A drone attack on a major Russian oil refinery highlights the growing risk to global energy supplies from targeted strikes on critical infrastructure.
A drone attack on a major Russian oil refinery highlights the growing risk to global energy supplies from targeted strikes on critical infrastructure.

A Ukrainian drone attack has halted operations at Rosneft’s Syzran oil refinery, taking 170,000 barrels per day of processing capacity offline and underscoring the vulnerability of Russia’s energy infrastructure. The May 21 strike damaged the plant's primary crude distillation unit, which could take more than a month to repair, according to industry sources.
The business of predicting geopolitical risk is itself risky, with many firms failing to anticipate major disruptions. “You have to model your business plan around the worst risks actually being much more likely than you thought,” said Sarah Kaplan, a professor at the Rotman School of Management, commenting on the difficulty of forecasting events in the current global climate.
The attack forced the shutdown of the CDU-6 crude distillation unit, which accounts for over 70% of the plant's 8.5 million metric-ton annual capacity, sources told Reuters. In 2024, the refinery processed 4.3 million tons of crude, producing 1.5 million tons of diesel and 800,000 tons of gasoline. The loss of this output will add pressure to domestic Russian fuel markets and could have knock-on effects on global diesel prices if the outage is prolonged.
The incident is the latest in a series of targeted attacks on Russian energy facilities, representing a significant supply-side shock that complicates the calculus for global oil markets. For corporations and boards, it’s a stark reminder of the need to integrate geopolitical risk into their strategic planning. As Siemens executive Khalil Dindarian notes, companies must “learn quickly and adapt quickly” in an environment where crises can compound, creating exponential uncertainty.
The strike on the Syzran refinery is part of a broader trend where non-state and state actors are increasingly targeting critical economic infrastructure. This shift is forcing a reassessment of risk management, moving it from a purely defensive, linear process to a more dynamic and adaptive one. “Traditional risk management is built on linear processes,” Dindarian argues, but in a world of cascading crises, “you have to accept uncertainty as part of every decision.”
This new reality has spurred a boom in geopolitical crisis consulting, with firms like Crisis24, a division of GardaWorld, partnering with data-analytics giants like Palantir to offer clients “actionable foresight.” Yet, even sophisticated AI-driven models can miss the mark. Crisis24’s own 2026 forecast, for instance, deemed a blockage of the Strait of Hormuz an “unlikely worst-case,” a scenario that has since unfolded. This highlights the challenge of modeling risk when unpredictable actors are involved. The attacks on Russian refineries show that previously unthinkable scenarios are now part of the standard operational landscape, forcing the energy industry and its insurers to price in a new, higher level of geopolitical risk.
This article is for informational purposes only and does not constitute investment advice.