AI could help win ‘race against extinction’ of vital plants, say botanists
TL;DR
Royal Botanic Gardens, Kew says AI and digitisation can speed up work on threatened plants and fungi by helping scientists identify, compare and protect species faster. Kew has digitised all 7.4m of its own specimens, while 145m digital herbarium specimens are online globally. The report says that is still under 16% of collections. An AI model analysed 8m digitised specimens and found flowering times have shifted by an average 2.5 days per decade, with some regions moving by weeks.
Nauti's Take
This is one of the better AI use cases: no chatbot wrapper, just sorting, comparing and recognising patterns across huge scientific collections. The Kew report is optimistic and has a PR sheen, but the evidence is not vaporware: 7.4m Kew specimens digitised and 8m plant records analysed is real infrastructure.
The bottleneck is still fieldwork, curation and local capacity, especially in the global south. Without funding for collections, labs and local teams, AI becomes a fast scanner pointed at an incomplete shelf.
Briefingshow
The hard problem is speed: many species may disappear before scientists can name them or assess their risk. AI helps only if archives, local collections and biodiversity hotspots are digitised with care. Otherwise the models learn mainly from wealthy institutions and reinforce the blind spots they are meant to close.
Sources