Answers are grounded in this paper and cite their sources.
The increasing use of digitized financial services has significantly enhanced efficiency and convenience of online transactions. Conversely, the situation tends to simultaneously create opportunities for criminals to commit financial fraud. This study sought to examine application of AI in detection of online financial fraudulent activities in Kenya using machine learning techniques. Specific objectives comprised: to identify common patterns associated with online financial fraud in Kenya, assess the use of machine learning techniques in fraud detection, examine the effectiveness of AI in detecting online financial fraud, and recommend strategies for improving AI-driven fraud detection systems. Descriptive cross-sectional research design was used, allowing application of both quantitative and qualitative data. Target population included 54 respondents comprising fraud analysts, cybersecurity professionals, information technology officers, risk and compliance officers, and mobile money managers. Out of these respondents, 41 provided usable responses, accounting for a 75.9% response rate. Data was analyzed using frequencies, percentages, means, standard deviations, and one-sample t-tests against a neutral midpoint of 3.00. Phishing, social engineering, account takeover, identity theft, and exploitation of mobile money and digital payment platforms were found to be prominent fraud patterns. The findings further showed that although real-time integration and frequent model retraining remained comparatively underdeveloped, machine learning was reportedly being utilized more often to detect suspicious transactions and unusual customer behaviour. More than 70% agreed that AI-based fraud detection was successful, especially when it came to increasing detection speed and accuracy, identifying new fraud trends, and preventing fraud in general. The study concluded that AI and machine learning provide significant potential for strengthening online financial fraud detection in Kenya. It recommended enhanced investment in real-time AI systems, continuous model retraining, improved data infrastructure, staff capacity development, human-AI collaboration. Finally, the study recommended stronger regulatory and governance frameworks for responsible AI utilization.
No takes yet. Share an insight, caveat, or question.
Sasha Moraa Siika (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: