Publications
The Poseidon Effect: How Nuclear Drones Are Redrawing South Asia’s Deterrence Architecture
Tahir Azad
Policy Brief. 18 April 2026
ABSTRACT
The emergence of nuclear-powered and nuclear-armed autonomous systems marks a fundamental inflection point in global deterrence theory. Russia’s successful test of the Poseidon autonomous nuclear torpedo in October 2025, combined with the unprecedented drone warfare witnessed during the May 2025 India-Pakistan crisis, reveals a strategic environment in which the traditional nuclear triad is being supplemented and potentially supplanted by a fourth dimension of autonomous nuclear delivery. This paper examines the strategic logic driving this technological development, analyses the role of conventional drones in the May 2025 crisis as a harbinger of more complex future scenarios, and critically assesses the implications for South Asian strategic stability. It argues that India and Pakistan, already locked in an asymmetric deterrence competition, face a qualitatively new set of escalation risks in the absence of any bilateral or multilateral governance framework for autonomous nuclear systems.
Keywords: nuclear drones, Poseidon, South Asian deterrence, India-Pakistan, autonomous weapons, strategic stability
Introduction
On 29 October 2025, Russian President Vladimir Putin announced the successful test of the Poseidon autonomous nuclear torpedo, describing a system that combines nuclear propulsion with a multi-megaton warhead capable of crossing entire oceans undetected.[i] The timing of the announcement, days after the successful test of the Burevestnik nuclear-powered cruise missile, was calibrated: a deliberate act of strategic signalling to Washington during a period of ongoing negotiations over Ukraine and the broader Euro-Atlantic security architecture. Yet the implications of what Putin described as a weapon with “no means of interception”[ii] extend well beyond the Euro-Atlantic theatre. For strategists in New Delhi and Islamabad, the Poseidon test and the global proliferation of autonomous weapons systems raise a question that neither capital has yet adequately confronted: what happens to South Asia’s already fragile deterrence equilibrium when nuclear delivery shifts from ballistic missiles to autonomous, long-endurance, difficult-to-attribute unmanned systems?
The question is not merely theoretical. Less than six months before the Poseidon announcement, India and Pakistan fought the most severe military engagement between the two nuclear-armed states since the Kargil conflict of 1999, one in which conventional drones occupied a central and unprecedented role.[iii] Operation Sindoor and the Pakistani counter-offensive under Operation Bunyan-un-Marsoos constituted, in the assessment of the Stimson Centre, the first drone battle between the two nuclear-armed nations.[iv] Strikes approached within kilometres of Pakistan’s Strategic Plans Division in Rawalpindi, nuclear signalling was ambiguous, and the United States tracked indicators of Pakistani nuclear readiness.[v]The crisis resolved through a ceasefire on 10 May 2025, brokered under sustained American diplomatic pressure, but it left both doctrines unsettled and both arsenals on high alert. The convergence of these two developments, the Poseidon-type autonomous nuclear system entering operational testing at the global level, and the demonstrated centrality of autonomous platforms in South Asian crisis escalation, demands serious analytical attention. This paper examines the strategic logic of nuclear drone development, its fourth-dimensional implications for deterrence theory, and what it portends for a region in which the margin for miscalculation is already perilously thin.
