USDA to Test Satellites and AI to Improve Crop Estimates After Farmer Criticism

The U.S. Department of Agriculture is turning to satellites and artificial intelligence as it tries to repair confidence in its closely watched crop estimates.

Agriculture Secretary Brooke Rollins announced a pilot project at the Farm Progress Show in Boone, Iowa. The department will work with NASA and other federal agencies to assess whether improved satellite imagery can make reporting more accurate.

The plan also calls for geospatial tools and crop models to be used alongside farmer surveys when estimating planted acreage and yields. USDA said it would examine artificial intelligence and machine learning as part of the modernization effort.

The announcement follows sharp criticism from farmers, grain traders and economists after staffing cuts and unusually large changes between initial and final estimates of the corn acreage planted and harvested in 2025. USDA’s National Agricultural Statistics Service produces reports that can move commodity markets, affect farm-gate prices and influence federal farm programs. After final estimates were released in January, grain prices fell more than 5%, while growers were already struggling with low returns.

USDA said it wants to reduce the amount of information it repeatedly requests from farmers while making its statistics faster and more accurate. It also plans to give farmers more information about its methodologies, survey response rates and data limitations. The department began listening sessions in Minnesota on Aug. 4 and plans to continue them through the fall. The sessions have highlighted frustration over data reliability, staffing reductions and other Trump administration policies affecting agriculture.

“

We will conduct a pilot to evaluate the use of improved satellite imagery, working with NASA and other federal government agencies across the country to figure out how we can make this reporting more accurate.