INVESTIGATION: 9,874.9 Kilograms on Paper, but No Current Iranian Inventory

Introduction

The International Atomic Energy Agency’s last quantified inventory put Iran’s enriched-uranium stock at 9,874.9 kg on June 13, 2025—not in August 2026. The public record is narrower than the rhetoric wrapped around it.

The critical fact is not that the agency lacks a number. It has a detailed old number. The failure is that the IAEA says it cannot verify what happened after inspectors lost broad access.

The last measured inventory

A number fixed in time

The dated record is specific: IAEA report GOV/2026/8, dated February 27, 2026, records Iran’s enriched-uranium inventory at 9,874.9 kg on June 13, 2025. For a number fixed in time, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.Pressure without a shared text is not progress; it is pressure.

Its consequence: The precision is real, but the date makes it a last known data point rather than an August 2026 measurement. The question here is a number fixed in time, and the available record reaches no further.

An inventory is not a forecast

That sequence is specific: The IAEA presents the number as an accounting record, not as a current assessment of Iran’s programme or intentions. For an inventory is not a forecast, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

That matters because: Turning a dated quantity into an estimate of present capability would add a conclusion the report does not make. The question here is an inventory is not a forecast, and the available record reaches no further.

What the total contained

Most was listed as UF6

The available account is specific: The IAEA broke the June 13, 2025 total into 9,040.5 kg of UF6 and 834.4 kg in other forms. For most was listed as uf6, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.A deadline can create urgency. It cannot manufacture consent.

The limit is concrete: That breakdown describes the old inventory in more detail; it does not restore physical access to it. The question here is most was listed as uf6, and the available record reaches no further.

The categories still add to a past total

One documented detail is specific: The two reported components equal the 9,874.9 kg total recorded by the agency. For the categories still add to a past total, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

This does not settle: Arithmetic can validate the old ledger without updating what happened after the ledger closed. The question here is the categories still add to a past total, and the available record reaches no further.

The strikes and the break in continuity

Israel struck facilities in June 2025

The next fact is specific: GOV/2026/8 confirms that Israel struck Iranian nuclear facilities between June 13 and June 24, 2025. For israel struck facilities in june 2025, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.The channel matters because the channel defines what was actually discussed.

The proper conclusion: The chronology explains disruption, but it does not reveal the fate of material inside or outside the struck sites. The question here is israel struck facilities in june 2025, and the available record reaches no further.

The United States struck three facilities

The public statement is specific: The same report says the United States struck three facilities on June 22, 2025. For the united states struck three facilities, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

No source here authorizes: A military strike is not a safeguard inspection and cannot substitute for a verified inventory. The question here is the united states struck three facilities, and the available record reaches no further.

Inspectors left

The withdrawal was complete by late June

This part of the file is specific: The IAEA had withdrawn all inspectors from Iran by the end of June 2025, according to GOV/2026/8. For the withdrawal was complete by late june, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.A draft in circulation is not an obligation in force.

The record therefore shows: Without inspectors, the agency cannot certify what it has not observed or accounted for. The question here is the withdrawal was complete by late june, and the available record reaches no further.

The gap has no substitute in the file

A separate source is specific: The dossier does not identify an equivalent remote or documentary mechanism that restored comprehensive verification after the withdrawal. For the gap has no substitute in the file, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The caution is not cosmetic: The present condition cannot be inferred from an absent monitoring chain. The question here is the gap has no substitute in the file, and the available record reaches no further.

Bushehr is the narrow exception

Inspections resumed at one plant

The chronology is specific: The agency resumed inspections at the Bushehr Nuclear Power Plant in late August 2025. For inspections resumed at one plant, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.An estimate is useful only when its author and limits stay attached.

What follows is narrower: Bushehr is identified as the sole facility with continued access, not a window onto the whole programme. The question here is inspections resumed at one plant, and the available record reaches no further.

One open site cannot represent four enrichment sites

The reported limit is specific: The IAEA separates Bushehr from other safeguarded facilities and related stocks. For one open site cannot represent four enrichment sites, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The status remains: Local access confirms the scale of the wider access problem. The question here is one open site cannot represent four enrichment sites, and the available record reaches no further.

The missing paperwork

No nuclear material accounting reports arrived

The mechanism is specific: GOV/2025/50 says the agency had received no nuclear material accounting reports from Iran since June 13, 2025. For no nuclear material accounting reports arrived, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.A public claim does not become independent confirmation by repetition.

That leaves: The absence blocks a documentary route to updating the inventory. The question here is no nuclear material accounting reports arrived, and the available record reaches no further.

Design information was not updated

The comparison is specific: The report also records no updated design-information questionnaires after that date. For design information was not updated, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The missing element: The gap concerns both quantities and the information needed to maintain oversight of installations. The question here is design information was not updated, and the available record reaches no further.

Safeguarded sites out of reach

The official exception is Bushehr

The unresolved point is specific: GOV/2025/50 says the agency had no access to safeguarded nuclear facilities in Iran after June 13, 2025, except Bushehr. For the official exception is bushehr, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.The missing access is not a footnote. It is the central fact.

This separates: The wording describes an inability to verify, not a temporary inconvenience. The question here is the official exception is bushehr, and the available record reaches no further.

Britain cited four enrichment facilities

The official record is specific: In June 2026, the United Kingdom told the IAEA Board that Iran had refused required information or access to four enrichment facilities and associated stocks. For britain cited four enrichment facilities, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The implication is limited: That is London’s stated position, consistent with the agency’s access problem, not a new physical inventory. The question here is britain cited four enrichment facilities, and the available record reaches no further.

