Nationwide Building Society has become the first organisation to deploy Aveni's financial services-focused large language model FinLLM.

The UK’s largest building society is currently testing the AI model across a range of compliance use cases, with plans to expand its use across other areas of the business following the completion of trials.

Nationwide invested in Aveni in 2024 and has worked closely with the Edinburgh-based FinTech on the development of FinLLM to shape its technical architecture, governance framework, and risk controls.

Aveni has adopted a "family of models" approach, allowing firms to deploy different AI models for specific use cases, including agent assurance, compliance monitoring, and risk detection.

The company added the model family has been designed specifically for financial services organisations, with a focus on compliance, safety, and regulatory requirements. Unlike general-purpose large language models, Aveni said the tool has been trained using financial services rules, workflows and supervisory expectations.

Aveni claims its LLM has been benchmarked against general-purpose AI models and consistently performs better on financial services-specific tasks.

The technology is already used within Aveni's Assist and Detect products, where it combines structured and unstructured data and operates under governance processes aligned with Financial Conduct Authority guidance and the EU AI Act.

Sri Kanisapakkam, chief data officer at Nationwide, said that since investing in Aveni and working together on co-creating the tool, Nationwide is excited to see how it develops.

“The industry-specific family of models approach that Aveni has taken is exactly what we need: the flexibility to apply the right model for the right task, at scale,” he added. “We’re excited by the early performance we are seeing during testing and the potential benefits FinLLM will bring both Nationwide and our customers, as we continually look to deliver better service and experience through the responsible adoption of new technologies.”

---