Year: 2024
Type: article
Abstract: Insider threats are one of the longstanding and most challenging issues in securing financial institutions, with employees, contractors, and trusted partners turning legitimate access into a liability for sensitive systems and data. Traditional security methods are usually unable to identify subtle malicious activities or negligent behaviors until after much damage has been done. Artificial intelligence (AI) provides transformative opportunities in this field with proactive detection via sophisticated behavioral analytics, anomaly detection, and hybrid modeling. This paper delves into the changing face of insider threats in financial institutions and discusses the role AI can play in strengthening detection capabilities while keeping up with regulatory frameworks. Approaches to the collection and analysis of heterogeneous data sources (from transaction logs to communication patterns) are described, as is the performance of supervised, unsupervised, and deep learning models. The results show both the potential of AI in uncovering hidden insider risks as well as the challenges involved with false positives, privacy and governance. Finally, this paper attends to the demand for AI systems to be explainable and robust governance systems to ensure that insider threat detection is adopted successfully and ethically for high-stakes financial use cases.
DOI: 10.64206/8eakz491
Author: Olatunji Olusola Ogundipe
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7079696283
FWCI:
Cites:
Cited by:
Related to: