Zest AI
Machine learning credit underwriting that helps lenders approve more borrowers without more risk.
About
Zest AI sells AI credit models to lenders, mostly credit unions, community banks, and fintechs. Where traditional credit scores rely on 15 to 20 variables, Zest's models analyze thousands, and the company says lenders get on average 25% more approvals with no added risk and 20% fewer defaults at constant approval rates. Every model ships with explainability tooling meant to satisfy fair-lending compliance, and the platform also includes Zest Protect for fraud detection and LuLu, a generative-AI lending intelligence layer for benchmarking against peers. The company was founded in 2009, originally as ZestFinance, and says nearly 300 lenders run more than 650 of its proprietary models, backed by 50+ patents.
In November 2025 it closed an oversubscribed customer-led financing round with investors including SchoolsFirst Federal Credit Union, Members 1st Federal Credit Union, ORNL FCU, Truliant FCU, and Citi, which the company said marked a significant valuation increase over its $200 million round in 2024. Having customers put money in is a strong signal in a sector where trust is everything. There is no public pricing and models require high-quality lender data to perform, so this is enterprise procurement, not a self-serve product.
What it can do
- ML credit underwriting models analyzing thousands of variables vs 15-20 in traditional scores
- Explainable AI outputs for fair-lending and regulatory compliance
- Automated underwriting integrated into loan origination workflows
- LuLu generative-AI lending intelligence platform with peer benchmarking
- Zest Protect for lending fraud detection
- 50+ patents; 650+ proprietary models deployed
- Fairness-aware modeling aimed at expanding inclusive approvals
Setup
- Contact Zest AI for an enterprise evaluation of your lending portfolio.
- Provide historical loan performance data for model training and validation.
- Work with their team to build and validate custom underwriting models with explainability reports.
- Integrate the models into your loan origination and decisioning systems.
- Monitor model performance and fair-lending compliance with ongoing reporting.
Prerequisites
A lending institution with historical loan data; enterprise contract with Zest AI.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: customer-led financing round led by SchoolsFirst, Members 1st, ORNL, Truliant, and Citi (Finextra, Business Wire, Nov 2025); ~300 lenders and 650+ models with 50+ patents (company-reported); founded 2009 with decade-plus track record. Strengths: customer-backed financing is a strong trust signal, real lender testimonials, long operating history. Docked for: (1) the 25%-more-approvals and 20%-fewer-defaults figures are vendor-stated averages; (2) no public pricing or open benchmarks. License: proprietary.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.0/10
- Maintenance · recency, releases, issue load
- 9.0/10
- Openness · license + docs
- 5.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · Customer-led financing round led by SchoolsFirst, Members 1st, ORNL, Truliant, Citi (Finextra, Business Wire, Nov 2025)
- · Founded 2009 (as ZestFinance); ~300 lenders, 650+ proprietary models, 50+ patents (company-reported)
- · Claims 25% more approvals at no added risk and 20% fewer defaults - vendor-stated averages
- · No public pricing; models require high-quality lender data; enterprise procurement only
- · Customer-backed financing is a strong trust signal in a trust-driven sector
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.