This research paper introduces a novel neuro-symbolic framework designed to enhance accountability in public-sector AI systems. The framework links AI-generated decision justifications to specific legal rules, using CalFresh, California's food assistance program, as a case study. It combines a structured ontology of eligibility requirements with a rule extraction pipeline and a solver-based reasoning layer to identify and correct legally inconsistent explanations, thereby supporting procedural accountability. AI
IMPACT Enhances transparency and contestability of AI decisions in public services, potentially setting a precedent for other government applications.
RANK_REASON Academic paper on a novel AI framework for accountability. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →