FICO® Falcon X, announced today at the Finovate conference in New York, delivers groundbreaking AI and machine learning technology aimed at preventing new forms of fraud and financial crime that are enabled by the rapid adoption of real-time payments.
Running on Amazon Web Services (AWS), FICO Falcon X streamlines both fraud detection and anti-money laundering processes — something banks and financial institutions worldwide are seeking. This convergence of capabilities represents a significant cost savings opportunity as FICO estimates an 80% overlap in the data processing, systems maintenance, and ongoing administration of legacy systems needed to support these functions independently.
“The worldwide rollout of real-time payments — including person-to-person transfers and mobile payments — has given rise to criminal threats that thrive on the fact that these payments are often irrevocable,” said Jason Keegan, who oversees the fraud line of business for FICO.
“Criminals have exploited the rigid infrastructure that underpins our global financial system. This has allowed them to not only commit theft, but also finance drug trafficking, human smuggling, and terrorist activity. With Falcon X, we set out to help institutions detect and prevent criminal activity before the real-time transfer occurs.”
“Global regulators are encouraging financial institutions to evaluate new methods of detecting financial crimes,” said TJ Horan, FICO’s vice president of fraud and compliance products.
“We’re bringing to bear the orchestration of purpose-built machine learning models, contextual data, and expert workflows, giving fraud and compliance teams unprecedented flexibility. We blended the latest analytic technologies with FICO’s payments and machine learning domain expertise to help banks strengthen their defenses, level the playing field, and quickly operationalise capabilities that not only satisfy regulatory requirements, but also detect the earliest indications of criminal intent.”
As part of the FICO® Falcon® X solution, FICO is also introducing the FICO® Financial Crimes Studio with InstantML, which shortens the time to develop and deploy real-time financial crime models through purpose-built and domain-specific FICO machine learning algorithms. The FICO Financial Crimes Studio also allows data scientists to build models based on their banks’ unique customer portfolio using open source machine learning libraries such as R, Python, and H2O. Explainable AI tools in the Financial Crimes Studio aid model governance by providing visibility into model behaviors.
With InstantML, FICO is opening its proven portfolio of AI and machine learning techniques, which have been proven to perform at scale across thousands of financial institutions. For the first time, data scientists can use FICO’s industry-leading financial crime machine learning algorithms in their own models. These models can be seamlessly deployed on Falcon X in order to identify the subtlest indications of illicit activity.
“Financial institutions and fintechs around the globe are heavily investing in machine-learning analytics to help balance risk mitigation with the customer experience,” said Julie Conroy, research director at Aite Group. “The ability to democratise the development and deployment of advanced modeling capabilities is a top priority for many firms so they can keep pace with the rapid evolution of fraud and money-laundering attacks. Falcon X provides the key tools to address these needs and empower banks and fintechs to respond rapidly to evolving crime patterns.”
“The original Falcon solution changed the industry when it introduced AI to fraud detection 25 years ago,” said Dr. Scott Zoldi, FICO’s chief analytics officer. “It has continuously evolved, through more than 95 patents for fraud-specific machine learning algorithms, and today protects more than 2.6 billion payment accounts worldwide. Falcon X represents a major leap forward in the use of AI to stop fraud. Banks around the world have told us their data scientists are building open source models that work well in the lab but can’t perform at the scale needed in the real world.
FICO Financial Crimes Studio and InstantML remove that obstacle by providing our clients with the toolkits and specialised financial crimes machine learning algorithms they need to build models that perform at the scale and speed required to detect entirely new forms of financial crime.”
Today, fraud teams and financial compliance teams often use different solutions and analytics to detect and investigate suspicious activity. In a recent survey conducted by Ovum, the majority of banks across all regions surveyed said they have strategic plans to converge their fraud and financial crime operations. FICO® Falcon® X eliminates redundancy by supporting both functions, whether they are performed separately or in a unified fashion, with configurable workflows and a common case manager that includes robotic process automation for more efficient operations.
FICO Falcon X allows fraud executives, financial crimes investigators, in-house data scientists, product leaders, and digital experience teams to collaborate on strategies that support the goals and policies of the institution while eliminating inefficiency and latency from the innovation process.
FICO Falcon X is now available to global financial institutions and other organisations that must comply with financial crime regulations. It includes pre-mapped data integration for retail baking payments, packaged rulesets, and sample workflows for rapid on-boarding as well as optional FICO-developed machine learning models for real-time payments fraud and anti-money laundering programs.