Russia’s Poseidon: Strategic Requirements and Nuclear Culture
To understand the Poseidon programme, one must situate it within the specific strategic logic and nuclear culture of the Russian Federation. Russia’s “super-weapon” programmes, of which Poseidon is one of six unveiled in Putin’s March 2018 address to the Federal Assembly, are fundamentally a response to the collapse of the Anti-Ballistic Missile Treaty architecture. The United States’ unilateral withdrawal from the ABM Treaty in 2001 removed the last formal constraint on American missile defence expansion, creating what Moscow assessed as a prospective first-strike environment in which American interceptors could degrade a retaliatory response.[vi] Poseidon, a nuclear-powered autonomous underwater vehicle assessed at approximately twenty metres in length and capable of operating at depths exceeding one thousand metres, is designed to render such defences irrelevant by approaching from below and from oceanic distances at which interception is not feasible.[vii]
Critically, Poseidon falls entirely outside existing arms control verification regimes. It is not covered by New START, which expired in February 2026, nor by any successor framework. Moscow has explicitly stated that it will not provide advance notification of Poseidon test launches.[viii] This deliberate exclusion from arms control accountability reflects what scholars of Russian strategic culture identify as a consistent preference for asymmetric capability development, maintaining escalation dominance through technological domains where Western superiority cannot be leveraged. Lawrence Freedman’s foundational observation that deterrence is ultimately a function of perceived costs and credibility applies with particular force here: Poseidon’s deterrent value derives not from transparency but from the certainty of catastrophic retaliation even under the most adverse conditions.[ix]
The system is also, by design, a weapon of compellence as much as deterrence. Russian state media has amplified the narrative of a “tsunami weapon”, capable of detonating a multi-megaton warhead near coastal population centres to generate radioactive tidal surges.[x] Whether or not this is operationally realistic, the psychological dimension serves a signalling function that augments its military value. This dual character, simultaneously a survivable second-strike system and a coercive instrument of nuclear statecraft, is precisely why other states will study its implications carefully. The strategic lesson Moscow seeks to transmit is that the deterrence architecture of the twenty-first century belongs to those who dominate autonomous undersea and aerial domains, not only to those who possess the most accurate ballistic missiles.
The May 2025 Crisis: Drones as Precursor
The May 2025 India-Pakistan crisis was, in fundamental ways, a laboratory for the kind of strategic environment in which nuclear drone development will eventually be contested. It began on 22 April 2025 with the killing of twenty-six tourists near Pahalgam in Indian-administered Kashmir, an attack attributed by New Delhi to Pakistan-based militant groups.[xi]India’s response, Operation Sindoor launched on the night of 6-7 May, consisted of missile and drone strikes on nine targets across Pakistan and Pakistan-administered Azad Kashmir, including infrastructure associated with Jaish-e-Mohammed and Lashkar-e-Taiba.[xii] What followed over four days was a conflict unlike any previous India-Pakistan military engagement: primarily conducted through standoff platforms, with no manned aircraft crossing the international border, and with drones occupying a central operational role on both sides.
India deployed a range of systems, including Israeli-manufactured IAI Heron and Harop platforms alongside indigenous systems such as the Nagastra-1 loitering munition and the Archer-NG armed tactical drone.[xiii] Pakistan countered with its own drone and missile assets, launching strikes on Indian air defence infrastructure across fifteen locations.[xiv]Critically, Indian strikes on 10 May targeted the Nur Khan airbase in Rawalpindi, located within the Chaklala military cantonment and in close proximity to the headquarters of the Strategic Plans Division, the secretariat of Pakistan’s National Command Authority.[xv] This proximity transformed what New Delhi characterised as conventional precision strikes into events with potentially existential implications in Islamabad’s strategic calculus.
The nuclear dimension was not merely structural. On 10 May, Pakistani Prime Minister Shehbaz Sharif convened an emergency meeting of the National Command Authority, the apex body responsible for all decisions relating to Pakistan’s nuclear arsenal.[xvi] Defence Minister Khawaja Asif subsequently denied that any such meeting had been formally constituted,[xvii] the denial arriving under apparent diplomatic pressure following calls from US Secretary of State Marco Rubio to both Army Chief General Asim Munir and Foreign Minister Ishaq Dar. Nevertheless, the reported convening marked the first publicly known NCA activation since February 2019, and the United States tracked a number of indicators associated with Pakistani nuclear readiness during this period.[xviii]
The May 2025 crisis demonstrated two things with uncomfortable clarity. First, conventional autonomous platforms can approach, and threaten, nuclear command-and-control infrastructure, blurring the boundary between conventional and nuclear operations in ways that generate acute escalation pressure. Second, the existing bilateral crisis management architecture between India and Pakistan remains grossly inadequate to the speed and opacity of modern drone warfare. Both findings acquire an order of magnitude greater urgency when the drones in question are nuclear-armed.
