On April 7, 2026, US President Donald Trump posted on Truth Social that “a whole civilization will die tonight, never to be brought back again”. The immediate context was Iran’s closure of the Strait of Hormuz. The broader context was the war Trump had launched five weeks earlier on the intelligence his own Director of National Intelligence subsequently contradicted before Congress. But the post did something beyond the immediate crisis: it placed civilisational annihilation on the table as a coercive instrument. In the grammar of deterrence, that kind of ambiguity is not rhetorical excess. It is a signal, and every government operating below the nuclear threshold received it as one.

The Iran crisis is the most visible consequence of a policy trajectory that began during Trump’s first term and accelerated under his second. In conversations I held recently in Washington, DC, with George Perkovich of the Carnegie Endowment for International Peace, Ankit Panda, author of The New Nuclear Age: At the Precipice of Armageddon, and Kunal Singh of the Harvard Kennedy School’s Belfer Center, a consistent analytical conclusion about this trajectory emerged: the US has spent eight years systematically dismantling the architecture of nuclear restraint that it had spent decades building.

Trump’s 2018 withdrawal from the Joint Comprehensive Plan of Action (JCPOA) is familiar history by now. But its precise consequences are stated less often. In 2015, the JCPOA capped Iranian enrichment at 3.67 per cent, reduced its centrifuge count by two-thirds, and put in place an inspection architecture the International Atomic Energy Agency (IAEA) certified as functional. When Washington walked away, Tehran remained in compliance. Writing in Bulletin of the Atomic Scientists in April, Perkovich argues that the US’ withdrawal cracked the brittle foundation of the global non-proliferation regime and that its full implications were not visible at the time. They are visible now.

The particular irony of the current moment is that the terms American negotiators brought to Islamabad—verified enrichment limits, centrifuge constraints, and robust IAEA access—were in substantial measure what the JCPOA had already secured. The demand for a 20-year enrichment moratorium and full removal of enriched uranium represents not an advance on the 2015 agreement but a maximalist restatement of its central logic, pursued without the diplomatic coalition, the multilateral framework, or the Iranian confidence in American commitment that made the original agreement achievable. Eight years and two wars later, Washington is negotiating to recover ground that it voluntarily surrendered.

Trump’s second term compounded the setback by shifting from coercion to compellence. The demand was for Iranian capitulation across enrichment, missiles, proxy networks, and revolutionary governance, with nothing substantive being offered in return. As Perkovich argues, the war on Iran has displaced the NPT’s foundational logic of reciprocity, substituting it with a stark demand for unconditional surrender. The damage is not to a single negotiation. It is to the underlying model on which non-proliferation has always depended.

The US’ intelligence record sharpens the picture further. US Director of National Intelligence Tulsi Gabbard testified to Congress that Iran’s enrichment programme was obliterated by the June 2025 strikes and showed no signs of reconstruction. The IAEA Director General, too, had said the same thing. These findings directly contradicted Trump’s stated justification for the February 2026 war that Iran was two weeks away from possessing a nuclear weapon. A war launched on contested intelligence, framed for the public as settled fact, sets a precedent whose consequences extend well beyond the Persian Gulf.

Ankit Panda, in the Washington conversations, was direct about the proliferation signal that the Iran case now broadcasts globally. The cumulative effect of the June 2025 Israeli strikes on Iran and the current conflict will increase interest in nuclear weapons possession worldwide. One of the lessons states are drawing from the episode is that Iran’s posture had been the most self-defeating. It had stayed proximate enough to a weapon to justify preventive attack, while remaining unwilling to cross the threshold that might have in fact prevented one. The result was strategic self-defeat at the worst possible moment.

Now, the likely end state end state makes its position worse. According to Panda, some version of the Iranian revolutionary regime will survive this conflict, and while some of its conventional deterrence will be degraded, approximately 400 kg of highly enriched uranium and centrifuge components will remain intact beneath the rubble. Weakened conventionally, humiliated politically, and retaining fissile material provide a near-perfect set of conditions for breakneck proliferation. The war may fail to terminate Iran’s nuclear ambitions but may produce a new Iran modelled on North Korea.

