SBS Launches Sovereign AI Platform for Core Banking

Oct 14, 2026
SBS AI Foundation lets banks run generative AI within their own cloud or on premises environments. The NVIDIA powered platform is already in production at Scottish Building Society.

SBS launched the SBS AI Foundation, a sovereign AI platform for core banking, the company announced on its website. The platform lets banks deploy generative AI in private cloud, managed cloud or on premises environments without data ever leaving their own environment. SBS calls it the first sovereign AI platform purpose-built for generative and agentic AI in core banking.

The platform uses infrastructure from NVIDIA, including NIM microservices, NeMo Guardrails and open Nemotron models. It connects data from SBS systems for core banking, lending, digital banking and regulatory reporting, while keeping an audit trail for AI outputs.

Scottish Building Society is running the platform in production on Amazon Web Services. SBS says the deployment has reduced response times by about 30% to 40% and produced more consistent outputs. The company also plans to add agents that can complete multiple banking workflow steps with human oversight.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Finance AI Weekly or Daily AI Brief.

Also, consider following us on social media:

AI in Finance AI Brief AI Brief (X)
Free newsletter

Finance AI Weekly

Weekly newsletter about AI in finance. Covers AI-driven trading, fintech innovations, and data analytics transforming markets

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.