The CIA secretly built a ship to raise a sunken Soviet submarine from the Pacific floor
By Colin PetersonJuly 8, 2026
Where the evidence lands: Proven
That the CIA, under the codename Project Azorian, spent roughly $800 million building a one-of-a-kind deep-sea recovery ship disguised as a commercial mining vessel, sailed it to the wreck of the Soviet submarine K-129 in the summer of 1974, and used a giant mechanical claw to try to raise the entire submarine (nuclear missiles, cryptographic material, and crew remains included) from a depth of roughly 16,000 feet, all without the Soviet Union ever learning what the ship actually was.
Proven. Declassified by the CIA itself in 2010: an internal history confirms the agency built the Hughes Glomar Explorer under a fake Howard Hughes mining cover to recover the Soviet submarine K-129, though how much of the sub's code material and remains actually came up is still only partly public.
Believed by: Widely accepted as historical fact once the CIA confirmed it
The full story
A submarine nobody could find, except the Americans
In March 1968, a Soviet Golf-II class submarine designated K-129 left its base on the Kamchatka Peninsula on a routine ballistic-missile patrol and never returned. It carried nuclear-armed missiles and a crew of 98 men. The Soviet Pacific Fleet searched for weeks, in growing secrecy and growing panic, and found nothing. What the Soviets did not know was that the United States Navy's classified undersea listening network, SOSUS, built to track exactly this kind of submarine, had recorded an acoustic signature consistent with a catastrophic failure at sea. Cross-referencing recordings from multiple listening stations, US naval intelligence worked out a probable location the Soviets never managed to pinpoint themselves.
To confirm it, the Navy sent a submarine, the USS Halibut, converted for deep ocean intelligence work, to photograph the wreck directly. Over the course of 1968 it took more than twenty thousand close-up images of K-129, resting broken on the Pacific floor roughly 16,000 feet down, nearly three miles below the surface, in water so deep and dark that no recovery of this kind had ever been attempted. The pictures were detailed enough that analysts believed at least one missile compartment might have survived the sinking intact, along with code equipment and other material of enormous intelligence value. That assessment is what turned a quiet naval curiosity into a presidential-level covert operation.
What the CIA's own declassified history confirms
Take the case at its strongest, because with Project Azorian the strongest case is also the officially confirmed one. On 12 February 2010, the National Security Archive published a fifty-page CIA internal history, Project Azorian: The Story of the Hughes Glomar Explorer, that the agency had just declassified in response to a Freedom of Information Act request filed by researcher Matthew Aid in 2007. The document had originally been written in 1985 (more than a decade after the operation, while the program was still fully classified) for the CIA's in-house journal, Studies in Intelligence, by a participant in the operation whose identity remains classified to this day. This was not a leak, a rumor, or a foreign intelligence claim. It was the CIA describing its own operation, in its own words, decades after the fact.
The declassified history lays out the mechanics in detail: a Special Projects Staff stood up within the CIA's Directorate of Science and Technology to run the program, a contract with Global Marine Development to design a one-of-a-kind ship at Sun Shipbuilding in Pennsylvania, and (the detail that made the whole scheme work) a cover story built around the reclusive billionaire Howard Hughes. His Summa Corporation publicly fronted the new vessel, the Hughes Glomar Explorer, as a private venture to mine manganese nodules from the deep ocean floor, a real, if speculative, area of commercial interest at the time, which is exactly why the cover held. Hughes's well-known reclusiveness meant no one expected to see him personally confirm or deny anything about his own company's ship, an arrangement later accounts, including the CIA's own declassified history, describe as one of the most effective covers the agency built during the Cold War.
The recovery attempt itself is equally well documented. After President Nixon approved the mission in 1970 and the ship was completed in 1973, the Glomar Explorer sailed to the wreck site in the summer of 1974 and, in early August, lowered a massive purpose-built capture vehicle (nicknamed Clementine by the crew) on a pipe string nearly three miles long to grip the submarine and lift it toward the ship's hidden internal moon pool. That a civilian-flagged ship spent over a month conducting deep-ocean salvage work directly above a sunken Soviet military vessel, without the Soviet Navy ever mounting a challenge, is itself a remarkable feat of Cold War tradecraft that the CIA's own history confirms happened essentially as planned.
