Generative AI Transaction Simulation for Accurate Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing transaction computing systems face challenges in efficiently detecting fraud and errors without burdening users with high bandwidth requirements and inconvenient verification processes, often incorrectly flagging legitimate transactions as fraudulent.
Innovation Solution
Utilizing a generative AI model to simulate transactions and compare simulated parameters with submitted transaction parameters, determining legitimacy values to accurately identify fraud and errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fraud detection methods are used, then fraud detection capability is provided, but user convenience deteriorates due to high bandwidth requirements and burdensome verification processes
Solution Approach 1:
The system creates simulated transactions that replicate the structure and parameters of actual transactions. These simulated transactions are then analyzed to generate legitimacy values without requiring users to provide additional verification data or bandwidth, thus maintaining fraud detection reliability while improving user convenience
Solution Approach 2:
The system introduces an intermediary analysis layer that processes transactions through simulation and comparison mechanisms. This intermediary layer generates legitimacy values by comparing simulated transaction parameters against actual transaction parameters, enabling fraud detection without direct user intervention or additional bandwidth consumption
2Reliability
If traditional fraud detection methods are used, then fraud detection capability is provided, but transaction efficiency deteriorates due to incorrect flagging of legitimate transactions
Solution Approach 1:
The system implements a feedback mechanism where simulated transactions are continuously compared against actual transaction parameters. This feedback loop allows the system to adjust and refine its detection accuracy, reducing false positives and improving transaction efficiency by allowing legitimate transactions to proceed without unnecessary delays
Solution Approach 2:
The system changes the approach from direct transaction validation to simulated transaction parameter comparison. By generating legitimacy values through parameter comparison between simulated and actual transactions, the system achieves more accurate fraud detection that reduces incorrect flagging and improves overall transaction efficiency
Data Source
AI summary
A provider computing system includes a processing circuit having at least one processor coupled to at least one memory device and at least one artificial intelligence (AI) system. The processing circuit performs operations including receiving a first request for a first transaction having one or more first parameters; analyzing a transaction history comprising one or more previous transactions having at least one of the one or more first parameters; determining a response to the first request; and transmitting the response to the first request. The at least one AI system is configured to perform operations including: simulating one or more transactions; identifying one or more second parameters of the one or more simulated transactions; comparing the one or more second parameters to the one or more first parameters; and determining a legitimacy value associated with the first transaction based on the comparison.