. Glaser, Steve Fetter, “Should the United States Reject MAD? Damage Limitation and U.S. Nuclear Strategy toward China,” International Security, Vol. 41 (2016).

15. “New ICBM Interceptor to Cost $18 Billion,” Arms Control Today, June 1, 2021.

16. Lockheed Martin, “Golden Dome for America Revolutionizing U.S. Homeland Missile Defense,” accessed tktk 2025; Ken Bedingfield and Ed Zoiss, “Space-based capabilities are critical to enabling a missile shield for America,” Breaking Defense, March 6, 2025; “Exclusive: Musk’s SpaceX is frontrunner to build Trump’s Golden Dome missile shield,” Reuters, April 17, 2025.

17. “Remarks by President Trump and Vice President Pence Announcing the Missile Defense Review”, January 17, 2019; Shelby Talcott, “There is no greater danger: Trump unveils a 2024 missile defense plan,” Semafor, January 27, 2023.

18. William Langewiesche, “The Secret Pentagon War Game That ​Offers a Stark​ Warning for Our Times,” The New York Times, December 2, 2024; Speech by Gen. Hyten for the Mitchell Institute Triad Conference, U.S. Strategic Command, July 17, 2018. See also, “Trump wants denuclearization talks with Russia and China, hopes for defense spending cuts,” AP, February 14, 2025.

19. Federation of American Scientists, “Nuclear Notebook: Russian nuclear forces, 2024”, March 7, 2024; Hans M. Kristensen, Matt Korda, Eliana Johns, Mackenzie Knight, “Chinese nuclear weapons, 2025,” Bulletin of the Atomic Scientists, March 12, 2025.

20. Unshin Lee Harpley, “DOD Official Confirms Russia Is Developing an ‘Indiscriminate’ Space Nuke,” Air & Space Forces Magazine, May 2, 2024. See also, Dmitry Antonov, “Russia condemns Trump missile defence shield plan, accuses US of plotting to militarise space,” Reuters, January 31, 2025.