LO
DeploymentCognitive automation2025-10-22
The US Geological Survey reports producing its National Land Cover Database with three linked deep learning models, delivering the map faster and more cheaply
GIS analyst / cartographeroccupation page →
Event date / reported
2025-10-22
Evidence stage
DeploymentAn employer has put it into production. Can move the baseline — weighted by scale and how similar the setting is.
Tasks this bears on
Spatial databases and map production
Maintaining the spatial data and designing the maps and reports people use.
Being augmented✓ Evidence-backed
Where this applies
United States. The Department of the Interior's own AI use-case inventory lists the USGS National Land Cover Database as deployed and operational since 22 October 2025 and developed in-house; it says the database is among the most widely used datasets the department produces, that revamping its methodology with a series of three linked deep learning models has improved the product, reduced the latency of delivery and produced it more cheaply, that its output is maps of land cover change over time, and that accuracy is assessed with high-resolution data and manual interpretation by land cover experts. One national dataset; it says nothing about how much analyst time changed.
What this means
A national mapping agency now builds and updates a flagship map database with deep learning in production.
What it does not yet show
One dataset, with no measure of analysts' time or of human review.
What you can check
Open the Department of the Interior's 2025 AI use-case inventory (individual use cases, CSV) and find "a series of 3 linked deep learning models".
Does it change the assessment?
No. The impact index is never moved by a single event. What this record did: the 1 linked task judgement above now rest on evidence instead of inference.
Source
U.S. Department of the Interior — 2025 AI Use Case Inventory, individual use cases (CSV; entry DOI-0066, National Land Cover Database, U.S. Geological Survey) · verified 2026-10-09 · Claude (VOLO agent) · interpreted 2026-10-09 · Claude (VOLO agent)
Primary source — published by the party that did this, or the authority of record. No co-signature needed.