Generating scenarios for extreme events, without extreme data

Researchers at MIT have developed a novel algorithm designed to simulate 'black swan' events, even when historical data is scarce. Traditional AI models often fail to predict rare but catastrophic disruptions because they rely on patterns found in common datasets. This new approach bridges that gap by effectively generating high-risk, unprecedented scenarios. By stress-testing critical infrastructure and supply chain logistics against these synthetic extremes, companies can better fortify their operations against global shocks before they actually occur.