TRUST ACCURACY
AI-powered customer intelligence
At GBG, we deploy artificial intelligence to advance solutions and combat threats, helping businesses to make faster and more accurate decisions about their customers.
VERIFY
Identity fraud detection
Imagine a world where businesses can trust that the people they interact with online are who they claim to be. That’s the vision of GBG Trust; a unique identity data network uniting more than 650+ organisations across 24 sectors and more than 80 countries.
Powered by AI, this smart network combines millions of identity data records, applying expert-derived pattern matching, data mining and machine learning to trust-test identity claims. It detects inconsistencies, anomalies, and other signs of fraud, delivering an identity trust score in seconds without ever divulging protected data.
This onboarding intelligence protects businesses from identity fraud while ensuring genuine customers can quickly proceed to complete onboarding journeys.
GBG Trust builds on years of experience in identity data science and practical AI. The intelligent algorithms that power Trust perform hundreds of tests identified by machine learning and designed by identity fraud experts. These tests are constantly updated as threats evolve, revealing positive and suspicious patterns and activity hidden deep in the data. The result is intelligent fraud detection beyond the reach of human perception.
Show More
PROVE
Deepfake and tamper detection
Identity proofing against a government-issued ID provides a high degree of assurance that we are who we claim to be, wherever we present ourselves online.
Tamper detection is an essential AI-driven security measure. Algorithms are trained to conduct a pixel-by-pixel document image analysis to identify signs of face substitution and other alterations in seconds.
With the rapid evolution of GenAI, deepfake video injection attacks on identity-proofing systems are increasingly common. Determined fraudsters will bypass a device camera during biometric authentication, ‘injecting’ a deepfake video purporting to present the genuine owner of the ID.
While AI generates new identity security threats, it also advances countermeasures to neutralise those threats. GBG IDscan detects and defends against deepfakes with AI-powered video analysis of lip movement and mouth shapes to detect synchronisation discrepancies, failing identity proofing when these are discovered.
Show More
PROTECT
Transaction fraud detection
Beyond customer onboarding, businesses must continue to protect themselves and their customers from fraud, monitoring transactions as they occur. With digital payments set to exceed $14tn worldwide by 2026, AI is an essential ally in the fight against fraud.
GBG transaction monitoring deploys AI and machine learning to analyse transaction events and complex behavioural patterns, isolating suspicious activity without interrupting the flow and timely approval of legitimate digital transactions.
Working alongside expert fraud rules based on deep domain knowledge, GBG AI/ML models increase fraud detection by up to 34% while reducing false positive alerts by up to 51%. Together, these enhanced results represent millions in savings and revenue for our banking and financial services customers.
Customers deploy GBG expert pre-trained transaction monitoring models or develop and train custom models with our simple model builder. This artificial intelligence continues to learn as fraud patterns evolve, powering the detection of financial crime.
Show More
MATCH
Address parsing
The quality of location data is critical to all businesses that need to know exactly where their customers are. Thanks to AI-powered address parsing, our Loqate solution is setting the global standard for data precision, enhancing address verification accuracy by as much as 18%.
Address parsing breaks down an address into discrete components, such as street name, city, state, and postal code. It’s essential to location data curation. Loqate’s AI address parser deploys deep learning models to process location data, rapidly extracting and organising information from unstructured or complex formats to standardise addresses.
Parsing millions of addresses every day in 250 countries and 6,500 languages, Loqate continually learns, achieving the highest quality premise-level data of any global address verification platform. This delivers substantial improvements in address matching: 7.8% in the United States, 13.9% in Germany and 18% in hard-to-address markets like Indonesia.
By continually improving our location intelligence with AI address parsing, Loqate delivers for our customers all around the world. And by combining a global AI graph with other advanced techniques we can create new addresses that don’t exist in any supplier dataset.
Show More
Secure and trusted
As trust builders, we place people at the heart of what we do, so our use of AI is always fair, transparent and consistent.
Tracking global regulations
We track global regulatory developments relating to AI, adopting and adapting to controls used within legal services, privacy, and information security.
Staying ahead of AI
AI presents important opportunities as well as new security risks and fraud attack vectors. We get ahead of opportunities and stay ahead of threats.
Maintaining human oversight
We maintain human oversight for all AI systems, especially in decision-making contexts, to ensure accountability and manage complexities and risks.
We believe that AI is a transformational technology that will provide compelling benefits to our customers, our team members and our business, through its capacity to assist, complement, empower and inspire people in the way we approach and solve real-world problems.
Mark Mamone, CIO
