Integrating AI into integrity monitoring

Download PDF

Artificial Intelligence (AI), including Generative AI (GenAI) and Agentic AI, is transforming the landscape of integrity monitoring by enhancing analytical precision, automating repetitive tasks, and improving decision-making transparency. Integrating these AI capabilities into integrity monitoring workflows can significantly improve the efficiency and effectiveness of fraud, waste, and abuse (FWA) detection, prevention, and response. Download our paper for a deep dive into how these technologies can be strategically applied across the entire monitorship lifecycle—from risk identification to reporting—to drive trust and accountability.

AI across the integrity monitoring lifecycle

Risk identification: By leveraging pattern recognition, natural language processing, and predictive modeling, AI systems can rapidly analyze vast volumes of structured and unstructured data at speed to identify potential FWA risks early in a project’s lifecycle.

Risk assessments: AI can streamline the risk assessment process by automating data aggregation, providing dynamic, real-time risk scoring, and generating hypothetical scenarios to test controls, which enables integrity monitors to shift their focus from manual data compilation to interpreting and acting on AI-driven insights.

Identification of high-risk areas: AI systems enhance monitoring efficiency by holistically evaluating project data through network analysis to uncover collusion or conflicts of interest, process mining to identify procedural deviations, and comparative benchmarking to surface anomalies, thereby directing resources to the most critical risk areas.

Actionable field observations: AI can convert qualitative field observations into actionable intelligence by automatically summarizing unstructured data, using natural language processing to extract recurring themes and risks from various reports, and creating knowledge graphs that link observations to specific risk categories.

Monitoring plans and reports: AI can streamline the creation and management of monitoring plans and reports by using GenAI to draft plans based on project scope and risks, employing Agentic AI to dynamically track progress and update plans in real-time, and automating the generation of clear, auditable integrity reports, which reduces administrative burden and enhances the timeliness of reporting to stakeholders.

Invoice review and contract compliance: AI can improve the accuracy and speed of invoice review and contract compliance by using GenAI to extract key contract terms, Agentic AI to cross-reference invoices against contractual obligations and supplemental documentation to identify discrepancies, and machine learning models to flag anomalous billing patterns, allowing monitors to shift their focus from manual reconciliation to investigating substantive issues.

KPMG helps clients integrate AI into integrity monitoring by combining our deep experience in compliance, forensic accounting, and regulatory oversight with tested methodologies and a structured implementation approach. Our approach helps ensure AI tools are used ethically and effectively, giving clients confidence through a process grounded in human oversight, robust model governance, and client assurance.