Nonproliferation policy rests on an assumption so basic it is rarely stated: that safeguards and enforcement work together. Inspections detect a violation; enforcement — sanctions, diplomacy, in the extreme, force — corrects it. The two are understood as sequential parts of a single system, the second acting on what the first reveals.
The Iran campaign of 2026 has produced the first clear empirical case in which they did not cooperate but cancelled. Enforcement did not act on what the safeguards system revealed. It destroyed the safeguards system’s ability to reveal anything.
The sequence is now documented in the International Atomic Energy Agency’s own reporting. The last verified figure for Iran’s most sensitive material — uranium enriched to 60 percent — was 440.9 kilograms, recorded on the date of the final pre-strike safeguards report. Operation Epic Fury began the following day.
On July 10, 2026, the Agency told the UN Security Council that it had lost continuity of knowledge across all of Iran’s declared nuclear facilities — not only the enriched uranium inventory but the production and current inventory of centrifuges, rotors and bellows, heavy water, and uranium ore concentrate.
The phrase “continuity of knowledge” is a term of art. It means the chain-of-custody record for specific nuclear material has been broken, and, as former senior IAEA safeguards official Laura Rockwood has emphasized, once broken it cannot be reconstructed from external observation alone.
Even if every door in Iran opened tomorrow, the Agency could not certify what happened to that material during the blackout. The knowledge is not suspended. It is gone.
This is the paradox in its exact form. The world knew about 440.9 kilograms of 60 percent HEU because Iran was a safeguarded state. That number was a product of the inspection regime — the one instrument that generated the very information the enforcement action was justified by. The strikes were premised on the danger that inventory represented. And the strikes ended the regime that could account for it.
What began the campaign with a specific, inspected, quantified stockpile has ended it with a figure that is now four months stale, unverifiable, and — because Iran likely relocated the Esfahan stock into a tunnel complex before the strikes — quite possibly no longer where the last inspection found it. Enforcement was supposed to reduce the proliferation risk the safeguards system had measured. Instead it dismantled the measurement.
Read More: Pakistan Warns Diplomacy Breakdown Complicating Iran Nuclear Issue
The objection to this argument is obvious and must be met directly: the loss of continuity is Iran’s fault, not the strikes’. Iran suspended cooperation, passed a law barring inspectors, and has refused access since. All true, and the Agency continues to find Iran in breach of its safeguards obligations. But this objection misunderstands the claim.
The argument is not that the strikes were the sole cause of the monitoring collapse — Iranian obstruction and the June 2025 strikes had already degraded the baseline. The argument is structural: that military enforcement against a safeguarded nuclear program, whatever its justification, predictably terminates the safeguards relationship that made the program legible in the first place, and that this cost is not incidental but intrinsic.
A state that has just been bombed will not admit the inspectors of the powers that bombed it. That is not an Iranian idiosyncrasy. It is what any state would do, and it is therefore a general property of the instrument, not a feature of this case.
Why does this matter beyond Iran? Because it inverts the incentive the nonproliferation regime is built to create. The regime offers a bargain: accept safeguards, gain legitimacy and the benefits of the peaceful-use framework. The implicit promise is that transparency is safer than opacity.
The Iran case, read by every threshold state, suggests something closer to the reverse. Iran accepted enrichment monitoring and was found, through that monitoring, to hold material for which there is no civilian justification — and was then struck, after which the monitoring ended and the material vanished from the ledger.
Read More: Vance Says Iran Must Cut Nuclear Enrichment Capacity to End War
A state weighing whether to accept intrusive verification now has a worked example in which verification produced the intelligence that justified an attack and then evaporated when the attack came. The lesson is not that safeguards are worthless. It is that safeguards without a guarantee against their being used as a targeting database are a gamble, and the regime has no mechanism to offer that guarantee.
There is a second-order consequence that the current diplomacy has not absorbed. The Islamabad memorandum’s nuclear track opened a 60-day clock toward a nonproliferation agreement. But an agreement presupposes a baseline — an agreed account of what exists and where — and that baseline no longer exists.
The negotiation is being conducted over a stockpile whose size, composition, and location the Agency has told the Security Council it cannot establish. Every month of continued non-access subtracts permanently from what any future agreement can verify. Continuity of knowledge is the one variable in this crisis that only degrades and never recovers with time, and it is the one variable the diplomatic timetable treats as though it can wait.
Read More: Fresh US-Iran Rift Emerges Over Nuclear Inspections and Frozen Assets
The policy implication is narrow, specific, and unwelcome to all parties. Restoring inspector access must be decoupled from every other item on the negotiating agenda — sanctions relief, enrichment limits, regional questions — because it is the only item that becomes irreversibly harder with delay.
It should not be a concession Iran trades for relief or a chip the United States withholds for leverage. It is the precondition for there being anything to negotiate about, and treating it as bargaining currency guarantees that whatever agreement eventually emerges will certify less than the one that could have been reached a month earlier.
The deeper lesson is for the architecture itself. The nonproliferation system has always assumed that its detection function and its enforcement function are allies. The Iran case demonstrates that under military enforcement they can be adversaries — that the act of enforcing can consume the evidence base that enforcement depends on, leaving the international community with less knowledge after the intervention than before it.
That is a structural defect, not a policy error, and it will recur wherever force is used against a safeguarded program. A regime that cannot see this contradiction will keep producing interventions that are justified by inspection and end by destroying it. Four hundred and forty kilograms is not hidden. It is unaccounted for — and it was the accounting, not the hiding, that the enforcement destroyed.
*The views presented in this article are the authors’ own and do not necessarily reflect the views of The Diplomatic Insight.*
Aleena Saif Ullah
Aleena Saif Ullahis an MPhil Scholar in International Relations at the University of Punjab, Lahore. She can be reached at[email protected]
