SBS launches AI platform for core banking

SBS says its AI Foundation is designed to run generative and agentic AI applications in core banking.[1] The platform is built on NVIDIA infrastructure and supports private cloud, managed cloud and fully on-premises deployment.[1] It uses NVIDIA NIM microservices for its AI layer, NVIDIA NeMo Guardrails to keep responses within each bank’s regulatory parameters, and open Nemotron models.[1][2] SBS is making it available to the hundreds of financial institutions it serves globally.[1] The report also says SBS serves more than 1,500 financial institutions.[2]

The platform is already in production at Scottish Building Society, an existing SBS client in the UK, and runs on Amazon Web Services.[1] SBS says that deployment has delivered approximately 30% to 40% faster response times and more consistent outputs.[2] It also says the deployment maintains the explainability and audit trail regulators require.[2] Those results are SBS’s account of one deployment; the evidence does not establish that other institutions will see the same performance or outcomes.[2]

The platform connects data across SBS core banking, lending, digital banking and regulatory reporting systems for use by generative AI applications and autonomous agents.[1] A customer-focused use case applies generative AI to a bank’s own data to support onboarding and engagement, and can use customer records and the bank’s product portfolio to identify suitable products for individuals.[1] SBS says this can assist with timing, personalisation, operational efficiency, product design and time-to-value through Explainable AI.[1] The co-pilot converts queries over structured and unstructured data into insights; SBS says it can replace multi-day reporting cycles with insights from complex data queries.[1][2]

The risk-intelligence function monitors risk and compliance exposure and generates a regulatory audit trail while keeping risk data in the institution’s controlled environment.[1] SBS says autonomous agents will handle multi-step banking workflows in a governed environment, with human oversight and controls.[1] SBS cites data-residency rules, including GDPR and country-specific requirements, as potential limits on processing customer and transaction data and routing sensitive data through external AI providers.[2] It also cites Gartner data indicating that 84% of finance organisations have implemented or plan to implement AI, while 7% report high or very high impact from those investments.[2]