The claw that broke, and the Glomar response it left behind
The recovery did not go entirely as hoped, and the CIA's own account does not pretend otherwise. During the lift, a large section of K-129's hull, structurally weakened by six years underwater and the pressure of the recovery effort, failed and fell back to the ocean floor. Only the forward 38 feet of the bow was successfully brought inside the ship. Along with it came the remains of six Soviet crewmen, given a formal military burial at sea aboard the Glomar Explorer itself on 4 September 1974, sealed in a metal casket, reportedly out of concern about radioactivity, with prayers, flags, and national anthems for both countries. A CIA camera crew filmed the ceremony, and that roughly fourteen-minute burial reel is the only part of the recovery footage the agency has ever made public, and even that took until 1992, when it quietly handed a portion to the Russian government as a gesture of goodwill after the Cold War had ended.
The secret's first crack came from an ordinary crime, not an intelligence leak. On 5 June 1974, burglars broke into the Los Angeles offices of Hughes's Summa Corporation and made off with cash and internal papers, at least one of which tied Hughes to the CIA's secret submarine-recovery operation, and reportedly tried to sell that connection back to the Hughes organization rather than to a newspaper. Rumors circulated among reporters through the rest of 1974, and on 7 February 1975 the Los Angeles Times published the first account tying the burglary, Hughes, and a CIA salvage operation together. CIA Director William Colby personally asked several news organizations to hold the fuller story on national-security grounds, and several did, until columnist Jack Anderson named the ship on his national radio broadcast on 18 March 1975, after which Seymour Hersh published the fullest account the next day on the front page of The New York Times, under the headline “C.I.A. Salvage Ship Brought Up Part of Soviet Sub Lost in 1968, Failed to Raise Atom Missiles.” Once the story was public, journalist Harriet Ann Phillippi filed a FOIA request seeking any records of the CIA's attempts to suppress the press coverage itself. The agency's answer was neither yes nor no: it would not confirm or deny that such records even existed. When Phillippi sued, the US Court of Appeals for the D.C. Circuit upheld that answer in Phillippi v. CIA in November 1976, reasoning that in narrow cases the very existence of a record can itself be a properly classified fact.
“When the Agency's position is that it can neither confirm nor deny the existence of the requested records, there are no relevant documents for the court to examine other than the affidavits which explain the Agency's refusal.”
That ruling is the direct origin of the term still used across FOIA law today: the “Glomar response.” Every time a federal agency responds to a records request with a flat refusal to confirm or deny, used constantly in national-security and surveillance cases that have nothing to do with submarines or the Pacific Ocean, it is invoking a legal doctrine that traces back, by name, to this one operation.
What is still classified, and what is genuinely disputed
Here is the honest complication, and it is a real gap in the declassified record rather than a manufactured one: exactly what came up inside that recovered, 38-foot bow section is still not fully public, and accounts that treat every detail as settled are outrunning the primary sources. The CIA's own 2010-released history is heavily redacted in places, and even independent researchers disagree on specifics. The most thorough outside account, Norman Polmar and Michael White's 2010 book Project Azorian, draws on participant interviews and on-board video to conclude that two nuclear-armed torpedoes and some cryptographic material came aboard along with the crew remains, while the missile compartment, control room, and most of the submarine's codebooks and decoding equipment were lost when the hull broke apart and fell back to the seafloor. A handful of other researchers have gone considerably further, arguing the failed lift was itself a cover story and that far more of the submarine (potentially the missiles and cryptographic material themselves) was quietly recovered; that version has no support in the declassified record or in the eyewitness accounts behind Polmar and White's book, but the real redactions in the official history are part of why the claim hasn't gone away. Even the wreck's depth and the submarine's exact sinking date are given slightly differently across otherwise reliable sources, a reminder that “declassified” here means partially declassified, not fully closed.
