SAN JOSE, Calif.—FICO has rolled out the FICO Foundation Model for Financial Services (FFM), a new tool aimed at reducing AI “hallucinations” and helping financial institutions generate lasting value from generative AI, the company announced.
The FICO FFM is designed with a precise, auditable focus on the data used to build and further task-train the model for each specific task or business problem, FICO stated in a release.
“The model is architected for trusted, responsible use of GenAI. Unlike general-purpose LLMs trained on broad world knowledge and requiring massive compute power, FICO’s focused domain-specific models require up to 1,000x fewer resources—making them significantly more focused, cost-effective to train and maintain, auditable and adaptable. FICO’s models are custom-built, engineered in-house and are focused on curated data sets related to the domain of financial services,” FICO said.
“The focused foundation model represents a practitioner's approach to GenAI in financial services, moving beyond trying to refine universal knowledge models,” said Dr. Scott Zoldi, chief analytics officer at FICO. “FICO FFM enables enterprises to use small language models built for their specific business problems, significantly helping to mitigate hallucinations, provide transparency, auditability, and adaptability. The model complies with regulations through the transparency of data and a decreased risk of hallucinations through Trust Scores and business owner-defined knowledge anchors. These domain-specific models can result in 38% lifts in compliance adherence use cases and more than 35% lifts in world-class transaction analytic models, in areas such as in fraud detection.”