Plaid has introduced a suite of new credit and fraud detection models designed to leverage real-time cash flow data. The Instant Link tool allows consumers to share financial account insights with lenders via the Plaid Consumer Reporting Agency, providing a more comprehensive view of repayment capacity for those outside traditional credit scoring.
The company released LendScore 2, which utilizes cash flow underwriting to predict loan repayment ability with 42% greater accuracy than traditional data. Specialized versions of this model are now available for auto, home, and short-term lending sectors. Additionally, the new LendScore Arc model employs transformer-based technology to analyze the sequence and timing of borrower transactions.
For fraud prevention, Plaid launched an AI foundation model trained on patterns from the Plaid Network. This model enhances the existing Protect solution and now powers the company’s ACH risk assessment capabilities. These tools aim to address the growing demand for real-time financial behavior analysis in lending and payment processing.
The updates arrive as firms face pressure to comply with new Nacha regulations regarding active fraud monitoring. According to research from PYMNTS Intelligence and Plaid, 94% of surveyed firms are not yet fully prepared for these requirements. Plaid executives noted that these models provide the financial context necessary to improve decision-making for both lenders and consumers.
