Source: site
FICO has been granted ten new patents in the areas of Responsible Artificial Intelligence (AI), applied intelligence, and transaction analytics by the US and Canadian intellectual property authorities.
The new patents bring FICO’s portfolio to 231 active US and international patents, with an additional 79 patent applications pending. These patents support FICO solutions such as Falcon Fraud Manager and FICO Platform, which are deployed in billions of enterprise decisions globally.
Patent details
The newly received patents address a range of industry challenges, including bias detection, fraud prevention, and data privacy. They contribute to the improvement of core FICO technologies designed to serve sectors such as financial services, telecommunications, healthcare, retail, transportation, and supply chains.
Among the patents is a “Method for Real-Time Enhancement of a Predictive Algorithm by a Novel Measurement of Concept Drift Using Algorithmically-Generated Features”, which allows real-time monitoring of machine learning models by detecting changes in behaviour compared to the models’ original training data. According to FICO, this supports both its Responsible AI strategy and Falcon Fraud Manager product.
Another patent, “False Positive Reduction in Abnormality Detection System Models”, aims to reduce the rate of false positives in fraud detection, particularly for repeat transactions at the same merchant or location. This technology is integrated into Falcon Fraud Manager and FICO Fraud Predictor with Merchant Profiles.
The “Supervised Machine Learning-Based Modeling of Sensitivities to Potential Disruption” patent enables organisations to evaluate how sensitive their predictive models are to various possible disruptions, using historical data. FICO states that this is core to its Resilience Index offering.
A patent for “Training Artificial Neural Networks with Constraints” builds technical improvements into machine learning models so they can better meet specific business requirements and improve explainability. This work is implemented across several FICO products, including Open Banking Analytics and Transaction Scores.
The “Attributing Reasons to Predictive Model Scores with Local Mutual Information” patent provides institutions with enhanced tools to explain why an AI model produced a particular fraud alert. This, according to FICO, supports transparency and regulatory compliance needs in financial services.
Other patented technologies address problems such as managing missing data, processing large data streams efficiently, and enhancing software verification outcomes. Patents also focus on defending against adversarial attacks and improving large-scale identity resolution.
Patent portfolio significance
“Our patent portfolio continues to grow, and it reflects our commitment to innovation and developing technological advancements for our clients that are not just powerful, but transparent and trustworthy. We are helping clients solve problems responsibly in identifying why an AI model made a specific prediction or in detecting subtle biases across complex datasets. We’re not just advancing AI technology-we’re making Responsible AI the new competitive standard,” said Dr. Scott Zoldi, Chief Analytics Officer at FICO.
The accumulation of patents further consolidates FICO’s place in the development of Responsible AI systems, where transparency in model decisions and bias mitigation are seen as essential by both regulators and industry stakeholders.
FICO’s patent strategy supports not only advances in model accuracy and security but also broader efforts to align AI technology with business and regulatory requirements. The patented inventions also enable safer, more explainable deployment of AI within complex transactional environments, including payment systems, supply chains, and telecommunications networks.
FICO’s approach to Responsible AI is reflected in the direct implementation of these patented advances into products deployed at scale. Patents related to adversarial robustness and fraud detection directly support secure transactions for institutions dealing with billions of data points and significant operational risks.
The company stated that its efforts to expand and diversify its patent portfolio are closely aligned with the ongoing need for responsible, explainable, and reliable AI systems in industries handling sensitive transactions and large-scale datasets.
Upcoming Events