The full operational film of the recovery, as opposed to the burial-at-sea footage released in 1992, has never been made public. Nor has a complete, itemized inventory of what the CIA actually brought back, beyond the general acknowledgment that the haul was smaller than hoped. A planned second recovery attempt at the site, intended to retrieve the section that fell back to the seafloor, was abandoned after the 1975 press disclosures made further secret operations there impossible, meaning the part of the submarine that broke away in 1974 was, by design, never retrieved at all and presumably remains on the ocean floor to this day. One tangible piece of the recovered hardware is on public view: a submarine control panel from the operation is displayed at the International Spy Museum in Washington, D.C.
It is also worth being precise about what is not in serious dispute: that the operation happened, that the CIA built and ran it, that Howard Hughes was a knowing but non-operational cover, and that a partial recovery took place in August 1974. Those facts come from the agency's own declassified account and are not contested by independent historians. What remains open is the finer-grained inventory of intelligence gained: a gap that is honestly acknowledged by the same declassified history that confirms everything else, rather than a hole conspiracy theorists have had to invent.
Why this confirmed story still fascinates
Project Azorian occupies an unusual place on a site built around evaluating unproven claims, because here the extraordinary claim turned out to be true, confirmed by the very government agency that ran it. Part of why it still fascinates is scale: an $800-million ship (an extraordinary sum for a single covert operation even today) built from scratch, disguised as a mining venture convincing enough that actual mining companies took the manganese-nodule premise seriously, all to reach three miles down and grab hold of a single sunken submarine before a rival superpower noticed. Cold War history is full of claims that strain credulity; this is one of the few where the declassified paper trail says the strain was justified.
The other reason it endures is the redactions themselves. Because the CIA's own account stops short of a complete inventory, and because the full operational film has never surfaced, the story leaves just enough of a documented gap for reasonable speculation to live in: did the mission also serve as cover for other undersea intelligence work already underway in the Pacific during the same years, such as cable tapping operations the US ran elsewhere against Soviet communications? The CIA's declassified Azorian history does not address that question either way, which is a genuinely different situation from a conspiracy theory built on silence: here, official silence coexists with an official, on-the-record confirmation of the core operation itself.
There is also a quieter, more human thread running under the tradecraft: the CIA camera crew that filmed a formal military burial at sea for the Soviet sailors recovered in that bow section, and the decision, made across a Cold War divide, and not required by any treaty, to hand that footage to Russia in 1992. It is a small gesture inside an enormous act of espionage, and it is one of the few parts of the story that both governments have, in their own ways, quietly acknowledged.
A real recovery, a myth around it
On the core claim, that the CIA secretly built and operated the Hughes Glomar Explorer under a fabricated Howard Hughes mining cover to recover a sunken Soviet ballistic-missile submarine from the Pacific floor, the rating is Proven. This is not an inference built from suspicious circumstances or a journalist's leak alone. It is the conclusion of the CIA's own declassified internal history, released in 2010, corroborated by a federal appellate court record from 1976 that exists precisely because of this operation, and by three decades of independent reporting and research that lines up with the official account on every major point.
What remains genuinely open is not whether the operation happened, or whether Howard Hughes was a knowing cover rather than a participant, but the finer detail of what the recovered bow section actually contained, and what became of the intelligence gained from it, questions the CIA's own partial declassification leaves honestly unresolved rather than conclusively answered. Take the primary record on its own terms and the story survives fully intact either way: a nearly three-mile-deep recovery attempt, built on one of the most elaborate cover stories of the Cold War, that succeeded partially, leaked publicly, and left behind a legal doctrine, the Glomar response, still invoked in FOIA disputes that have nothing to do with submarines at all.
