A group of Democratic members of Congress from Florida is demanding answers from state officials regarding a rapid push to deploy artificial intelligence, or AI, to reduce errors in the state's food assistance program.
In a letter sent Monday to Florida Department of Children and Families Deputy Secretary Kathryn Williams and Gov. Ron DeSantis, seven lawmakers expressed reservations over a $4 million line item in this year's state budget. The provision requires DCF to find an AI vendor by Sept. 1 to carry out data analytics and machine learning on Supplemental Nutrition Assistance Program, or SNAP, eligibility determinations.
About 3 million people in Florida — about 13% of the state population — receive SNAP benefits.
The move by state officials stems from President Donald Trump's signature domestic policy bill, the One Big Beautiful Bill Act. It requires states for the first time to pay a portion of the cost of the historically federally funded benefits based on their error rate.
Under the law, states with SNAP payment error rates above 10% will be forced to cover 15% of total benefit costs, beginning in October 2027. Florida's current SNAP error rate stands at 12.97%. It could cost the state $1 billion.
The lawmakers say they are concerned with using "new and relatively unproven technology of AI to identify SNAP errors."
The lawmakers — Reps. Maxwell Alejandro Frost, Kathy Castor, Jared Moskowitz, Lois Frankel, Darren Soto, Debbie Wasserman Schultz, and Frederica S. Wilson — have given state officials until Aug. 10 to respond to their questions.
"The state is moving on an extraordinarily accelerated timeline to procure an AI vendor by September 1, yet Floridians have received little information about how that vendor will be selected or what safeguards will be in place," said the lawmakers.
This is a News In Brief report. Visit WLRN News for in-depth reporting from South Florida and Florida news.
Copyright 2026 WLRN
