Satellites can reveal global extent of forced labor in the world’s fishing fleet

Significance Forced labor in fisheries is increasingly recognized as a human rights crisis. Until recently, its extent was poorly understood and no tools existed for systematically detecting forced labor risk on individual fishing vessels on a global scale. Here we use satellite data and machine learning to identify these high-risk vessels and find widespread risk of forced labor in the world’s fishing fleet. This information provides new opportunities for unique market, enforcement, and policy interventions. This also provides a proof of concept for how remotely sensed dynamic individual behavior can be used to infer forced labor abuses.