What's still unexplained
- Exactly what came up inside the recovered bow section (and what intelligence the United States actually drew from it) is still not fully public. The CIA's own 2010-declassified history is redacted at close to the points where a complete inventory would appear, and outside accounts disagree with each other: participant interviews and CIA-shot video used in Norman Polmar and Michael White's 2010 book describe two nuclear-tipped torpedoes and some cryptographic material coming aboard, while other credible summaries of the same events describe cryptographic material as among what was lost. More than fifty years on, no complete public accounting exists either way.
- Which section of K-129 was actually recovered and which fell back to the seafloor is itself disputed in the details. The mainstream account (the CIA's own history and the eyewitness testimony behind Polmar and White's book) holds that only the forward 38 feet of the hull, holding the torpedo room, came aboard, while the missile compartment and control spaces most likely to hold codebooks and targeting data broke away and were lost. A minority of researchers have argued closer to the opposite: that the valuable compartment was in fact recovered and the public failure was itself a cover story. Nothing in the declassified record settles this, and a planned second attempt that might have resolved the question was abandoned once the 1975 press leaks made further secret work at the site impossible.
- The CIA has confirmed that six Soviet crewmen's remains were recovered and given a formal, filmed military burial at sea on 4 September 1974, sealed in a metal casket reportedly because of radioactivity concerns, with prayers and honors conducted in both English and Russian. Some published accounts identify three of the six by name; the agency has never released a complete roster, medical findings, or the full recovery footage (only a portion of the burial film was given to Russia in 1992), so exactly how the men were identified, and what happened to the rest of the film, remains outside the public record.
- A second recovery attempt aimed at the section that broke away and fell back to the seafloor was reportedly planned before the 1975 press leaks made further secret operations at the site impossible. Whether that fallen section, potentially still holding the submarine's missiles and codebooks, has ever been quietly revisited by any nation, or simply remains where it fell in 1974, has never been addressed in any declassified document.
Point by point
The claim: The CIA built the Glomar Explorer to secretly recover a sunken Soviet submarine.
What the record shows: Confirmed by the CIA's own declassified internal history. The 2010-released Studies in Intelligence article lays out the agency's Special Projects Staff, its contracting of Global Marine Development, and the ship's true purpose: intelligence recovery, not commercial mining. This is the US government's own account of its own operation, not an outside allegation.
The claim: Howard Hughes was a real front for a fake mining operation, not an actual participant.
What the record shows: Confirmed. Multiple declassified and journalistic accounts agree the reclusive billionaire's Summa Corporation lent its name and public reputation for eccentric, secretive ventures to make a manganese-nodule mining cover story plausible; Hughes had no operational role, and the ruse was convincing enough that other mining companies reportedly explored their own seabed-nodule ventures in response.
The claim: The recovery operation actually happened, at the depth and on the date claimed.
What the record shows: Confirmed by the CIA history and corroborated by independent researchers and journalists: the Glomar Explorer sailed to the site in the summer of 1974 and lifted a section of the K-129 from a wreck roughly three miles down in early August 1974, using a purpose-built capture vehicle lowered on a pipe string.
The claim: The entire submarine, including its nuclear missiles and code materials, was successfully raised intact.
What the record shows: Not established, and this is the honest limit of the case. The CIA's own history and independent accounts agree that a large section of the submarine broke apart and fell back to the seafloor during the lift; only a smaller forward portion was brought aboard. Exactly what was inside that recovered section, beyond the remains of several crewmen, is described inconsistently across sources and remains only partly declassified.
The claim: The government's 'we can neither confirm nor deny' non-answer began with this operation.
What the record shows: Confirmed by the federal court record itself. When journalist Harriet Ann Phillippi sought records of CIA efforts to suppress press coverage of the story, the agency's refusal to confirm or deny the existence of any such records was upheld by the D.C. Circuit in Phillippi v. CIA (1976): the ruling that gave the phrase 'Glomar response' its name, still cited today in FOIA law.
The claim: The Soviet Union never learned what actually happened until after the fact.
