An AI-Based Security Architecture for Fraud Detection in Cloud Call Centers for Low-Resource Languages: Arabic as a Use Case

Cloud-based telephony platforms face growing fraud risks including voice phishing (vishing), subscription abuse, and organizational impersonation, with detection being especially challenging in low-resource languages such as Arabic. This paper presents an Artificial Intelligence (AI)-based security architecture for fraud detection in Arabic cloud call centers, integrating onboarding verification, behavioral monitoring, domain-adapted Automatic Speech Recognition (ASR), semantic transcript search, and Large Language Model (LLM)-based entity verification. The domain-adapted Langa ASR model achieves a Word Error Rate (WER) of 41.0% and Character Error Rate (CER) of 18.2%, outperforming all evaluated commercial baselines. LLM-based entity extraction with multi-call consensus achieves 97.3% company-name accuracy (Generative Pre-trained Transformer 4, GPT-4) and 92.0% in the cost-effective deployed configuration (GPT-3.5 with log-probability filtering). Evaluated on production data from a Middle East and North Africa (MENA)-region provider spanning more than 1,000 accounts, the pipeline flagged 47 accounts of which 41 were confirmed fraudulent (precision 87.2%, 95% confidence interval (CI): 74.3%–95.2%; estimated recall 51%–82%), demonstrating the viability of a unified, threat-model-driven architecture for low-resource telephony fraud detection.