Uneven research attention to biodiversity loss revealed by large language models

Large language model–driven mapping of 2.4 million bibliographic records found 253,081 biodiversity-loss articles, with chordates receiving 34.8% of attention despite only 4.6% of described species.

• Why it matters: Biodiversity loss threatens ecosystem functioning and human well-being, but existing syntheses cover only parts of the evidence base, leaving uneven research attention across ecological contexts.
• What they did: The team built a systematic mapping framework using large language models to screen and synthesize the literature, classifying articles by geography, ecosystems, taxa, direct drivers, and specific threats.
• The result: The map reveals comparatively limited attention to non-chordate animals, studies in Africa, and direct exploitation in marine systems, helping identify evidence gaps for future research priorities.