What the record shows: Substantially confirmed but not perfectly certain. Declassified US records show real anxiety inside the Ford administration about a Soviet reaction after the 1975 leaks, and no formal Soviet protest followed publicly, but exactly what Moscow's intelligence services privately concluded, and when, has never been fully documented on the public record from the Soviet or Russian side.
Timeline
- 1968-03The Soviet Golf-II class ballistic-missile submarine K-129 is lost with all hands in the Pacific, roughly 1,600 miles northwest of Hawaii, after leaving its base on the Kamchatka Peninsula. The Soviet Navy searches for weeks and cannot find it.
- 1968The US Navy's classified SOSUS hydrophone network, built to track Soviet submarines, has recorded an acoustic event consistent with an implosion; naval intelligence uses multiple listening stations to triangulate a probable wreck location the Soviets never find.
- 1968-07The submarine USS Halibut, converted for deep-sea intelligence work, locates and photographs the K-129 wreck (over 20,000 close-up images) lying broken on the seafloor at a depth of roughly 16,000 feet.
- 1970Defense Secretary Melvin Laird and National Security Advisor Henry Kissinger bring a recovery proposal to President Nixon, who approves a CIA-run effort to raise part or all of the submarine.
- 1971The CIA's Special Projects Staff contracts Global Marine Development to design and build a purpose-built recovery ship at Sun Shipbuilding in Chester, Pennsylvania, under the cover that it is a private commercial vessel.
- 1972–1974The agency arranges for reclusive billionaire Howard Hughes and his Summa Corporation to publicly front the ship (named the Hughes Glomar Explorer) as a deep-sea mining venture harvesting manganese nodules from the ocean floor, a cover story real enough that mining firms took it seriously for years.
- 1974-06/07The Glomar Explorer sails from Long Beach, California, to the recovery site, arriving in early July after President Ford (having succeeded Nixon in August 1974) and his administration continue backing the mission.
- 1974-08Over roughly a week in early August 1974, the ship lowers a massive mechanical claw nearly three miles to the seafloor and begins lifting a section of K-129 toward the surface.
- 1974-08-08During the lift, a large portion of the submarine's structure fails and falls back to the ocean floor. A smaller forward section is successfully brought inside the ship's hull.
- 1975-02/03The Los Angeles Times first publishes a partial account of the secret operation in February; after CIA Director William Colby persuades several outlets to hold the story, columnist Jack Anderson breaks it open in March and other outlets follow.
- 1975Journalist Harriet Ann Phillippi files a Freedom of Information Act request seeking records of the CIA's attempts to suppress press coverage of the operation; the CIA refuses to confirm or deny that such records even exist.
- 1976-11In Phillippi v. CIA, the US Court of Appeals for the D.C. Circuit rules the CIA may lawfully refuse to confirm or deny the existence of the requested records, the ruling that gives rise to the term 'Glomar response.'
- 1992The CIA releases footage of the burial-at-sea ceremony for the Soviet crewmen recovered aboard the Glomar Explorer to the Russian government, a quiet diplomatic gesture two decades after the operation.
- 2010-02-12After a 2007 FOIA request from researcher Matthew Aid, the CIA declassifies a 50-page internal history of the operation, 'Project Azorian: The Story of the Hughes Glomar Explorer,' originally written for its in-house journal Studies in Intelligence; the National Security Archive publishes it.
From the case file
The actual records: declassified, released, or leaked. We link straight to each document in its official archive, so you never have to take our word for it. Read the originals yourself.
Project Azorian: The Story of the Hughes Glomar Explorer
The CIA's own internal history of the operation, written for its classified in-house journal and declassified in 2010: the government's first official confirmation that it built the Glomar Explorer to recover the Soviet submarine K-129.
Read the document: CIA Reading Room →Project Azorian: The CIA's Declassified History of the Glomar Explorer (Electronic Briefing Book No. 305)
The National Security Archive's compiled collection of the declassified Azorian history and related records, with editorial context on what the CIA released and what remains redacted.
