FICO Unveils Domain-Specific GenAI Models for Financial Services: FICO FFM and FICO FLM

FICO has introduced a new domain-specific FICO Foundation Model for Financial Services (FICO FFM) product offering, consisting of FICO Focused Language Model for Financial Services (FICO FLM) and FICO Focused Sequence Model for Financial Services (FICO FSM). These models are designed to reduce hallucinations and achieve precise accuracy over conventional Generative AI models, with a precise, auditable focus on data used to build and task-train the model for specific tasks or business problems. The models are custom-built, engineered in-house, and focused on curated data sets related to financial services. They are smaller, more accurate, and lower in cost, driving a practical approach toward agentic AI. FICO FLM enhances performance, streamlines operations for financial institutions, and empowers customers to uncover transactional behavior patterns. FICO FSM plays a crucial role in improving real-time detection accuracy across financial services transaction analytics. The models are built with patented and patent-pending Trust Scores, providing a risk-ranking of the reliability of the output generated by these models. By using a Trust Score, organizations can set their own risk thresholds, reduce hallucinations, and operationalize model outputs with confidence. FICO has filed multiple patent applications on FICO FLM and FICO FSM core technology and specific use cases.

Fair Isaac Corporation (FICO) has recently unveiled two new domain-specific AI models designed to enhance financial services: the FICO Foundation Model for Financial Services (FICO FFM) and its components, the FICO Focused Language Model for Financial Services (FICO FLM) and the FICO Focused Sequence Model for Financial Services (FICO FSM). These models are aimed at reducing hallucinations and achieving precise accuracy over conventional Generative AI models, focusing specifically on financial services data.

The FICO FLM and FICO FSM are custom-built, engineered in-house, and trained on curated datasets related to financial services. They are designed to be smaller, more accurate, and lower in cost, offering a practical approach to agentic AI. The FICO FLM enhances performance and streamlines operations for financial institutions, empowering customers to uncover transactional behavior patterns. Meanwhile, the FICO FSM plays a crucial role in improving real-time detection accuracy across financial services transaction analytics.

One of the key features of these models is the use of patented and patent-pending Trust Scores. These scores provide a risk-ranking of the reliability of the output generated by the models. By using Trust Scores, organizations can set their own risk thresholds, reduce hallucinations, and operationalize model outputs with confidence. FICO has filed multiple patent applications on the core technology and specific use cases of FICO FLM and FICO FSM.

These new models represent FICO's ongoing commitment to innovation and its strategic response to emerging competitive pressures and regulatory challenges. They are designed to complement FICO's existing credit risk models and analytics platforms, further solidifying its position in the financial services industry.