CRIF – the global leader in credit and insurance information, analytics and solutions – has today launched its AI-powered fraud detection services in the UK, helping financial providers tackle tampering during customer onboarding.
Designed specifically for banks and insurers, the tool helps strengthen onboarding checks, identifying potentially fraudulent information before businesses are accepted onto financial providers’ systems.
The service utilises the latest innovations in Generative AI (GenAI) to analyse a wide range of documents including IDs, bills and bank statements to check for visual manipulation and issues with underlying metadata.
Fraudulent documents can be used to falsely alter a business’s appearance, influencing more favourable risk assessment outcomes and access to financial services they may not have otherwise qualified for. Manipulated documents come in an array of shapes and sizes, from concealing poor credit history to gain access to a loan, to disguising high-risk operating sectors to gain more favourable lending terms.
CRIF’s solution combats this by identifying potential document tampering, utilising advanced neural networks, large language models and deep domain expertise. This proprietary methodology uncovers subtle signals and traces, often invisible to the human eye, left by editing and manipulation activities generated through traditional software or AI-driven techniques. The solution also integrates specialised deepfake detection models to identify patterns typical of AI-generated content.
Currently, manual fraud checks and document reviews are highly resource-intensive, time-consuming and risk human error, consuming up to 5% of total operating costs for banks.
CRIF’s anti-fraud solution offers a clear traffic light system to indicate a business’s level of risk. While analysing large amounts of data at pace, it still allows for clear human oversight and final judgement, resulting in a quicker and more accurate onboarding process.