The June 2026 report

Reuters described little change in the access finding

The practical consequence is specific: Reuters reported on June 4 that the first IAEA report since the February strikes showed little change: access to bombed sites remained unavailable. For reuters described little change in the access finding, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.No map, count, or warning becomes final because one party says so.

A claim cannot replace: “Little change” refers to the verification impasse, not proof that the material itself stayed unchanged. The question here is reuters described little change in the access finding, and the available record reaches no further.

Stocks were not notified

The evidence on hand is specific: Reuters reported that Iran had not notified the agency about its enriched-uranium stocks, including earlier reporting of material enriched up to about 60%. For stocks were not notified, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The evidence stops: That purity reference comes from earlier reporting, not a new physical check. The question here is stocks were not notified, and the available record reaches no further.

Grossi’s stated condition

A strong verification system was requested

The distinction is specific: On June 26, Rafael Grossi said “We have to have a very strong verification system in place.” For a strong verification system was requested, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.A reported event can be grave and still require attribution.

The result is: The director general was stating a requirement for credible monitoring, not announcing that the system had been installed. The question here is a strong verification system was requested, and the available record reaches no further.

The stocks remained unaccounted for

The current file is specific: Grossi said Iran had not granted access to bombed sites and that the stocks remained unaccounted for. For the stocks remained unaccounted for, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

No arithmetic can create: The agency does not fill that gap through conjecture. The question here is the stocks remained unaccounted for, and the available record reaches no further.

Vance and Tehran’s denial

A U.S. promise of immediate inspections

The stated position is specific: Vice President JD Vance said on June 22 that IAEA inspections could start “as early as today,” according to the BBC. For a u.s. promise of immediate inspections, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.The record documents a problem. It does not license a guess.

The difference is: That was a stated possibility, not verified evidence of general access. The question here is a u.s. promise of immediate inspections, and the available record reaches no further.

Iran denied a new commitment

The documented gap is specific: Iran’s foreign ministry denied any “new commitment” on inspections, and Esmaeil Baqaei said there was no intention to permit access to facilities hit by U.S. and Israeli strikes. For iran denied a new commitment, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The fact remains: The opposing accounts cannot be merged into a confirmed reopening. The question here is iran denied a new commitment, and the available record reaches no further.

The British warning

The director general reported inability for a second time

The measurable part is specific: Britain’s June 2026 statement said the IAEA director general had again reported being unable to conduct verification activities because of lack of Iranian cooperation. For the director general reported inability for a second time, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.A condition is not an agreement, and a promise is not execution.

The threshold for proof: That statement belongs to the British government but points to a documented institutional impasse. The question here is the director general reported inability for a second time, and the available record reaches no further.

The duration was nearly a year

The remaining question is specific: London said the missing information or access had lasted nearly a year by then. For the duration was nearly a year, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The passage from report: Duration deepens the uncertainty; it does not tell us whether the inventory rose, fell, or moved. The question here is the duration was nearly a year, and the available record reaches no further.

The cost of restoring verification

The annual requirement is 10.4 million euros

The earlier precedent is specific: GOV/2025/50 estimates 10.4 million euros per year for implementing the Additional Protocol and verifying Iranian nuclear commitments. For the annual requirement is 10.4 million euros, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.Procedure protects facts from being turned into slogans.

The relevant restraint: The figure prices a possible verification effort rather than demonstrating that the effort is under way. The question here is the annual requirement is 10.4 million euros, and the available record reaches no further.

Part of the budget is extra-budgetary

The later development is specific: The report says 4.6 million euros of that amount would come from extra-budgetary contributions. For part of the budget is extra-budgetary, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The political cost: Funding can prepare an inspection system, but it cannot create consent to enter closed facilities. The question here is part of the budget is extra-budgetary, and the available record reaches no further.

What 60% does not prove

The level is an earlier reported estimate

The named source is specific: The reference to enrichment up to about 60% is relayed by Reuters from an earlier estimate, not a recent physical verification. For the level is an earlier reported estimate, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.What has not been verified must remain unverified in print.

The public should see: It must not be presented as an August 2026 measurement. The question here is the level is an earlier reported estimate, and the available record reaches no further.

No further capability claim is supported

The final check is specific: The assigned sources do not convert the reported enrichment level into a timetable, operational capability, or intention. For no further capability claim is supported, the source is inseparable from its date and stated scope; it cannot answer a larger question by itself.

The conclusion is: Anxiety does not license an invented calculation. The question here is no further capability claim is supported, and the available record reaches no further.

Conclusion

The number is not a lie, an estimate of today, or a measure of intent. It is an authoritative historical inventory with a date attached. Nearly every serious question begins after that date, where the accounting reports stopped, most inspections stopped, and the IAEA’s own director general demanded a strong verification system.

The next meaningful development would be a jointly confirmable act, not another unilateral description of events. Until then, the record requires disciplined language about what is known, who says it, and what remains outside independent confirmation. Facts do not become weaker when they are qualified. They become harder to misuse.

Sources

Primary sources

The primary material below preserves the official statements and institutional records used for the factual backbone.

These sources establish their own statements and documents; they do not erase the limits noted in the article.

• IAEA — GOV/2026/8 verification report — February 27, 2026
• IAEA — GOV/2025/50 safeguards report — 2025
• UK Government — IAEA Board statement — June 2026

Secondary sources

The reporting below provides the contemporaneous accounts, competing descriptions, and context attributed in the text.

No link is used to claim more than the described source supports.

• Reuters — First IAEA report since February strikes — June 4, 2026
• Reuters — Grossi on strong verification — June 26, 2026
• BBC — Dispute over inspector access — June 22, 2026
• Arms Control Association — IAEA investigations in Iran