At the Sevmash shipyard in Severodvinsk, on Russia’s White Sea coast, a submarine unlike any other is preparing to enter the water for its first sea trials. The Khabarovsk, a purpose-built vessel designed from the keel up to carry Russia’s Poseidon nuclear-powered underwater drone, is expected to begin those trials in 2026, according to reporting from Naval News and analysis of commercial satellite imagery by open-source intelligence researchers. If the boat performs as intended, it will become the world’s first dedicated launch platform for a weapon that no existing arms-control treaty covers: an autonomous, nuclear-armed torpedo designed to detonate near enemy coastlines and blanket them in radioactive contamination.
The paper trail
The Poseidon program entered public view through what the Kremlin initially called an accident. During a November 2015 meeting between Vladimir Putin and senior military officials, cameras from the NTV television network captured a document describing a new “oceanic multi-purpose system.” The visible text outlined a weapon intended to create zones of extensive radioactive contamination along enemy coasts, rendering ports and naval bases unusable. Western analysts, including researchers at the Federation of American Scientists, noted at the time that the “leak” appeared suspiciously well-framed, suggesting it may have been a deliberate signal to NATO governments about Russia’s expanding strategic toolkit.
Putin removed any remaining ambiguity during his March 2018 state-of-the-nation address. In a presentation that also unveiled hypersonic missiles and a nuclear-powered cruise missile called Burevestnik, he described an autonomous underwater vehicle with intercontinental range, powered by a compact nuclear reactor, capable of carrying nuclear or conventional payloads at extreme depth. Animated graphics showed the drone gliding beneath the ocean and detonating near a coastline. The message was blunt: Russia was building a weapon that could bypass every missile-defense system the United States had deployed or planned.
The most recent official milestone came when Putin stated publicly that the Poseidon had been tested successfully on nuclear power. No independent body has verified that claim, but it represents the strongest official confirmation that the drone’s miniaturized reactor has reached a functional stage. Satellite imagery reviewed by analysts at the James Martin Center for Nonproliferation Studies has shown activity at Russian naval facilities consistent with handling large experimental undersea systems, including specialized piers and support vessels at Severodvinsk and the Olenya Guba submarine base on the Kola Peninsula.
The submarine: Khabarovsk vs. Belgorod
Russia has two submarines linked to the Poseidon program, and understanding the difference between them matters. The Belgorod, commissioned in 2022, is a converted Oscar II-class guided-missile submarine. At roughly 184 meters long, it is the largest submarine in the world and has served as the initial test platform for Poseidon prototypes. But the Belgorod is a retrofit. Its cavernous hull was adapted from a Cold War-era design originally built to carry anti-ship cruise missiles, not autonomous drones.
The Khabarovsk is different. It was designed from scratch as a Poseidon carrier. Smaller than the Belgorod, it reportedly features dedicated launch tubes sized specifically for the drone, a layout that should allow faster loading, more reliable deployment, and the ability to carry multiple units on a single patrol. According to assessments published by the U.S. Congressional Research Service, the Khabarovsk-class is expected to form the backbone of Russia’s Poseidon deployment capability, with multiple hulls potentially planned.
Sea trials are the critical gate between construction and operational service. During this phase, engineers test the hull’s integrity under pressure, validate propulsion and power systems, and confirm that weapons-integration systems function in real ocean conditions. For a submarine carrying nuclear-armed drones, those trials also involve verifying that the launch tubes, handling equipment, and fire-control systems work together safely. If the Khabarovsk passes, Russia would have a credible, dedicated platform ready to carry Poseidon on operational patrols.
What the weapon is supposed to do
The Poseidon, designated “Status-6” in early Russian planning documents and later given the NATO reporting name “Kanyon,” is not a torpedo in the traditional sense. Conventional torpedoes are launched from submarines at ranges of tens of kilometers and travel at speeds between 40 and 60 knots. They are designed to sink ships. The Poseidon is designed to destroy cities.
According to the 2015 document and Putin’s subsequent descriptions, the drone is propelled by a compact nuclear reactor, giving it effectively unlimited range. Russian state media has claimed it can travel at speeds exceeding 100 knots and operate at depths beyond 1,000 meters. Those figures deserve skepticism. No known underwater vehicle has sustained 100 knots for extended periods; the fastest confirmed torpedo, Russia’s own VA-111 Shkval, uses a supercavitation rocket motor to reach roughly 200 knots but only over very short distances and at shallow depths. A large, reactor-powered drone operating at 1,000 meters would face enormous hydrodynamic and structural challenges. Independent naval engineers, including analysts at the Royal United Services Institute in London, have suggested that a more realistic sustained speed might be in the range of 50 to 70 knots, still fast enough to outrun most anti-submarine torpedoes.
