Hear from Emily Sessa, Ph.D., Patricia K. Holmgren Director of the William and Lynda Steere Herbarium, on how the Garden is using a grant from the Bezos Earth Fund to revolutionize collections-based discovery.
Accelerating Species Discovery with AI
NYBG is honored to be selected as one of 15 global awardees in the Bezos Earth Fund’s AI for Climate and Nature Grand Challenge to harness the power of cutting-edge, responsible Artificial Intelligence (AI) to unlock the vast, underutilized data in herbarium collections. Herbaria are scientific collections of plant specimens that contain data about species’ distributions, traits, and environmental responses that are critical for conservation, restoration, ecology, and foundational biology. Using AI, we are building scalable, open-access tools that can transform biodiversity science and conservation planning by unlocking the extraordinary amount of data contained in herbaria and other biodiversity collections worldwide. This grant is part of NYBG’s cross-cutting initiative to responsibly utilize Artificial Intelligence to unlock the power of plants and fungi.
About the Project
Our project at NYBG will harness the power of cutting-edge, responsible AI to unlock the vast, underutilized data in herbarium collections. Herbaria are scientific collections of plant specimens that contain data about species’ distributions, traits, and environmental responses that are critical for conservation, restoration, ecology, and foundational biology.
Herbaria are the major frontier for discovery of new plant species and for understanding already known and threatened species. Over the next few years, we will develop advanced vision-language models and computer vision tools to unlock the extraordinary amount of data contained in herbarium specimens, at a scale and rate that would be impossible through manual methods. We will dramatically increase the quantity, quality, and availability of plant data.
Using AI to build scalable, open-access tools, will allow us to accelerate biodiversity discovery, support conservation actions for species and geographic areas, and inform environmental policy and decision-making faster and more effectively.
Our goals are
- To accelerate threat detection and improve habitat assessments for conservation and other environmental practitioners, particularly those advancing International Union for Conservation of Nature (IUCN) Red List assessments, Key Biodiversity Areas (KBAs), and Important Plant Areas (IPAs).
- To enable policymakers to make timely, evidence-based recommendations on land use, protected species and areas, and conservation investment; and
- To inspire new, high-quality research studies grounded in robust scientific evidence that can open new novel pathways for innovative ecological, evolutionary, and climate response studies.
Featured Publication
New Phytologist: “Herbariograph: a deep-learning tool to classify specimen images”
Fabio Andrés Ávila, John Y. Park, Leanna Feder, Damon P. Little, NYBG
Learn more about the digitization of herbaria utilizing AI tools.
Dive Deeper
New York Times: A.I. Program Aims to Break Barriers for Female Students
A new program, backed by Cornell Tech, M.I.T. and U.C.L.A., helps prepare lower-income, Latina and Black female computing majors for artificial intelligence careers.
