Artificial Intelligence in Financial Analytics: An Integrative Review of Financial Modeling, Forecasting, and Financial Controls

Abstract

Artificial intelligence (AI) is increasingly applied across finance, including fraud detection, risk assessment, and forecasting, yet forecasting and internal controls are often governed in separate initiatives. This paper examines two pillars of AI use in finance: AI-driven financial modeling and forecasting, and AI-driven reconciliation within internal control over financial reporting (ICFR). The paper is based on a literature review of 23 academic, regulatory, and standards sources, including materials from the SEC, PCAOB, and COSO. The review indicates that the two pillars are interdependent: reconciled, well-governed data underpins reliable forecasting, and continuous reconciliation combined with predictive modeling can support pre-filing risk detection in SEC reporting. Model risk management, explainability, and human sign-off emerge as governance requirements common to both pillars. Building on these insights, the paper proposes a five-stage maturity framework that sequences AI-driven reconciliation ahead of predictive forecasting, together with shared governance practices. The findings suggest that managing forecasting and controls under a unified governance approach offers a coherent pathway for AI adoption in finance. JEL classification numbers: C53, G17, G32, M41, M42. Keywords: Artificial Intelligence in Finance, Financial Analytics, Automated Reconciliation, Internal Controls, AI-Driven Finance.