Trolley Problem

The Trolley Problem is a thought experiment that asks you to imagine a runaway trolley heading toward five unsuspecting people. You can pull a lever, diverting the trolley onto another track where it will instead hit a single individual. The core of the problem lies in deciding whether it is permissible to actively cause harm to one person to prevent greater harm to many others.

This scenario matters because it isolates the tension between consequentialist reasoning—judging actions by their outcomes—and deontological principles, which focus on duties and rights regardless of results. By stripping away real‑world complexities, the Trolley Problem provides a clear laboratory for probing moral intuitions, testing ethical theories, and exposing hidden assumptions about responsibility and value.

Although originally a philosophical puzzle, the idea recurs in fields that must codify decision‑making under uncertainty: autonomous vehicle programming, medical triage protocols, military rules of engagement, and AI alignment research. In each case, designers confront choices that echo the trolley’s dilemma—balancing harms and benefits while grappling with whether to intervene directly or let events unfold.