The Fourth Dimension of Deterrence
For most of the nuclear age, deterrence has been organised around the concept of the nuclear triad: land-based ballistic missiles, submarine-launched ballistic missiles, and nuclear-capable aircraft. This tripartite architecture was designed to ensure survivability, guaranteeing that no first strike could eliminate all retaliatory options. Vipin Narang’s influential analysis identifies three broad postures within this framework, asymmetric escalation, catalytic deterrence, and assured retaliation, each calibrated to different strategic environments and capabilities.[xix] What nuclear drones introduce is a potential fourth dimension that does not map cleanly onto any of these postures.
Autonomous nuclear systems, whether Poseidon-type underwater vehicles or nuclear-powered cruise missiles such as Burevestnik, share several characteristics that distinguish them from conventional triad elements. They offer potentially unlimited range and endurance, decoupling nuclear delivery from the logistical and basing constraints that have historically structured deterrence geography. They operate in domains, deep oceans and very low altitudes, where existing detection and interception architectures perform poorly. And their autonomous navigation raises profound questions of attribution, command and control, and the conditions under which a weapon, once launched, can be recalled or overridden. As Esra Serim has argued, increased autonomy in drone systems could inadvertently escalate nuclear tensions if AI-driven threat assessments misinterpret signals or inaccurately identify hostile intentions during crises and integrating autonomous drones into nuclear arsenals risks lowering the threshold for nuclear engagement by blurring distinctions between conventional military threats and genuine nuclear escalations.[xx]
This erosion of distinction is the central strategic risk. A drone does not announce its payload. An adversary observing an autonomous system approaching sensitive infrastructure cannot easily determine whether it carries a conventional warhead or a nuclear one. In a compressed, high-tempo crisis of the kind witnessed in May 2025, this ambiguity may be irresolvable within the decision timelines available to national command authorities. The fourth dimension, therefore, is not simply a new delivery vector; it is a structural alteration of the epistemics of nuclear deterrence, one in which the certainty upon which credible deterrence depends is systematically degraded.
South Asian Implications: Architecture Under Pressure
For India and Pakistan, the proliferation of nuclear autonomous systems creates a new set of pressures on already strained deterrence architectures. India maintains a declared No First Use policy combined with a doctrine of massive retaliation, though post-2019 political rhetoric has cast sustained doubt on the long-term credibility of that commitment.[xxi]Pakistan operates under Full Spectrum Deterrence, a posture explicitly designed to deter India across the entire escalation ladder from tactical to strategic levels, rejecting No First Use and reserving the right to first employment of nuclear weapons in the event of conventional military defeat or threat to territorial integrity.[xxii] These asymmetric doctrines, already generating instability in the context of Pakistan’s tactical nuclear weapons programme and India’s Cold Start concept, become considerably more complex when autonomous delivery systems are introduced.
Neither India nor Pakistan currently possesses nuclear-armed or nuclear-powered autonomous underwater vehicles. Pakistan’s maritime nuclear capability remains in development, centred on the Babur-III submarine-launched cruise missile,[xxiii] while India has operationalised INS Arihant and is commissioning a second nuclear-powered ballistic missile submarine. The question is not whether either state currently possesses Poseidon-equivalent capability but whether the global normalisation of nuclear autonomous systems will accelerate indigenous development programmes and whether the proliferation of conventional autonomous platforms creates the conditions for nuclear use through the misidentification and compressed decision-making dynamics already partially visible in May 2025.
There are at least three distinct escalation pathways. The first is direct emulation: if Russia’s Poseidon is assessed as operationally credible, and if China, Pakistan’s principal strategic partner, develops analogous systems, Islamabad will face significant institutional pressure to seek similar capabilities, either indigenously or through technology transfer. The second is the conventional-nuclear interface problem, already partially visible in May 2025: conventional drones operating against nuclear-associated infrastructure, creating escalation pressure through proximity rather than intent. The third is the attribution problem: in a densely contested environment of multiple autonomous platforms operating at high tempo, the risk that a nuclear command authority misidentifies a conventional drone swarm as a nuclear delivery system cannot be dismissed.
