Key takeaway: Data networks like Plaid embed fraud and ACH risk scoring directly into core lending and payments infrastructure. As a result, merchants and processors connected to that network should see faster fraud signals, but most firms still have real work ahead to meet Nacha’s new monitoring requirements.
Plaid has rolled out a set of new artificial intelligence models covering credit, fraud and payments decisioning, including a foundation model built specifically to flag fraud, the company said in a Tuesday (Oct. 6) press release.
A new feature called Instant Link lets eligible consumers share cash flow insights with lenders in seconds. It connects their financial accounts through Plaid’s Consumer Reporting Agency. This gives lenders a fuller repayment picture for borrowers who fall outside traditional credit scoring. Alongside it, Plaid introduced two credit models. The first is LendScore 2, which the company said predicts repayment ability 42% more strongly than traditional credit data alone. Then, there’s the LendScore Arc, a transformer-based model that learns from the order and timing of a borrower’s transactions.
On the fraud side, Plaid added a new AI foundation model trained on patterns across the Plaid Network to strengthen its fraud detection product, Protect. Additionally, a sequential foundation model now also powers Protect’s ACH risk model.
Plaid Credit Product Lead Michelle Young said in the release that the new models are meant to help lenders act on cash flow data they previously could not easily access. She added that the update gives lenders “new tools to expand access to more affordable credit.”
Why it matters: The launch arrives alongside a PYMNTS Intelligence and Plaid study, “Payment Protection: Why Firms Still Aren’t Real-Time Ready,” which found that 94% of surveyed firms are not yet fully compliance-ready for new Nacha rules — phased in between March and June 2026 — that require banks and businesses on the ACH network to actively monitor for fraud.
Source: PYMNTS