The scale and complexity of U.S. federal financial management is growing to a level that is increasingly testing the conventional oversight, audit, and accountability mechanisms. Even though artificial intelligence and sophisticated business analytics have become an appealing tool to deal with fraud, improper payments, and performance monitoring, their implementation in federal programs exceeded the creation of governance frameworks to guarantee transparency, auditability, and democratic accountability. The current literature on AI in government finance is disconnected, as technical literature focuses on analytical performance and policy literature on governance principles, but with limited integration between public finance accounting and audit systems. This study addresses this gap through an integrative review of peer-reviewed research, federal oversight reports, and policy frameworks published between 2020 and 2025. The review integrates insights from public finance, audit practice, and AI governance to examine how AI-driven business analytics are currently applied in U.S. federal financial management and where accountability failures persist. The literature indicates that AI initiatives strengthen oversight only when embedded within existing financial control systems and supported by robust data governance, explainability mechanisms, auditable model outputs, and clearly defined human oversight. Analytics deployed as standalone technical solutions frequently increase opacity and weaken audit assurance. Based on these insights, the paper proposes a multi-layered AI-driven business analytics framework that integrates technical capability with federal accounting standards, audit requirements, and governance structures, repositioning AI as a governance instrument rather than a purely technological tool.