This is the conclusion that travels across the globe. North Korea has been studied closely by US adversaries since 2003, when the US invaded Iraq and attempted to kill Saddam Hussein at the outset. Contrasting Iran with North Korea’s trajectory—defined by the latter’s three summits with a US President and its uninterrupted regime survival—has always been instructive. The Iran war makes it definitive.

What the US has unsettled

Kunal Singh’s research on counterproliferation strategy identifies five approaches that states have historically used to prevent nuclear acquisition: kinetic reversion, military coercion, diplomatic inhibition, pooled prevention, and accommodation. The successful ones have depended on coordinated pressure and credible security guarantees operating together. Singh’s core finding is that states are concerned about proliferation, but they are not so concerned that they will destabilise the present international order to prevent a future threat. That calibration—careful, institutionally grounded, and multi-instrumental—is what the current US approach has unsettled.

The effect on allies is already measurable. The two forces that historically constrained South Korean and Japanese nuclear ambitions were a credible US nuclear umbrella and an active US non-proliferation policy. Both are now in question. Roughly 70 per cent of the South Korean public now supports some role for an independent nuclear capability. Saudi Arabia has publicly committed to matching any existing Iranian enrichment. These are rational recalculations by states reassessing the reliability of American commitments under the transactional foreign policy framework.

Tong Zhao’s work on Chinese nuclear modernisation provides a systemic perspective that completes this picture. China’s build-up, Zhao has demonstrated, is driven substantially by domestic political imperatives rather than simply by external threat perception. Its approach to nuclear transparency is oriented towards delivering deterrence signals rather than strategic reassurance, a posture that forecloses the kind of confidence-building that arms control requires. The US’ concern, which Zhao has documented carefully, is that China’s developing capabilities like the Fractional Orbital Bombardment System could enable strikes against US leadership targets and command-and-control infrastructure with warning times too compressed for managed escalation.

In a February 2026 conversation with the foreign affairs think tank Stimson Centre’s Daniel Markey, Zhao situated China’s nuclear modernisation in the context of regional dynamics of launch-on-warning postures, the China-India nuclear dyad, and the Chinese military support to Pakistan during the May 2025 crisis. The picture that emerges is of interlocking competitive pressures across every regional nuclear theatre, with no arms control architecture capable of restraining any of them.

The lapse of New START—the nuclear arms reduction treaty between the US and Russia signed on April 8, 2010—and the failure to renew it, as well as the complete absence of any multilateral framework that includes China, means this competition is now structurally unconstrained. Zhao’s analysis confirms from Beijing’s side what Perkovich, Panda, and Singh identify from Washington’s: the architecture is gone, and the space it occupied is being filled with something considerably less stable.

The non-proliferation regime has absorbed serious shocks before: India and Pakistan’s 1998 tests, North Korea’s sustained defiance, and the NPT’s failure to achieve universality. But what it has not previously encountered is the withdrawal of its principal architect from the foundational commitments that sustained the treaty. The unity among the five nuclear weapons states in their response to proliferation threats was, for decades, the central mechanism of deterrence and coercion. As that unity erodes, the capacity to impose costs on would-be proliferators diminishes, and the appeal of the nuclear option correspondingly grows globally.

Trump’s Truth Social post is not, in formal terms, a nuclear threat. But in the strategic environment that his administration has constructed—by withdrawing from the JCPOA with Iran, by launching wars based on contested intelligence, by making alliance guarantees transactional, and by allowing the arms control architecture to lapse—Trump’s post lands as a confirmation that the boundary between coercion and annihilation is no longer drawn where it was. The state that built that boundary appears to be erasing it.

The Iran crisis will not be the last test of what remains. It may, however, mark the moment when the scale of the transformation becomes impossible for serious analysts to set aside in future conversations around the nuclear order.

Anil Raman is a retired Army Brigadier who focusses on US politics and foreign policy in the Geostrategy Programme at the Takshashila Institution.

Also Read | Nuclear deterrence in South Asia looks stable only on paper

Also Read | A pause that is not a return to normalcy