Plaid Expands Credit and Payment Risk Models
Start here
AI-assisted source synthesis · not an independent investigation.
At 21:00 on October 6, 2026, PYMNTS reported that Plaid launched Instant Link, LendScore 2, and LendScore Arc, enabling financial institutions to use cash flow data and transaction sequences to support credit decisions, fraud detection, and payment risk decisions. A follow-up report at 23:32 that day cited data showing that LendScore had an approval rate of 73%, higher than the traditional benchmark of 65%; at the same approval rate, risk was 41% lower. Personify Financial said that verified bank transaction data increased approval rates for underserved consumers by up to 8%; Affirm said its latest underwriting model increased completed purchases by 3.4% at comparable risk. On October 9, Payments Cards & Mobile reported that Plaid was expanding AI into credit, fraud detection, and ACH payment risk, and said that LendScore 2 was 42% more predictive of repayment than traditional credit data used alone; in tests, LendScore Arc improved predictive performance for deep-subprime and superprime borrowers by 20% and 24%, respectively.
AI-written from reports · updated 12 days ago
Developments
- Oct 6 23:32 · 1 reportsLenders discover high-quality borrowers overlooked by traditional credit assessments in the dataPYMNTS：Lenders find high-quality borrowers missed by traditional credit assessments
- Oct 6 21:00 · 2 reportsPlaid launches credit and anti-fraud modelsPYMNTS：Plaid launches credit and fraud models based on cash flow data
Timeline
Follow the reporting to see the different sides of the story.
- Payments Cards & MobilePlaid expands AI to credit, fraud detection, and ACH payment risk
Plaid launched a cash flow underwriting tool, as well as AI foundation models for fraud detection and ACH payment risk that analyze long-term financial behavior rather than isolated data points. Its LendScore 2 claims 42% greater predictive power for repayment than traditional credit data used alone, while LendScore Arc improved prediction performance for deep-subprime and superprime borrowers by 20% and 24%, respectively, in testing.
- PYMNTSLenders find high-quality borrowers missed by traditional credit assessments
Bank transactions, cash flow and detailed credit histories are helping lenders supplement traditional credit scores; Personify Financial says verified bank transaction data increased approval rates for underserved consumers by up to 8%. Affirm’s latest underwriting model increased completed purchases by 3.4% at comparable risk, while Plaid’s LendScore had an approval rate of 73%, above the traditional benchmark of 65%, with 41% lower risk at the same approval rate.
- PYMNTSPlaid launches credit and fraud models based on cash flow data
Plaid has launched models including Instant Link, LendScore 2, and LendScore Arc, providing financial institutions with credit, fraud, and payment risk decisioning capabilities based on cash flow and transaction sequences.