Customer Service Automation · Document Processing
BBVA wanted to scale safe, responsible generative AI across the bank so frontline and specialist teams could solve knowledge-heavy tasks faster, including answering high volumes of client legal questions, accelerating company credit analysis from many document sources, and extracting insight from large volumes of customer feedback.
Reported impact
BBVA achieved broad internal adoption of generative AI, with most licensed users incorporating it into weekly work, thousands of custom GPTs created, faster legal responses for clients, quicker credit-analysis preparation, and improved ability to derive customer-service insights from survey data.
BBVA rolled out ChatGPT Enterprise across the organization and used it as the shared generative AI platform for employees to build task-specific GPTs. The implementation distributed 3,000 enterprise licenses and allowed teams across countries to create custom assistants close to their business processes. Within that environment, BBVA published around 700 reusable GPTs in its internal BBVA GPT Store. Concrete implementations include the Retail Banking Legal Assistant GPT, which uses documentation maintained by a nine-lawyer advisory team to answer client legal questions routed through branch managers; the Credit Analysis Pro GPT, which pulls relevant information from annual reports, sustainability evaluations, and media documentation to support credit risk assessments; and the Client Experience Assistant GPT in Mexico, which processes thousands of satisfaction surveys to generate conclusions and improvement actions for customer service. BBVA also embedded legal, compliance, and IT security oversight into adoption.
Legal documentation from BBVA's advisory team is used by the Retail Banking Legal Assistant GPT to answer client questions for branch managers; annual reports, sustainability evaluations, and media documentation feed Credit Analysis Pro GPT for analyst review; satisfaction surveys feed Client Experience Assistant GPT for service-improvement recommendations. These GPTs are built and accessed through ChatGPT Enterprise, with many shared through BBVA GPT Store.
"700 are available on BBVA GPT Store."
"83 percent of these users use the tool on a weekly basis, and they have created nearly 3,000 GPTs."
"The GPT was trained to respond to these questions with all the documentation from this team of nine attorneys, thus making it possible to provide faster, more thorough responses to clients, many of them in less than 24 hours."
"The bank started to distribute 3,000 ChatGPT Enterprise licenses in May 2024 at all levels of the organization."
"This GPT automates the extraction of the most relevant data on a company and compares it to its peers in the sector."
"a GPT that processes thousands of satisfaction surveys, extracting the main conclusions and proposing steps to continuously improve customer service"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
OpenAI Sample Oct 16, 2026 Itaú Itaú needed to make financial guidance simpler and more personalized for millions of customers while also scaling document-heavy internal workflows, all under the governance, observability, and economic controls… 100% of the customer base is the goal by the end of 2026 Customer DNA · IARA · OpenAI models OpenAI Sample Mar 28, 2024 Oscar The healthcare industry is notoriously complex. Oscar Health, a healthcare technology company, believes in leveraging technology and data to serve patients at every step of their care. Clinical documentation. Documenting a single conversation between a patient and the medical team can take a… GPT-4 · OpenAI API OpenAI Sample Sep 25, 2026 Wayfair Wayfair needed to keep product attributes accurate across a massive catalog and handle large supplier-support volumes without relying on slow, error-prone manual review. 1,200 ChatGPT Enterprise seats deployed ChatGPT Enterprise · ChatGPT Work · OpenAI API 