Last updated: 13 October 2026. Our AI helps banks make faster, fairer credit decisions, and every decision stays explainable, auditable and under human control.
1. Where we use AI. Document reading (OCR), credit scoring in AI-Underwriting, rule suggestions in the AI Rule Engine, early-delinquency prediction in AI-Collection, and assistants that help staff work faster.
2. Our principles.
- Explainable. Every score comes with the main factors behind it, in language a credit officer, customer or regulator can understand.
- Fair. Models are tested for bias across relevant groups before release and monitored after. Protected characteristics are not used as model inputs where the law prohibits it.
- Human in control. Clients set credit policy, approve rule changes and can override any AI decision. Policy changes use maker-checker approval.
- Accountable. Every model version, rule version, decision and override is logged for audit.
- Private. A client's data is used only for that client. It is never used to train models for other clients without written agreement.
- Safe and secure. Models run inside the client's isolated environment with the same security controls as the rest of the platform.
3. Governance. An internal AI governance group names/roles reviews new models and material changes, including validation results, bias testing and documentation. Clients receive model documentation to support their own model-risk management.
4. Regulation. We design our AI to help clients meet applicable rules, including explainability and adverse-action requirements and emerging AI regulation such as the EU AI Act, which treats creditworthiness assessment of individuals as high-risk.
5. Questions or concerns. ai@bstech.world.