Large language models unlock the ecology of species interactions.

Species interactions shape population dynamics, geographic distributions, evolutionary trajectories, and responses to environmental change. Yet data on these interactions remain scarce across broad spatial, temporal, and taxonomic scales because they are difficult to collect in the field. One promising source of interaction data is citizen science platforms, which contain billions of biodiversity observations, often accompanied by unstructured text comments that may document interactions among organisms. Advances in large language models (LLMs) make it increasingly feasible to identify, extract, and categorize biotic interactions from these unstructured data at scale. Here, we present an LLM workflow that collects species interaction observations from multilingual citizen science comments. Using two case studies-bird-bird and plant-pollinator interactions-we show that LLMs can rapidly extract interaction types and participating species with high accuracy. These data can greatly expand the spatial, temporal, and taxonomic coverage and resolution of species interactions data, enable new tests of long-standing ecological questions, and improve our ability to track ecological changes. With appropriate validation, expert review, and attention to data privacy for both users and sensitive species, this approach opens new opportunities to characterize, forecast, and conserve biodiversity under global change.