Analysis

$12 Trillion on Weapons We Can Never Use

The Most Expensive Insurance Policy in Human History — Or the Most Dangerous Bet.

Since 1945, the United States has spent over $12 trillion (in 2024 dollars) on nuclear weapons — designing them, building them, testing them, maintaining them, and planning to use them. It currently possesses 5,500 warheads, enough to destroy every major city on Earth several times over. It has never used one in combat since Nagasaki. And now it plans to spend $1.7 trillion more replacing the entire arsenal with new weapons. The logic of nuclear deterrence is simple: we must spend trillions on weapons we can never use, because if we ever do use them, civilization ends. The question nobody in Washington asks: is there a cheaper way to not end civilization?

By the Numbers

$12.3T

Total US nuclear weapons spending since 1945 (2024 dollars)

Brookings Institution, Atomic Audit

$1.7T

Planned modernization cost over next 30 years (2023-2053)

CBO, 2023 Report

5,500

Nuclear warheads in the current US stockpile

Federation of American Scientists

70,300

Peak US stockpile (1967) — enough to destroy civilization many times over

DoE/NNSA

1,150+

US nuclear tests conducted (1945-1992)

CTBTO

22+

Known nuclear close calls where accidental launch was narrowly avoided

Union of Concerned Scientists

$12 Trillion Over 80 Years: A History of Nuclear Spending

Manhattan Project (1942-46)

$30BStockpile: 3

Two bombs used. One tested. The beginning of everything.

Early Cold War (1946-55)

$280BStockpile: ~2,400

Hydrogen bomb development. Castle Bravo (15 MT). Arms race begins.

Massive Retaliation (1955-65)

$1.4TStockpile: ~31,000

Eisenhower's "more bang for the buck." ICBMs, SLBMs, B-52s. Peak production.

MAD Era (1965-80)

$2.8TStockpile: ~70,000 peak

Mutual Assured Destruction. MIRV warheads. Trident submarines. Minuteman III.

Reagan Buildup (1980-90)

$2.1TStockpile: ~60,000

Star Wars (SDI), MX Peacekeeper, B-1B, stealth bomber development.

Post-Cold War (1990-2010)

$2.2TStockpile: ~10,000→5,000

Stockpile reduction. But infrastructure maintenance costs stayed high.

Modernization Era (2010-24)

$1.1TStockpile: ~5,500

New ICBM (Sentinel), new sub (Columbia), new bomber (B-21), new warheads.

Planned (2024-53)

$1.7TStockpile: ~3,700 (projected)

Full triad replacement. Every major system replaced simultaneously.

The $1.7 Trillion Modernization: Replacing Everything, All at Once

The United States is replacing every major component of its nuclear arsenal simultaneously — new ICBMs, new submarines, new bombers, new warheads, new command and control. This is the most expensive military procurement program in history, and it's already over budget before most programs have entered production.

Sentinel ICBM (GBSD)

$141B

Replaces: Minuteman III (deployed 1970)

Timeline: 2029-2075

Over budget (85% cost overrun). Nunn-McCurdy breach in 2024. Continuing anyway.

Columbia-Class Submarine

$132B (12 subs)

Replaces: Ohio-class (deployed 1981)

Timeline: 2031-2042

First boat delayed 1-2 years. Each sub carries 16 Trident II D5 missiles (192 warheads per sub).

B-21 Raider Bomber

$203B (100 aircraft + operations)

Replaces: B-2 Spirit (deployed 1997)

Timeline: 2026-2060+

First flight December 2023. Unit cost: $753M each (before overruns).

W93 Warhead

$14.8B

Replaces: W76/W88 submarine warheads

Timeline: 2034-2040

New warhead design — the first new nuclear warhead in 35+ years.

W87-1 Warhead (for Sentinel)

$12.1B

Replaces: W78/W87-0 ICBM warheads

Timeline: 2030-2036

New pit production required. LANL/Savannah River capacity inadequate.

Long-Range Standoff Weapon (LRSO)

$16B

Replaces: AGM-86B Air-Launched Cruise Missile

Timeline: 2030+

Nuclear-armed cruise missile. Launched from B-21 or B-52.

NC3 (Nuclear Command & Control)

$58B

Replaces: Cold War-era communications

Timeline: 2024-2035

Currently relies on 1960s-era technology. Floppy disks used until 2019.

22 Times We Almost Ended the World

Nuclear deterrence works — until it doesn't. At least 22 times since 1945, technical failures, miscommunication, or human error brought the world to the brink of nuclear war. We survived not because the system worked, but because individual humans — often low-ranking officers — chose not to follow protocol.

Nov 1961

BMEWS False Alarm

NORAD lost contact with all three BMEWS radar stations simultaneously. SAC went to highest alert. Cause: relay station failure in Colorado.

Oct 1962

Cuban Missile Crisis — Vasili Arkhipov

Soviet submarine B-59, depth-charged by US Navy. Two of three officers voted to launch a nuclear torpedo. Vasili Arkhipov was the lone dissent. One man saved civilization.

Nov 1979

NORAD Training Tape

Someone loaded a training tape showing a Soviet first strike into the live NORAD system. Zbigniew Brzezinski was moments from waking Carter to authorize retaliation.