Read the document: National Security Archive →Foreign Relations of the United States, 1969–1976, Vol. XXXV: record on the K-129 recovery (Glomar Explorer)
A formerly Top Secret record in the State Department's official documentary history covering high-level deliberations over the Glomar Explorer operation and the handling of its public disclosure.
Read the document: State Dept. Office of the Historian →Other case files that cite the same sources
Proven. Declassified by the CIA itself in 2010: an internal history confirms the agency built the Hughes Glomar Explorer under a fake Howard Hughes mining cover to recover the Soviet submarine K-129, though how much of the sub's code material and remains actually came up is still only partly public.
Reviewed by Colin Peterson · Last reviewed July 12, 2026 · How we rate
Common questions
Is Project Azorian true?
Proven. Declassified by the CIA itself in 2010: an internal history confirms the agency built the Hughes Glomar Explorer under a fake Howard Hughes mining cover to recover the Soviet submarine K-129, though how much of the sub's code material and remains actually came up is still only partly public.
What is Project Azorian?
Not a rumor but a declassified CIA operation: in the early 1970s, the agency secretly built a purpose-designed ship, the Hughes Glomar Explorer, to lift a sunken Soviet ballistic-missile submarine from nearly three miles down in the Pacific, all under the cover story that reclusive billionaire Howard Hughes was mining seabed minerals. Le…
What does the evidence show?
Confirmed by the CIA's own declassified internal history. The 2010-released Studies in Intelligence article lays out the agency's Special Projects Staff, its contracting of Global Marine Development, and the ship's true purpose: intelligence recovery, not commercial mining. This is the US government's own account of its own operation, no…
Why do people believe it?
This is one of the rare cases on this site where 'the CIA secretly did something this audacious' is not a theory but a matter the agency has confirmed in its own declassified history, so the pull here is less about doubting the official story and more about marveling that a true story this strange was kept secret for years.
What is still unresolved?
Exactly what came up inside the recovered bow section (and what intelligence the United States actually drew from it) is still not fully public. The CIA's own 2010-declassified history is redacted at close to the points where a complete inventory would appear, and outside accounts disagree with each other: participant interviews and CIA-…
Sources
- 1.Project Azorian: The Story of the Hughes Glomar Explorer, Central Intelligence Agency, Studies in Intelligence (declassified 2010) (2010)
- 2.Project Azorian: The CIA's Declassified History of the Glomar Explorer (Electronic Briefing Book No. 305), National Security Archive, George Washington University (2010)
- 3.Phillippi v. Central Intelligence Agency, 546 F.2d 1009 (D.C. Cir. 1976), U.S. Court of Appeals for the District of Columbia Circuit (1976)
- 4.Project Jennifer / Hughes Glomar Explorer (declassified document collection), Federation of American Scientists, Intelligence Resource Program (1998)
- 5.Glomar Explorer, K-129 recovery: Foreign Relations of the United States historical documentation, Office of the Historian, U.S. Department of State (1974)
- 6.Project Azorian: The CIA and the Raising of K-129, Norman Polmar and Michael White, Naval Institute Press (2010)
- 7.During the Cold War, the CIA Secretly Plucked a Soviet Submarine From the Ocean Floor Using a Giant Claw, Smithsonian Magazine (2019)
- 8.Document Friday: The Origins of Glomar Declassified, UNREDACTED: The National Security Archive Blog, George Washington University (2012)
Embed this verdict on your site
Paste this snippet to show our sourced verdict as a small card, with a link back to the full case file. Free to use.
<iframe src="https://theconspiratory.com/embed/project-azorian" title="The Conspiratory verdict" width="520" height="190" style="border:0;max-width:100%" loading="lazy"></iframe>Help us investigate
This is a living case file. If you spot an error or know evidence we missed, tell us, and weigh in on where you land.
Where do you land?
Cast your read on this one.
Comments
Add your take. Comments are read and approved by a human before they appear, so keep it on topic and civil. Please do not accuse named, living people of crimes.