Report: AI Use Rises, But So Do Fraud Teams

- The global market for AI in fraud detection is projected to grow from $15.6 billion in 2025 to $119.9 billion by 2034, expanding at a compound annual growth rate of 25.4%. This growth is driven by the increasing digitization of financial services and the need for more sophisticated, automated security systems. - In India, digital financial frauds are a significant and growing problem, with 2.4 million incidents resulting in losses of ₹4,245 crore in the first ten months of the 2025 fiscal year. For every rupee lost to fraud, Indian businesses incur an additional cost of ₹4.00, which includes investigation, recovery, and other external fees. - The rise of generative AI has led to new fraud tactics, including AI-assisted document forgery and highly convincing phishing attacks. Phishing attempts linked to generative AI surged by 1,265% in the past year, and 60% of email recipients now fall for these sophisticated scams. - In response to rising fraud, 99% of financial crime leaders acknowledge flaws in their detection abilities, citing siloed data and a lack of real-time visibility as key issues. This is a primary reason why, despite 98% AI adoption, teams and budgets are still growing to manage the complexity and fragmentation of security systems. - For WhatsApp-based businesses in India, AI chatbots are crucial for scaling customer interactions, with some D2C brands reporting that 25-40% of support queries are fully resolved by bots. Vernacular language chatbots have been shown to increase qualified leads from Tier 2 cities by as much as 40%. - In the hyperlocal commerce sector, Cash-on-Delivery (COD) fraud is a major challenge, leading to inflated return-to-origin (RTO) rates. Fraudsters often place multiple rapid orders from new accounts or use recycled phone numbers, patterns that AI systems are being trained to detect before dispatch. - The Unified Payments Interface (UPI) system in India is a major target for fraud, with cases increasing by 85% in 2023-24. Modern fraud detection for UPI now integrates machine learning to analyze transaction parameters like frequency, geolocation, and device characteristics in real-time to identify suspicious activity. - Looking ahead, security leaders are shifting from static, rules-based fraud detection to "continuous behavioral intelligence." This involves using AI to model normal user and device behavior in real-time to spot subtle anomalies, reducing false positives and improving the customer experience.