Plaid Launches Credit and Fraud Models Fueled by Cash Flow Data

AI summary of the source article

Plaid has launched several new AI models utilizing cash flow data to aid lenders and financial firms across credit, fraud, and payments. New additions include Instant Link, which connects consumer accounts to Plaid Consumer Reporting Agency, and LendScore 2, which claims to predict repayment ability with 42% greater strength than traditional credit data alone. The firm also debuted LendScore Arc, a transformer-based model analyzing transaction order and timing, along with specialized models for auto, home, and short-term loans. Additionally, Plaid announced a foundation model trained on network data to bolster its fraud and ACH risk solution, Protect.

Why it matters

The new cash flow and transformer-based models aim to help lenders assess borrowers who fall outside traditional credit metrics and assist businesses in monitoring ACH transaction fraud.

Key facts

• Plaid introduced LendScore 2, which predicts a borrower's repayment ability with 42% greater strength than traditional credit data alone.
• LendScore Arc is Plaid's transformer-based credit risk model that evaluates the order and timing of transactions.
• A new AI foundation model trained on Plaid Network data has been deployed to enhance Plaid's Protect fraud and ACH risk detection solution.