The compressed geography of the India-Pakistan nuclear environment makes each of these risks more acute. As the Belfer Center’s analysis of the May 2025 crisis concludes, the concept of fighting a limited nuclear war between India and Pakistan is a dangerous illusion.[xxiv] Given flight times measured in minutes and the logic of Full Spectrum Deterrence, the decision window for discriminating between conventional and nuclear autonomous systems may simply not exist. Pakistan’s doctrinal posture of escalation dominance at every rung of the conflict ladder creates an institutional incentive to respond early and decisively to any autonomous platform penetrating sensitive airspace, further compressing the available decision-making time.
The Governance Gap: An Absent Framework
What makes this landscape particularly alarming is the near-total absence of bilateral or multilateral governance mechanisms adequate to address it. India and Pakistan maintain a limited set of confidence-building measures, including an agreement on non-attack of nuclear installations and facilities dating from 1988 and a hotline between the Directors-General of Military Operations, both of which proved insufficient to prevent escalation in May 2025.[xxv] Neither measure contemplates autonomous systems, and neither establishes protocols for notification, de-escalation, or communication in the event of ambiguous autonomous platform activities near nuclear infrastructure.
At the multilateral level, the situation is no less precarious. The PIR Center’s 2026 analysis of the arms control landscape notes that the collapse of the New START framework in early 2026, combined with the failure of any successor agreement to emerge, has created a structural vacuum precisely at the moment when novel autonomous systems are achieving operational maturity.[xxvi] Poseidon is explicitly outside these regimes. No existing international legal framework addresses autonomous nuclear systems, and the Nuclear Non-Proliferation Treaty, to which India and Pakistan are not parties, provides no applicable constraint. The broader architecture of strategic stability, built on the assumption of identifiable, attributable, and interceptable delivery systems, is structurally ill-adapted to platforms designed to be none of these things.
The Toda Peace Institute has warned that, without robust international frameworks for autonomous nuclear systems, the destabilising effects of these technologies are more likely than the stabilising ones.[xxvii] For South Asia, where conflict cycles have accelerated from years to weeks and where nuclear thresholds have been the subject of sustained ambiguity, this governance gap represents a structural hazard of the highest order. The May 2025 crisis was resolved, narrowly, by American diplomatic intervention. It is far from clear that such intervention would be timely or effective in a scenario involving nuclear-armed autonomous systems operating without human authorisation at each step of their mission profile.
Policy Recommendations
Several urgent measures follow from this analysis. First, India and Pakistan should move without delay to expand and modernise their bilateral communication architecture, specifically to include protocols governing autonomous platform identification, mandatory notification of exercises involving autonomous systems operating near sensitive nuclear infrastructure, and dedicated channels for crisis communication in high-tempo autonomous conflicts. The existing hotline mechanisms were designed for a slower era and are inadequate to the operational realities of modern drone warfare.
Second, both states should engage in Track 1.5 and Track 2 dialogue specifically on the implications of autonomous nuclear systems for South Asian stability, building a common analytical framework before either state acquires nuclear autonomous delivery capability. Engagement through established venues, including Carnegie’s Nuclear Policy Programme and the regional dialogue infrastructure supported by the Arms Control Association, offers one institutional pathway for such conversations.[xxviii]
Third, the international community should prioritise negotiating a governance instrument, outside the NPT framework, if necessary, that addresses autonomous nuclear systems specifically, establishing verification requirements, notification protocols, and deployment restrictions equivalent to those applied to ballistic missiles under existing treaties. The exclusion of Poseidon from New START was a deliberate design feature. Any credible successor framework must close that gap.
Fourth, India and Pakistan should bilaterally agree on a minimum stand-off distance for autonomous platforms operating near the other state’s nuclear facilities, analogous to the 1988 agreement on nuclear installation non-attack, but adapted to the operational realities of modern unmanned systems. This is a modest but potentially decisive measure that requires no fundamental concession from either party on doctrine or capability.
