UBS undertook a massive data transformation effort to support the creation of two domain-specific AI assistants, collectively known as UBS Red.
Developed using Azure AI Search and Azure OpenAI Service, these AI-powered tools provide support to client advisors during customer interactions. UBS Red offers multi-language capabilities, improving efficiency in addressing client needs.
Michel Neuhaus, head of AI, data & analytics of Personal & Corporate Banking at UBS, underscores the scale of the challenge: “The internal information that we have, the financial expertise, the insights into financial markets, what’s happening in the world, it’s vast. We have to find ways to make this content easily accessible to each and everyone in the bank, and Microsoft helps us with that.”
To address this wealth of information, UBS created a queryable knowledge base. Azure AI Search was implemented to support retrieval strategies and to uncover new insights from unstructured data using language model reasoning. The improved vector search capabilities made it possible to find information based on conceptual similarity across various data types, allowing UBS to extract more value from its vast data resources.
Claire Corish, head of analytics and sales platforms of the Operations and Technology Office at UBS, explains, “We’ve digitized approximately 60,000 investment advice and product documents. This queryable knowledge base built on Azure is now easily accessible to our employees, saving considerable time in meeting preparation and research.”
The unlocking of its data depository has enabled UBS to establish an internal AI Hub for exploring new use cases.
Looking toward the future, Jonas Isliker, head of AI, data & analytics of Global Wealth Management at UBS, says, “Our biggest priority is rapidly moving into AI transformation. While delivering flagship AI solutions like Red is crucial, it’s equally important to build a scalable organization that can accelerate the delivery of AI use cases. This involves implementing a data mesh, moving to the Azure cloud, establishing new AI risk governance, and creating AI factories for quicker, more agile solution delivery. These elements are essential in addition to meeting the growing demand for AI use cases across the organization.”
In 2022, UBS set out plans to have more than 50% of its applications, including critical workloads, running on Microsoft Azure within five years. Within the past ten months, Azure OpenAI Service has been rolled out across all of the bank's key divisions globally, reaching 30,000 employees as of December 2024.