Darwinium Scam Detection is a behavioral intelligence platform designed to identify and prevent social engineering-driven fraud, including authorized push payment (APP) fraud, purchase scams, and travel scams. The platform monitors user behavior across complete digital journeys — from pre-login activity through to payment — to detect anomalies that may indicate coercion, social engineering, or fraudulent intent. Rather than relying solely on identity signals, it focuses on behavioral context and user intent. Key detection capabilities include: - Financial scam detection: Identifies APP fraud linked to investment, romance, and impersonation scams, including coercion signals, payment anomalies, and remote access tool presence. - Purchase scam detection: Uses text and image similarity analysis to expose fake listings, flags sellers using proxies or multiple accounts, and identifies buyers bypassing platform rules. - Travel scam detection: Flags suspicious listings, fake reviews, linked buyer/seller accounts, and anomalous booking patterns. The platform uses behavioral biometrics (e.g., hesitation, dwell time, page delays) and mobile sensor data (orientation, accelerometer) to detect live calls and signs of coaching through a payment journey. It also performs payer/payee risk assessments using historical behavioral analytics, transaction velocities, and beneficiary intelligence. When deployed via CDN on the edge, Darwinium can inject dynamic, tailored warning messages into the customer journey in real time to disrupt scam payments and introduce friction for high-risk transactions. Third-party data can be ingested via API to augment risk decisions, including call centre data, phone intelligence, and remote access session detection.