FICO-LOGO-1_Picture

FICO

HQ: San Jose, CA, United States

Founded: 1956

FICO provides fraud and financial crime solutions that apply machine learning to interpret behavior across multiple dimensions including transactions, authentication services, mobile devices, and P2P beneficiaries.  We protect more than 2.6 billion payment cards from fraud and also offer AI-powered solutions for real-time payments, account takeover, and AML / KYC compliance.

Fraud Solution Profile

FICO

Financial institutions are investing in new types banking interactions such mobile-first (or mobile-only) enhancements, real-time payments and person-to-person apps, to name a few examples. In many cases, the drive to satisfy consumer expectations is outpacing the sense of urgency (and spend) being applied to new fraud detection strategies. This often results in gaps, where fraud defenses lag the introduction of new consumer-focused banking enhancements. Institutions facing these challenges routinely select the Falcon Platform for reasons including:

  1. Depth and Breadth of Machine Learning Capabilities: The Falcon Platform provides real-time behavioral profiling as well as a portfolio of supervised, unsupervised, and semi-supervised machine learning techniques crafted specifically to separate legitimate and fraudulent financial transactions. FICO applies the optimal analytic technique for each type of transaction rather than using a one-size-fits- all approach to model development.
  2. Behavioral profiling: FICO applies fraud-centric machine learning and AI to extract frauds from flowing payment streams with extraordinary precision. This unique approach is based on interpreting the behaviors of each transaction element, such as merchants, devices, payees, and individual consumers, then identifying atypical behaviors of these elements both separately and in aggregate. These real-time behavioral profiles are machine learning models in their own right and work in concert with Falcon’s supervised, unsupervised, and semi-supervised fraud models. They adapt on the fly with each interaction to clearly identify the more subtle indications of fraud that would otherwise go unnoticed. Examples of these behavioral profiling techniques include:
  • Transaction Profiles – A detailed understanding of each consumer’s behavior, updated in real time with each transaction.
  • Merchant Profiles – Assesses the behavior of a specific merchant (including online properties) and combine this information with individual consumers’ behavior data to derive a more comprehensive risk assessment.
  • Self-Calibrating Profiles – Detects behavioral outliers in real-time, even with limited or no data to train the model, and automatically adjusts to accommodate new behavioral patterns. Advanced instances use deep learning to further improve pattern recognition.

These real-time behavioral profiling techniques work with a portfolio of supervised, unsupervised, and semi-supervised fraud models to provide multi-layered fraud defenses.

  1. Omnipresent Coverage: The FICO Falcon Platform is extensible and flexible in order to support all channels and interactions aimed at providing world-class consumer experiences. Each incremental channel or service provides economies of scale by supplementing deep behavioral profiles with new information, which forms a more informed view of enterprise risk. In addition to channel-specific transaction monitoring, these capabilities detect sophisticated indications of account takeover, social engineering and coordinated multi-channel attacks.
  2. Collaborative Networking: The Falcon Intelligence Network supports continuous machine learning innovation with billions of anonymized payment details from a global consortium of more than 9,000 contributing institutions. This unique data asset allows FICO to pursue fraud-specific R&D aimed at developing new AI techniques that accurately separate fraudulent and legitimate consumer behaviors. The Falcon Intelligence Network is the only global resource for monitoring and advancing the performance of machine learning in payment fraud operations.
Customers

UBS Card Center
FIS
EnterCard
Garanti Bank
Nordea


Industry
Ecommerce, Financial Services

Primary Functionality
Fraud Platform

Fraud Type
Account Takeover, KYC & AML, New Account Fraud, Payment Fraud, Synthetic Identity Fraud

Technology
Machine Learning, Rules Engine