Kenneth Glad
The closest brushes with nuclear war rarely looked dramatic at the time. No explosions, no public warnings. Just quiet rooms, flickering screens and people forced to decide, quickly, whether the data in front of them was real. Only years later did the scale of those moments become clear.
Shortly after midnight in September 1983, Soviet officer Stanislav Petrov was on duty when warning systems indicated a US missile launch. Procedure left little room for interpretation.
He paused.
Petrov later said the alert did not “feel right.” The system was reporting only a handful of missiles, an unlikely opening for a full-scale strike, writes the Daily Express.
He chose not to escalate. Investigations confirmed a sensor error caused by sunlight reflecting off clouds, a detail preserved in Cold War archives.
In January 1995, something similar unfolded at sea. Russian radar detected a fast-rising object over the North Atlantic. For a few tense minutes, officials feared a high-altitude nuclear strike designed to disable communications. President Boris Yeltsin was alerted before analysts concluded it was a Norwegian research rocket.
When systems failed
Cold War warning networks were built for speed. Accuracy was sometimes secondary.
In 1979, a US training scenario was mistakenly fed into live systems, briefly convincing operators that a large-scale attack was underway. Cross-checks with radar and satellites told a different story.
A year later, a faulty microchip triggered repeated false alerts. The warnings were serious enough to wake senior officials in the early hours, unsure whether the readings were real.
Declassified Pentagon records also point to a 1967 solar storm that disrupted radar coverage. At first, the interference looked like deliberate jamming, raising fears of an incoming strike before scientists identified the cause.
And then there are the incidents that barely made headlines. In 1961, a US B-52 bomber broke apart over North Carolina, dropping two nuclear bombs. One came close to detonating after several safety mechanisms failed, according to later investigations.
Pressure and miscalculation in crises
Not every close call came from machines. Some developed in the fog of geopolitical tension, where information was incomplete and time was short.
During the Cuban Missile Crisis, a Soviet submarine, cut off from Moscow and under pressure from US depth charges, prepared to launch a nuclear torpedo. Officer Vasili Arkhipov refused to approve it.
That decision likely prevented escalation.
Other crises followed a similar pattern. In 1973, amid the Yom Kippur War, nuclear forces were quietly readied. Border clashes between China and the Soviet Union in 1969, and later confrontations between India and Pakistan, showed how quickly regional conflicts could spiral.
What links these episodes is not just proximity to disaster, but the conditions behind them: Fragmented information, technical limits and intense time pressure.
Today, analysts warn those pressures have not disappeared. Cyber threats, automated systems and faster decision cycles may reduce the space for hesitation even further.
Next time, there may be no pause.
Sources: Daily Express
