Generative Artificial Intelligence use cases gain traction in finance

Financial institutions are moving beyond experimentation as generative Artificial Intelligence becomes embedded in claims, customer insight, and service workflows. Firms are pairing foundation models with proprietary data and domain expertise to improve speed, consistency, and business outcomes.

Financial institutions are using generative Artificial Intelligence to sharpen operations, uncover new revenue opportunities, and improve customer experiences. In insurance, the combination of large internal data sets, subject-matter expertise, and generative Artificial Intelligence is emerging as a practical way to scale specialized knowledge rather than simply applying generic large language models. Leaders are increasingly focused on where the technology creates differentiated value, especially when paired with proprietary historical data and business-specific workflows.

Verisk illustrates that approach in insurance claims management. Its Discovery Navigator automates medical record review for bodily injury claims by extracting diagnoses, treatment dates, medical providers, prescriptions, and ICD-10 codes from complex unstructured files. Medical demand packages vary in size from a few pages to more than 1,000 pages of unstructured documents, but averaging about 300 pages. The system turns those records into searchable databases and uses generative Artificial Intelligence to summarize and interrogate documents, helping adjusters make decisions faster. According to Smith, Verisk clients typically see a 90% reduction in time spent reviewing these medical packages. The company uses Amazon Web Services Bedrock and continues to test and update models as new procedures and processes emerge.

Analysts say the finance sector has moved past treating generative Artificial Intelligence purely as an efficiency tool. Business leaders are under pressure to show measurable impact, and growing interest now centers on revenue generation as well as productivity. Early opportunities are appearing in marketing content, localization, translation, customer support, human resources, and finance functions. Quality, consistency, brand alignment, and regulatory compliance remain critical measures of return on investment as organizations scale deployments. 100% The percentage of institutions planning changes in their AI investment that want to increase projects.

Bud Financial applies generative Artificial Intelligence to customer transaction data, helping banks build richer pictures of customer behavior. Its products classify and enrich ambiguous transaction descriptions, identify merchants and locations, and surface insights for segmentation, personalization, marketing, and monitoring. The company combines in-house language models, custom word embeddings, recurrent neural networks, and Google Cloud tools including Vertex AI and Gemini. It positions generative Artificial Intelligence as the customer-facing layer on top of structured transaction intelligence, with a strong emphasis on explainability and limiting hallucinations by constraining responses to client data.

BankUnited is using generative Artificial Intelligence internally to improve service quality and speed. Its SAVI chatbot, built with AWS Bedrock and Claude 2, gives employees natural-language access to more than 400 internal documents. SAVI delivers answers in under 10 seconds, with a 95% accuracy rate. The bank says it has reduced back-office support calls by about 40%. BankUnited also reports gains in customer satisfaction, employee confidence, and training, while using the project to build governance, risk management, validation, and human-review processes that now support broader Artificial Intelligence initiatives across its 55 locations.

65

Impact Score

Supermicro co-founder charged in alleged chip smuggling scheme

US prosecutors have charged three men, including Supermicro co-founder Yih-Shyan “Wally” Liaw, with allegedly diverting high-performance servers containing sophisticated US Artificial Intelligence technology to China. Supermicro says the company is not named as a defendant and has placed two employees on leave while ending ties with a contractor.

Intel bets on GPUs to extend its turnaround

Intel is trying to build on early signs of recovery by pushing deeper into GPUs for Artificial Intelligence data centers and consumer devices. The strategy hinges on sustained demand, sharper execution, and a broader effort to restore confidence in the company.

Neuro-symbolic models cut energy use and improve robot task performance

Researchers at Tufts University report that neuro-symbolic Artificial Intelligence systems could sharply reduce power consumption while improving accuracy in robot control tasks. The approach combines neural network learning with symbolic reasoning to limit trial and error and improve reliability.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.