Conclusion
The Poseidon effect is not merely a Russian strategic development. It is a global inflection point that reveals the inadequacy of deterrence architectures built for a world of ballistic missiles and manned aircraft. For South Asia, where the May 2025 crisis demonstrated both the operational centrality of autonomous platforms and the terrifying proximity of conventional strikes to nuclear command-and-control infrastructure, the emergence of nuclear-armed and nuclear-powered autonomous systems represents a qualitatively new and inadequately governed threat.
India and Pakistan are not currently building nuclear drones. But the strategic pressure to do so will intensify as global norms around autonomous nuclear systems are established or fail to be established. Russia’s Poseidon normalises the concept; China’s parallel programmes will accelerate it; and Pakistan’s structural dependence on Chinese strategic technology means that the question of nuclear drone proliferation to South Asia is a matter of timing rather than probability. The governance vacuum in which such systems would operate already exists. The time to address that vacuum is before the race begins, not after. The fourth dimension of deterrence is opening. South Asia cannot afford to enter it unprepared.
[i]Euronews, “Moscow Says It Tested Poseidon Underwater Drone, Another Nuclear ‘Super Weapon’,” October 29, 2025, https://www.euronews.com/2025/10/29/moscow-says-it-tested-poseidon-underwater-drone-another-nuclear-super-weapon.
[ii]Army Recognition, “Russia Claims Successful Launch of Poseidon Nuclear Submarine Drone Said to Create ‘Tsunamis’,” October 29, 2025, https://www.armyrecognition.com/news/navy-news/2025/russia-claims-successful-launch-of-poseidon-nuclear-submarine-drone-said-to-create-tsunamis.
[iii]International Institute for Strategic Studies, “India-Pakistan Drone and Missile Conflict: Differing and Disputed Narratives,” May 2025, https://www.iiss.org/online-analysis/online-analysis/2025/05/indiapakistan-drone-and-missile-conflict-differing-and-disputed-narratives/.
[iv]Christopher Clary, “Four Days in May: The India-Pakistan Crisis of 2025,” Stimson Center, June 23, 2025, https://www.stimson.org/2025/four-days-in-may-the-india-pakistan-crisis-of-2025/.
[v]Congressional Research Service, “India-Pakistan Conflict in Spring 2025,” Library of Congress, 2025, https://www.congress.gov/crs-product/IF13000.
[vi]Observer Diplomat, “Putin Announces Successful Test of Nuclear-Capable Poseidon Drone,” November 7, 2025, https://observerdiplomat.com/putin-announces-successful-test-of-nuclear-capable-poseidon-drone/.
[vii]Milivox Media, “Poseidon Unveiled: Russia’s Nuclear-Powered Underwater Drone and Its Strategic Implications,” October 31, 2025, https://milivox.media/poseidon-russia-nuclear-underwater-drone-analysis/.
[viii]Observer Diplomat, “Putin Announces Successful Test of Nuclear-Capable Poseidon Drone,” November 7, 2025. Moscow has stated that it will not provide advance notification of Poseidon test launches, consistent with the system’s deliberate exclusion from New START and successor frameworks.
[ix]Lawrence Freedman, Deterrence (Cambridge: Polity Press, 2004), 26-29.
[x]Euronews, “Moscow Says It Tested Poseidon Underwater Drone,” October 29, 2025. The tsunami-generation narrative originates in Russian state commentary following Putin’s 2018 unveiling and has been consistently amplified since.
[xi]Congressional Research Service, “India-Pakistan Conflict in Spring 2025,” 2025.
[xii]“The May 2025 India-Pakistan Conflict: Neither Quite the Same Nor Quite Another,” National Bureau of Asian Research, 2025, https://nbr.org/publication/the-may-2025-india-pakistan-conflict-neither-quite-the-same-nor-quite-another/.
[xiii]National Bureau of Asian Research, “The May 2025 India-Pakistan Conflict.”
[xiv]Christopher Clary, “Four Days in May,” Stimson Center, June 23, 2025.
[xv]National Bureau of Asian Research, “The May 2025 India-Pakistan Conflict.” The Nur Khan airbase is located within the Chaklala military cantonment, in close proximity to the Strategic Plans Division, which serves as the secretariat of Pakistan’s National Command Authority.
[xvi]Xinhua, “Pakistani PM Calls Emergency Meeting to Discuss Potential Nuclear Response to Indian Strikes,” May 10, 2025, https://english.news.cn/20250510/0643eb97de0e4e57a4c771ff71f3ec41/c.html.
[xvii]The Express Tribune, “PM Convenes National Command Authority Meeting amid Ongoing Military Operation against India,” May 10, 2025, https://tribune.com.pk/story/2545033/pm-convenes-national-command-authority-meeting-amid-ongoing-military-operation-against-india. The denial came after reported calls from US Secretary of State Marco Rubio to both General Asim Munir and Foreign Minister Ishaq Dar.
[xviii]Congressional Research Service, “India-Pakistan Conflict in Spring 2025,” 2025.
[xix]Vipin Narang, Nuclear Strategy in the Modern Era: Regional Powers and International Conflict (Princeton: Princeton University Press, 2014), 3-18.
[xx]Esra Serim, “Drone Technology and the Future of Nuclear Weapons,” Toda Peace Institute, August 25, 2025, https://toda.org/global-outlook/2025/drone-technology-and-the-future-of-nuclear-weapons.html.
[xxi]“Escalation Gone Meta: Strategic Lessons from the 2025 India-Pakistan Crisis,” Belfer Center for Science and International Affairs, Harvard Kennedy School, May 14, 2025, https://www.belfercenter.org/research-analysis/escalation-gone-meta-strategic-lessons-2025-india-pakistan-crisis.
[xxii]Hans M. Kristensen, Matt Korda, Eliana Johns, and Mackenzie Knight, “Pakistan Nuclear Weapons, 2025,” Bulletin of the Atomic Scientists, September 2025, https://thebulletin.org/premium/2025-09/pakistan-nuclear-weapons-2025/.
[xxiii]Kristensen et al., “Pakistan Nuclear Weapons, 2025.”
[xxiv]“Escalation Gone Meta: Strategic Lessons from the 2025 India-Pakistan Crisis,” Belfer Center for Science and International Affairs, May 14, 2025.
[xxv]Congressional Research Service, “India-Pakistan Conflict in Spring 2025,” 2025. The Agreement on the Prohibition of Attack against Nuclear Installations and Facilities was signed December 31, 1988 and entered into force January 27, 1991.
[xxvi]PIR Center, “Nuclear Weapons: What Future for Arms Control?” Security Index Yearbook, No. 1(10) (2026), https://pircenter.org/en/editions/%E2%84%96-1-10-2026-nuclear-weapons-what-future-for-arms-control/.
[xxvii]Esra Serim, “Drone Technology and the Future of Nuclear Weapons,” Toda Peace Institute, August 25, 2025.
[xxviii]Daryl G. Kimball, “Pakistan and the Nuclear Danger in Asia,” Arms Control Today, January/February 2025, https://www.armscontrol.org/act/2025-01/focus/pakistan-and-nuclear-danger-asia.
Tahir Azad
Related posts
The Maritime Blockade and the Strategic Calculus of the U.S., Iran, and Israel Toward a Renewed Armed Confrontation
Understanding Asim Munir’s Doctrine: Regime Security in Place of a National Goal
The Data Dilemma: Navigating the AI Systems Matrix
The Anatomy of a Shortfall: Europe’s Nine Capability Areas and the Lessons for Pakistan
From Quiet Diplomacy to Attempted Maritime Security Governance: Oman’s Role and Limits in the 2026 Strait of Hormuz Crisis
Anatomy of Asymmetry: How a Weaker State Converts Vulnerability into Leverage
Integration for Stability: The SDF’s Dissolution and Syria’s Security Order
The Western Flank Unravelled: Cross-Border Militancy, TTP Resurgence, and Afghan Taliban Governance
The Time of Guards Tank Divisions is Gone, or Maybe It is Not?
Re-pricing Israel’s Corridor: The Mecca Alliance and IMEC
The SCO at 25: From Regional Security Mechanism to Platform for a Multipolar Eurasia
Imran Khan: The Last Hope of a Nation
