Quantifying Discrimination That Legacy Methods Miss
aequumAI builds the compliance infrastructure for institutions, regulators, and legal teams that need to measure protected-class disparities when self-reported race data is unavailable — and need the results to hold up in court.
Within five miles of a community bank — and the loan went elsewhere anyway.
During the Paycheck Protection Program, minority-owned businesses disproportionately borrowed through a nationwide platform channel that forgave loans at markedly lower rates than branch-based community banks. Most were not remote — they sat inside the everyday footprint of a community bank that, on its own record, forgave at a far higher rate.
Chalavadi, Leitch & Pastor. Decomposes the forgiveness gap into within-lender and between-lender components, and traces the between-lender share to sorting into a nationwide platform channel — not to physical distance from a community bank.
Read the PaperThe compliance layer, and the engine beneath it
CREST produces the defensible disparity estimate. STRATA is the validated inference engine that feeds it — available on its own as a hosted API or a private no-egress deployment.
CREST quantifies how much legacy proxy methods understate discrimination — producing defensible disparity estimates for compliance self-testing, regulatory exam support, and litigation. Built on a patent-pending methodology validated against peer-reviewed findings that BISG understates Black–White approval disparities by approximately 43% in real lending data.
- Financial institutions for fair lending self-testing and exam defense
- Plaintiff's attorneys quantifying discrimination harm in class actions
- Insurers meeting Colorado SB 21-169 and NAIC proxy-testing mandates
- Applicable across fair lending, employment, insurance, housing, healthcare, and criminal justice
- Enterprise — includes full STRATA imputation$100,000 / yr
- Standalone — litigation & consulting$15,000 / matter
The validated ML inference engine beneath CREST. LSTM + geographic → XGBoost pipeline, validated on a 981,000-record HMDA holdout and a 1.07M-record PPP external validation. Available as a hosted API or a private no-egress deployment for regulated institutions.
- Free tierto 100K rows/mo
- Starter$1.00 / 1K rows
- Growth — 1M–10M$0.75 / 1K rows
- Scale — 10M+$0.50 / 1K rows
- AWS Marketplace$0.03 / 1K rows
- STRATA Premium — no-egress$20,000–$45,000 / yr
The problem with BISG isn't accuracy — it's what the errors hide
BISG's errors are SES-correlated, not random
Minorities in wealthier areas are misclassified as White; Whites in poorer areas are misclassified as non-White. The result: measured Black–White approval disparities are understated by approximately 43% in small-business lending. CREST corrects for this structure — not just the headline accuracy number.
Patent-pending methodology
CREST's bias simulation and disparity estimation methodology is patent-pending (application 64/067,906). The underlying STRATA imputation engine outperforms BISG, BIFSG, ZRP, and all major open-source alternatives on HMDA and PPP ground truth — but the durable moat is the compliance layer, not the benchmark.
Built from federal-grade methodology
The STRATA pipeline grew out of SSA race/ethnicity modeling work. The same rigor that supports federal agency analysis — reproducible, versioned, auditable — is now available as a commercial API and compliance platform.
Two deployment modes for every buyer
Hosted API for researchers, fintechs, and advocacy organizations. Private AWS container for banks, insurers, and mortgage lenders where PII cannot leave the institution's environment. Data never leaves your infrastructure in the Premium tier.
Published and peer-reviewable
Ready to talk?
Whether you're a compliance team preparing for exam, counsel quantifying discrimination harm, an insurer facing proxy-testing mandates, or a researcher — we'd like to hear from you.