The warhead is the most alarming and least understood component. The 2015 document described a weapon intended to contaminate coastal zones, and some Western analysts have speculated that the design could incorporate a cobalt-salted nuclear device, a theoretical weapon that wraps a nuclear core in cobalt-59, which transmutes into cobalt-60 upon detonation and produces intense, long-lasting gamma radiation. Such a device has never been built or tested by any known state. Other analysts, including those at the Federation of American Scientists, have argued that even a standard thermonuclear warhead detonated underwater near a coast would generate a massive radioactive tsunami, dispersing irradiated seawater and sediment across a wide area. The practical difference between a “salted” bomb and a large underwater nuclear blast may matter less than the shared outcome: a coastal zone rendered dangerous for years.
Why it sits outside arms control
Every major nuclear arms-control agreement of the past half-century, from SALT to START, was built around a specific category of weapon: intercontinental ballistic missiles, submarine-launched ballistic missiles, heavy bombers, and their associated warheads. The Poseidon fits none of these categories. It is not a ballistic missile. It is not launched from a tube in the traditional SLBM sense. It is an autonomous, nuclear-powered drone that travels through the ocean under its own propulsion, potentially for thousands of kilometers, before detonating.
This means that even if the United States and Russia were to negotiate a successor to New START, which expired in February 2026, the Poseidon would not automatically fall under its terms unless both sides agreed to include it. Arms-control specialists at the Carnegie Endowment for International Peace have warned that weapons like Poseidon and the Burevestnik cruise missile represent a new class of “exotic” delivery systems that could undermine the entire framework of strategic stability. If one side deploys weapons the other cannot count, verify, or constrain through treaty mechanisms, the incentive to negotiate shrinks and the incentive to build competing systems grows.
The United States has not publicly announced a direct counterpart to the Poseidon, but the Pentagon has significantly increased investment in undersea surveillance. The Deep Reliable Acoustic Path (DRAP) sensor network, upgrades to the Sound Surveillance System (SOSUS) inherited from the Cold War, and new autonomous undersea vehicles designed for anti-submarine warfare all reflect a recognition that the deep ocean is becoming a contested domain in ways it has not been since the 1980s.
What remains genuinely unknown
For all the attention the Poseidon has received, the list of things that cannot be independently confirmed is long. No Russian Ministry of Defense document has been released publicly detailing the Khabarovsk’s sea-trial schedule, its exact launch-tube configuration, or the timeline for integrating live Poseidon units. Western intelligence assessments have surfaced in media reports, but the underlying classified data is unavailable for public review.
The drone’s actual performance envelope is similarly opaque. Putin’s public remarks confirm that a nuclear-propulsion test took place, but they do not specify range, top speed, maximum operating depth, or warhead yield with any precision. Until those details are supported by verifiable data, whether from observed tests, treaty inspections, or credible leaks with corroborating evidence, they remain claims rather than demonstrated capabilities.
Construction status is difficult to pin down. Satellite imagery of the Sevmash shipyard has shown the Khabarovsk hull at various stages, but as of mid-2026, no independent analyst has published confirmed imagery of the submarine in a configuration clearly ready for open-water trials. Reports of the boat’s transfer to outfitting piers have relied on anonymous sources or low-resolution commercial satellite photos. The gap between Russian official statements and verifiable physical evidence remains wide.
That gap is itself strategically useful for Moscow. By keeping the program partially visible but never fully transparent, Russian officials can suggest game-changing capabilities without bearing the full diplomatic cost of openly testing or deploying them. The ambiguity forces Western defense planners to prepare for a worst-case scenario, diverting resources and attention, while giving the Kremlin plausible deniability if the technology falls short of its advertised potential.
How to read what comes next
For anyone following this story, the most reliable framework is to separate what has been shown from what has been said from what has been inferred. The 2015 broadcast document is primary evidence: it shows an actual classified paper, not a secondhand description. Putin’s speeches are official policy declarations, attributable and on the record, even if their technical claims cannot be independently verified. Everything else, the speed figures, the depth ratings, the cobalt-bomb speculation, the precise construction timelines, rests on a thinner foundation of satellite analysis, unnamed intelligence sources, and defense-industry leaks. These are useful for tracking the program’s trajectory, but they should be read as informed estimates, not confirmed facts.
The next concrete signals to watch for are straightforward. A visible launch at Sevmash, publicized port calls, or further presidential statements would strengthen the case that Poseidon is nearing operational status. Observed test firings in the Barents Sea or Arctic waters, detectable by hydroacoustic monitoring stations operated under the Comprehensive Nuclear-Test-Ban Treaty Organization, would be harder to dispute. Continued silence, by contrast, would leave analysts parsing satellite shadows and extrapolating from incomplete data.
What is not in doubt is the intent. The documents, the speeches, the submarine construction, and the investment of billions of rubles all point in the same direction: Russia is building a weapon designed to hold coastal populations at risk of nuclear annihilation in a way that no existing defense system can reliably prevent. Whether the Poseidon works as advertised or falls short, the strategic calculation behind it is clear. And with the Khabarovsk moving toward the water, the window for addressing that calculation through diplomacy rather than deterrence is narrowing.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
