Legal Challenges of Using AI and Big Data in Public Administration: Administrative Liability, Data Protection, and Public Services Efficiency

This study investigates the evolving legal challenges posed by the integration of artificial intelligence and big data in public administration.Through a mixed-method approach, combining doctrinal legal analysis, case study review, and empirical dataset evaluation, it examines how current laws respond to emerging risks in administrative liability, data protection, and service automation.A novel Legal Risk Index (LRI) was developed to quantify regulatory sensitivity across jurisdictions and application domains, revealing that systems in welfare fraud detection and biometric surveillance face the highest legal scrutiny, with LRI scores reaching critical thresholds in over 70% of examined cases.The study analyzed a dataset of 140+ public sector AI deployments across Europe, offering a concrete empirical base.The paper compares AI governance strategies in the EU, UK, US, and MENA, highlighting disparities in oversight, enforcement, and public accountability.The findings show that legal maturity, not just technological advancement, is key to responsible deployment.Major contributions include the introduction of a cross-jurisdictional legal risk framework, evidencebased policy recommendations, and a structured model for liability allocation in AI-driven decisions.This research offers practical insights for regulators and public institutions seeking to balance innovation with rights-based governance.