On October 27, 1962, at the height of the Cuban missile crisis, American warships dropped small explosive charges near a Soviet submarine off the coast of Cuba. Their purpose was to force the B-59 submarine to surface. The Americans did not know that it carried a nuclear torpedo. Inside the submarine, officers who had been cut off from reliable communications struggled to determine whether war had begun. Captain Valentin Savitsky came close to ordering the torpedo’s use but another senior officer aboard, Vasili Arkhipov, helped prevent that decision. The episode was fully understood decades later, a reminder of how close the world came to the use of nuclear weapons through misperception, incomplete information, and compressed decision timelines under stress.

The incident, dramatised in The Man Who Saved the World, a 2012 television documentary in the PBS series Secrets of the Dead series, illustrates how nuclear escalation can begin without either belligerent choosing to go nuclear. One side conducts an operation it considers limited while the other encounters explosions, incomplete information and uncertainty about what will happen next. Decisions then depend on officers under tremendous pressure, interpreting intentions they cannot verify. The 88-hour long military conflict between India and Pakistan in May 2025 brought this problem into sharper focus, when conventional weapons fired by India reached deep into enemy territory, including Kirana Hills, associated with Pakistan’s nuclear infrastructure. In August 2026, former Indian chief of defence staff Anil Chauhan revealed that an Indian loitering munition searching for radars at Sargodha airbase in Pakistan had exhausted its endurance and struck a hill there unintentionally.

Chauhan’s account does not establish that India attacked a nuclear storage installation, but it strengthens the case that conventional operations produced effects near a sensitive nuclear location. Pakistan would have had to interpret those effects while attacks were continuing. An accidental impact can resemble reconnaissance, a warning or the opening move of a campaign against nuclear forces. India’s confirmed attack on Nur Khan airbase on the same day created a related problem. The base shares the wider Chaklala cantonment with Pakistan’s Strategic Plans Division, responsible for nuclear planning, although the offices are more than a kilometre away. Physical separation matters, but it cannot guarantee reassurance during the fog of war. Pakistan could have easily perceived the attack as more dangerous than India intended and moved on to take a nuclear posture.

Chauhan has described the military as having received a free hand to conduct operations, with decisions on escalation made under time pressure and with limited real-time political oversight once the campaign began. Hotline communications between the two directorates-general of military operations experienced reported disruptions at critical moments. These details, drawn from public accounts by participants, illustrate how operational tempo, incomplete information and the absence of tightly managed political control can produce near misses even when neither side seeks nuclear use.

The Cuban missile crisis’s enduring lesson regarding the gap between what military operations intend and what frightened nuclear-armed adversaries believe they mean has been forgotten by India and Pakistan. An attacker’s confidence in precision or its non-nuclear intentions does not determine the defender’s interpretation of danger. More crucially, both the Soviet Union and the United States learnt the right lessons from the Cuban crisis and chose to eschew all pathways that could lead to even inadvertent nuclear escalation. The baseline of conflict was reset between the two superpowers for the rest of the Cold War.

The military generals in Pakistan and the political leaders in India have chosen the opposite path. To them, the four days of fighting last May demonstrated that missiles, drones and aircraft could strike across international borders without immediately producing nuclear escalation. The geographical and operational boundaries of conflict widened rapidly, but militaries on both sides believed that they had total control over the escalation ladder. Both governments claimed success and were hailed by their domestic publics, which were infused with a high dose of nationalism. India now believes it has expanded the space for conventional military operations under the nuclear threshold while Pakistan wants its full spectrum deterrence to remain effective by shrinking that space. These incompatible assumptions make the next military conflict in South Asia harder to contain.

South Asia was lucky that the Trump administration negotiated and announced a ceasefire after four days of military exchanges between India and Pakistan. However, it didn’t resolve the problem, instead freezing a higher baseline for the next confrontation, by leaving the underlying conflict unsettled and competing narratives intact. Think of it this way. After the next big terror attack in India believed to be linked to Pakistan, any government in New Delhi will be under pressure to demonstrate that its military response exceeds previous operations. Pakistan’s military will face corresponding pressure to preserve its reputation and impose visible costs on India. Each side will therefore regard escalation as necessary to restore deterrence.

General Chauhan has suggested that Pakistan’s decision to escalate conventionally in May 2025 makes subsequent nuclear escalation in future conflicts less likely. No such sequencing rule binds Pakistan. Its choices would depend on battlefield developments, perceived threats and political judgments, and not on India’s declared policy of no first use for its nuclear weapons. If the Pakistani leadership believes Indian attacks threaten its strategic capabilities or military’s survival or could lead to a split in the country, it will not hesitate to signal the use of its nuclear option. That has been made explicitly clear on numerous occasions by the chief architect of Pakistan’s nuclear deterrence doctrine, Lt General Khalid Kidwai.

The Federation of American Scientists’ Nuclear Notebook estimates that India may have assembled up to 190 nuclear warheads by 2026, with sufficient plutonium for 140 to 225, and continues to field additional delivery systems across a maturing triad of aircraft, land-based missiles, and ballistic-missile submarines. Pakistan is assessed to hold a comparable number, around 170 warheads, and has accelerated modernisation of its own forces, including tactical nuclear weapons intended to offset India’s conventional superiority. Both sides are fielding faster, more accurate systems such as India’s BrahMos cruise missiles and expanding missile defences and dual-capable platforms. These developments compress warning and decision times. The 2022 accidental Indian BrahMos launch into Pakistani territory, which prompted Pakistan to suspend air traffic and raise alert levels for hours while India remained mum, demonstrated how a single technical failure during peacetime can produce acute tension. In a crisis environment, the same error could prove far harder to contain.

Analysts have long identified South Asia as the region where accidental or inadvertent nuclear use is most plausible. Barry Posen’s work on inadvertent escalation showed that conventional operations can threaten an adversary’s nuclear forces or command systems and thereby generate pressures for early nuclear employment even when neither party intends it. In South Asia the geography is unforgiving where short flight times, high population densities along the border, and overlapping conventional and nuclear force deployments leave little margin for error. Recurrent triggers such as terrorist attacks, artillery exchanges along the Line of Control and competing claims over Kashmir keep the two sides in a cycle of crisis. The absence of robust, institutionalised confidence-building measures, shared nuclear doctrines or reliable crisis communication protocols beyond a fragile military hotline further elevates the danger. Studies by scholars such as Vipin Narang and Scott Sagan have documented how Pakistan’s full-spectrum deterrence posture and India’s evolving dynamic response doctrine create competing incentives to demonstrate resolve while testing the other’s red lines.

These dynamics carry direct consequences for the United States, as the broader US-led international order rests on the residual authority of the nuclear non-proliferation regime. Preventing Iranian nuclear acquisition was an announced objective of the US and Israeli operations launched in February 2026. The credibility of that effort depends on the perception that existing nuclear-armed states, including those like India and Pakistan which are outside the Non-Proliferation Treaty, manage their arsenals with restraint and that crises among them do not produce uncontrolled escalation. A serious nuclear incident between India and Pakistan would undermine that perception, complicate US efforts to maintain alliance commitments in Asia and raise questions about the durability of the entire non-proliferation architecture.

Emerging domains add further complexity. The United States has confirmed for the first time that it has deployed weapons in space. Both India and China are developing space-based capabilities, including anti-satellite systems and dual-use technologies that could blur the line between conventional and strategic operations. Proposals for lunar reactors and space-based sensors or weapons, while still nascent, introduce new pathways for miscalculation. An incident involving space assets during a South Asian crisis could cascade rapidly, given the reliance of modern command-and-control systems on satellites. The United States, as the leading space power and the primary external actor with influence in both New Delhi and Islamabad, would face immediate demands to clarify its own red lines and to prevent the conflict from expanding beyond the subcontinent.

A wargaming of South Asia’s distinctive danger combines recurring terrorist triggers, rapid conventional retaliation, sensitive military geography and contested crisis narratives. The combination of faster weapons, deeper military attacks, less constrained operational practices, overconfidence in escalation control, incompatible victory narratives and thin crisis-management institutions have reduced the margin for error. Successful termination of an earlier conflict has given political and military leadership in India and Pakistan confidence that the same methods will work again, even when the next attack, political circumstances or battlefield losses differ dramatically. The resulting danger, of South Asia being the most dangerous global nuclear flashpoint, comes from these mutually reinforcing expectations of limited, manageable violence between two nuclear-armed states. The world cannot afford another confrontation in South Asia which tests assumptions that fortuitously survived once.