RBI wanted to use AI to automate repetitive tasks like documenting intelligence and more rapidly summarizing legal, regulation, and banking documents.
The company built the RBI ChatGPT, using Microsoft Azure OpenAI Service and Azure AI Search within Azure AI Foundry.
RBI employees report greater productivity and the ability to resolve customer issues faster. RBI has set the stage for using generative AI across the company.
Helping people connect to trusted financial opportunities has been the aim of Raiffeisen Bank International (RBI) since its inception. The cooperative bank, with its headquarters in Vienna, Austria, is “at the gateway to Central and Eastern Europe,” says Armin Woworsky, Distinguished Engineer at RBI. “We’ve been building bridges to both established and emerging markets for customers in these regions and internationally for decades.”
RBI’s 43,000 employees offer a wide range of financial services, including retail and corporate banking, leasing, asset management, and mergers and acquisitions, serving 16.5 million customers in 11 markets. “With ongoing growth and our mission of innovation and superior customer experiences, we wanted to be proactive in using new technologies to support our staff. The answer was a full digital transformation with AI at the core,” explains Woworsky.
Knowing that AI integration is a journey involving stages of innovation and adoption, Woworsky says RBI wanted to start by using AI to automate repetitive tasks and document intelligence. It also aimed to more rapidly summarize legal, regulation, and banking documents. RBI wanted to set the stage for the ability to build applications to enhance customer services. “We decided to build a ChatGPT that could free our employees to provide more personalized customer care, while also helping customers help themselves more effectively,” says Woworsky.
RBI was already using Azure as part of its cloud strategy. “With our need to ensure security, privacy, and compliance with European regulations while adopting robust, scalable AI technologies, Microsoft was the right provider—and we knew Azure AI Foundry would be instrumental in helping us realize our long-term goals,” says Woworsky.
With our need to ensure security, privacy, and compliance with European regulations while adopting robust, scalable AI technologies, Microsoft was the right provider—and we knew Azure AI Foundry would be instrumental in helping us realize our long-term goals.
Armin Woworsky, Distinguished Engineer, RBI
With Azure AI Foundry, RBI used Azure OpenAI Service and Azure AI Search as primary tools. RBI used the latest GPT-series models on Azure OpenAI Service. “With Azure OpenAI Service, we can build solutions to address our company’s unique needs and also specific issues within the banking industry, both of which require assurances for data quality and effective data management,” says Woworsky. With responsible use of AI top of mind for AI deployments, RBI uses Azure AI Content Safety to moderate content.
“Using Azure AI Foundry, we can build solutions to address our company’s unique needs and also specific issues within the banking industry, both of which require assurances for data quality and effective data management,” says Woworsky.
Using Azure AI Foundry, we can build solutions to address our company’s unique needs and also specific issues within the banking industry, both of which require assurances for data quality and effective data management.
Armin Woworsky, Distinguished Engineer, RBI
“Azure OpenAI Service and Azure AI Search are key enablers for us,” adds Woworsky. “They provided the agility and scalability needed to support AI solutions across the organization, helping us design RBI ChatGPT for the bank’s most common use cases.” These include line leveling in reconciliation reporting and extracting and summarizing public news articles and financial reports. Employees can now more rapidly search and access information, creating quick customer summaries, drafting emails, and answering regulatory questions. RBI is also employing automated responses to common customer inquiries and supporting employees in creating content.
The RBI ChatGPT was first deployed to 2,000 users and now has an active user base of over 20,000. Employees report that previously time-consuming tasks can be performed quickly. They now have the flexibility to spend time on more value-added tasks or helping customers with complex questions. “People are feeling better about their jobs; we’re seeing greater collaboration and faster resolution of customer issues,” says Woworsky.
Based on Azure OpenAI Service and Azure AI Foundry, RBI ChatGPT will support a variety of internal processes, from facility management to legal services and defining complex products for large corporate customers. RBI employees are empowered to support the ongoing evolution of RBI ChatGPT through a new feature called “yellowGPTs.” Designed by RBI ChatGPT users, yellowGTPs lets employees contribute solutions, content, and ideas for improving efficiency. Users can work with their data to build dedicated knowledge bases and share them on the platform.
Azure OpenAI Service and Azure AI Search are key enablers for us.
Armin Woworsky, Distinguished Engineer, RBI
In addition, as part of its ongoing digital transformation, RBI is developing a platform that can scale and bring RBI ChatGPT and other generative AI tools and functionality to the entire organization. RBI is also experimenting with tying the latest o1 model on Azure OpenAI Service into their application framework. “We hope to make everyone an AI power user,” Woworsky says. “Every employee needs to understand how to interact with language models to remain competitive and provide the best customer experiences.”
Woworsky adds that with Microsoft and AI, RBI’s commitment to customers and their care is stronger than ever. “We’ve set the stage for broad-scale use of advanced technology to provide superior services that can evolve with our customer needs—that's a true differentiator.”
Discover more about Raiffeisen Bank International AG on LinkedIn.
Follow Microsoft