Financial Services Turn to GenAI for Customer Insights and Faster Claims Processing
Generative AI is moving beyond efficiency gains in financial services. Banks and insurance companies are now using the technology to uncover new revenue opportunities and improve how employees serve customers.
Three companies show how this plays out in practice: Verisk uses GenAI to speed up claims review, Bud Financial extracts customer insights from transaction data, and BankUnited deployed an internal chatbot that cut back-office support calls by 40%.
Medical Records Review Gets Faster
Verisk Claims, which manages property and casualty claims for insurers, built Discovery Navigator to handle a task that once consumed hours of manual work: reviewing medical demand packages.
These packages are PDFs containing medical and legal information, typically 300 pages long but sometimes exceeding 1,000 pages. A claims adjuster had to read through unstructured documents to find diagnoses, treatment dates, and provider names.
Discovery Navigator feeds these documents through a series of AI models that extract and organize the information into searchable databases. The system breaks out diagnosis codes, prescriptions, and providers, then uses generative AI and LLM technology to create summaries by category.
The result: Verisk clients report a 90% reduction in time spent reviewing medical packages. Adjusters can now make decisions faster and with more accuracy.
Ryan Smith, CTO of Verisk Claims, said the real value comes from combining GenAI with the company's existing expertise in medical, legal, and data science domains. "If all you're doing is taking the latest LLM and running your customers' data through it, why should someone pay you to do that?" Smith said. "The key is understanding what's unique about your data sources and historical knowledge, then figure out how to scale that with GenAI."
Verisk uses Amazon Web Services Bedrock to access multiple AI models, including Anthropic's Claude. The company maintains ongoing investment in updating the models as new medical procedures emerge.
Banks Mine Transaction Data for Customer Understanding
Bud Financial helps banks understand their customers by analyzing transaction data. The company built its own language models and neural networks to categorize spending, identify merchants, detect locations, and spot transaction patterns.
One product, Drive Copilot, uses GenAI as an interface that lets bank employees ask questions about customer segments, marketing opportunities, and spending behavior. The system is designed to work only with specific data sets to avoid hallucinations and maintain explainability-critical for financial institutions.
Jakub Piotrowski, vice president of product at Bud Financial, said banks don't need better customer relationship management tools. "Getting quality data is the big thing," Piotrowski said. "They need to make better use of the data they already have to understand their customers."
Internal Chatbot Transforms Customer Service
BankUnited, a regional bank in Miami Lakes, Florida, faced a problem: employees couldn't quickly answer policy questions from small-business customers, leading to long call times and inconsistent responses.
The bank built SAVI, an internal chatbot using AWS Bedrock and Claude 2. SAVI has access to more than 400 internal policy documents and answers employee questions in under 10 seconds with 95% accuracy.
The impact extended beyond speed. BankUnited reduced back-office support calls by 40%. Customer satisfaction improved. New hires felt more confident because they had a tool to guide them.
Jeiner Morales, senior vice president of data analytics and business systems at BankUnited, said the shift was fundamental. "We've gone from being reactive and spending time looking things up to being proactive," Morales said. "Our bankers can now get quick, accurate answers right in front of them. This allows them to spend less time searching for information and more time actually listening to customers."
Building SAVI required BankUnited to establish new governance practices. The bank worked closely with risk and compliance teams to set guardrails around data access, model validation, and human review. Those controls became the framework for future AI initiatives.
GenAI Becomes Competitive Necessity
For financial services companies, GenAI adoption has moved past optional efficiency tool to competitive requirement. Lisa Gately, principal analyst at Forrester, said the shift happened as businesses discovered new revenue opportunities.
"There's more urgency around generative AI now," Gately said. "Leaders are feeling a lot of scrutiny. With AI in place and some efficiencies gained, businesses are starting to see new revenue opportunities emerge. That creates excitement."
Quick wins exist in specific areas: marketing content campaigns, customer support, HR, and finance functions are seeing early success. But efficiency alone isn't enough. Companies need to track whether quality and regulatory compliance hold steady as they scale operations.
For those in AI for customer support roles, the pattern is clear: GenAI works best when paired with domain expertise, proper governance, and focus on how the technology changes employee workflows-not just automating tasks, but freeing people to do more meaningful work.
Your membership also unlocks:
