Setting the stage for AI for biodiversity
summary
Global demand for finer, actionable biodiversity data drives use of scaled deep learning to collect (camera traps, eDNA, bioacoustics), integrate (GBIF >2 billion observations, iNaturalist now supplies roughly half of those), and support decisions (NatureAlpha packages data for finance). Immediate claims: AI can densify and forecast species maps via species distribution modeling and LLMs for scraping historical surveys. Big hurdles remain — severe geographic and taxonomic gaps, slow updates, and largely unknown uncertainty — but community pilots already show practical downstream value.
AI-written summary · about
chapters (9)
- 0:00ACM workshop welcome
- 2:28Keynote: Drew Purves
- 9:48Biodiversity and threats
- 16:25Biodiversity data gaps
- 23:25AI for ecological data
- 29:36Geotemporal modeling with AI
- 39:28LLMs and ecological data
- 44:44AI for biodiversity policy
- 47:00Programming languages for biodiversity
related talks
11mAI and Bioacoustics: Tracking Biodiversity Progress with BirdNET
05mBringing AI into the Wild
41mSpotlight on AI use cases and innovators around the world
11mUsing AI to track wildlife patterns
19mHow to Connect Data & AI with Dog Intelligence
01h 51mAI and Social Entrepreneurship
30mHow Technology Can Save Humanity
28mData Curation: Transforming Bytes into AI Gold