Sep 1983

Stanislav Petrov

Soviet early warning system showed 5 US ICBM launches. Petrov's job was to confirm and trigger retaliation. He decided it was a false alarm. It was. Sunlight reflecting off clouds.

Nov 1983

Able Archer 83

NATO exercise simulated nuclear release procedures so realistically that the Soviets believed it was a real first strike. They put nuclear forces on alert. The world was minutes away.

Jan 1995

Norwegian Rocket Incident

Russia detected a Norwegian research rocket and believed it was a US submarine-launched missile. Yeltsin activated the nuclear briefcase. The first time a Russian president opened the "football."

The Global Nuclear Club: 12,500 Warheads, 9 Countries

CountryTotalDeployedAnnual BudgetTrend
Russia6,2571,674$8.6B (est)Modernizing. New Sarmat ICBM, Poseidon torpedo, Burevestnik cruise missile.
United States5,5001,770$51.5BFull triad replacement. $1.7T over 30 years.
China500~350$14B (est)Expanding rapidly. May reach 1,000+ by 2030. New silos in Xinjiang.
France290~280$5.6BMaintaining. New M51 SLBMs. Air-launched cruise missiles.
United Kingdom225~120$3.0BIncreasing cap from 180 to 260. New Dreadnought-class subs.
Pakistan170N/A$1.2B (est)Expanding. Tactical nuclear weapons. Nasr short-range missile.
India172N/A$2.3B (est)Expanding. Agni-V ICBM. Nuclear triad nearly complete.
Israel90 (est)N/AClassifiedNeither confirms nor denies. Dimona reactor. Jericho III missiles.
North Korea40-60N/A$500M (est)Testing. Hwasong-18 ICBM. Submarine capability under development.

The Poisoned Land: Environmental Cost of Nuclear Weapons

Nuclear weapons production contaminated vast swaths of the United States. The cleanup — which will take centuries and cost hundreds of billions — is the largest environmental remediation effort in human history. Some sites will never be fully clean.

Nevada Test Site

928 nuclear tests. Fallout across the western US. Downwinders developed cancer at 2-3x normal rates. $2.3B in compensation paid.

Cleanup: $1.2B spent, cleanup "indefinite"

Hanford, Washington

Produced plutonium for 60,000+ warheads. 56 million gallons of radioactive waste in underground tanks. Largest environmental cleanup in history.

Cleanup: $68B spent since 1989. Estimated $300-600B more needed.

Rocky Flats, Colorado

Plutonium trigger manufacturing plant 16 miles from Denver. Contaminated soil and groundwater. Fires in 1957 and 1969 released plutonium into the air.

Cleanup: $7B. Now a "wildlife refuge" where the DOE says it's safe. Locals disagree.

Marshall Islands

US tested 67 nuclear weapons (1946-58) including Castle Bravo (15 MT). Bikini Atoll still uninhabitable. Rongelap islanders exposed to lethal fallout.

Cleanup: $759M compensation. Estimated $2.3B needed. Bikini Atoll still has 6x safe cesium levels.

Savannah River Site, SC

Produced tritium and plutonium. 35 million gallons of radioactive waste. Groundwater contamination extending offsite.

Cleanup: $10B+ spent. Estimated $100B more over next 50 years.

What $1.7 Trillion Could Build Instead

End world hunger for 30 years

$340B

Less than 3 years of nuclear spending

Universal healthcare for all Americans

$1.5T/year

The 30-year modernization plan could fund it for 1+ year

Clean energy transition (entire US grid)

$1.2T

Less than the modernization plan

Cure for cancer (estimated R&D)

$150B

One submarine program

Free college for every American for 40 years

$1.6T

Roughly equal to the modernization plan

Rebuild all US infrastructure (ASCE estimate)

$2.6T

1.5x the modernization plan

The Question Nobody Asks

Deterrence requires enough nuclear weapons to guarantee retaliation. The question is: how many is enough? The US currently has 5,500 warheads. Experts estimate that 100-300 would be sufficient to deter any rational adversary. The difference between 300 and 5,500 is not security — it's bureaucracy, contractor profit, and Cold War inertia.

The $1.7 trillion modernization plan replaces every component of a triad designed to fight the Soviet Union. The Soviet Union no longer exists. China, the rising competitor, has 500 warheads. The US is spending $1.7 trillion to maintain a 10:1 advantage over an adversary with 500 weapons. Deterrence could be maintained for a fraction of the cost.

But nobody in Congress will say so, because voting against nuclear modernization is political suicide — even when the alternative is spending $1.7 trillion on weapons designed to end the world.

Sources

  • Brookings Institution, Atomic Audit: The Costs and Consequences of U.S. Nuclear Weapons Since 1940
  • Congressional Budget Office, “Projected Costs of U.S. Nuclear Forces, 2023 to 2032”
  • Federation of American Scientists, Nuclear Weapons Stockpile Data
  • Union of Concerned Scientists, “Close Calls with Nuclear Weapons”
  • SIPRI Yearbook 2024, World Nuclear Forces
  • GAO, Sentinel ICBM Nunn-McCurdy Breach Report (2024)
  • DoE/NNSA, Stockpile Stewardship and Management Plan
  • Eric Schlosser, Command and Control (Penguin, 2013)
  • Annie Jacobsen, Nuclear War: A Scenario (Dutton, 2024)
  • Daniel Ellsberg, The Doomsday Machine (Bloomsbury, 2017)