The Value of Using A Novel Decision Support Risk Assessment Tool When Investigating Internal Theft in a Gaming Establishment

—A Case Study

Abstract

The average US gaming establishment with at least 1,000 employees loses roughly $3.3 million annually as a result of employee fraud.  Current internal theft identification strategies and methodologies are neither quick nor sufficient enough to curtail the financial damage, as it typically takes 12-16 months to detect and resolve most internal theft issues.  The current applied research case study examines a novel artificial intelligence (AI)-enabled voice screening tool and describes its capability in identifying previously undetected internal theft knowledge and involvement within a particular employee population. Compellingly, 95.5% of risk-flagged interviews were confirmed and verified.  In the total sample of 99 consenting study volunteers with no known history of internal theft, 0% initially admitted to a fraudulent offense.  During the automated telephone interview, 3% of the entire subject pool provided a total of four admissions to either knowing about or being involved in internal theft.  During the posttest follow-up interview, 16.2% of the subject pool made additional knowledge and/or involvement-based disclosures, germane to theft onsite.  Critically, as a result of the automated interview process, previously undiscovered internal theft details were identified.  Finally, this study’s description of specific characteristics and predictive qualities of high-risk employees should benefit gaming establishment security and investigative teams alike.