LexisNexis launches AI fraud tool that adapts to attacks

LexisNexis Risk Solutions has launched Emailage Adaptive, an AI-based fraud detection product that automatically adjusts risk scoring as fraud patterns change.

The product uses each organisation's transaction data to build tailored fraud models without manual recalibration. It analyses signals including email, IP address, phone and postal address data to assess activity risk across account opening, account management and payments.

In testing, the model captured about 90% of fraud among the highest-risk transactions and reduced false positives by more than 80%. The goal is to cut the number of legitimate transactions flagged for review while improving fraud detection.

The launch comes as businesses face rising fraud losses and faster-changing attack methods. LexisNexis cited its Cybercrime Report 2026, which found that one in 11 new account creations globally is fraudulent. Separate retail and eCommerce research found that the total cost of fraud to US businesses is now more than USD $5 for every USD $1 of direct loss.

Customer example

One early user is Dell Technologies, which has used the product to improve fraud screening. According to LexisNexis, Dell doubled its rate of high-risk fraud capture and reduced manual reviews of low-risk transactions by 83% after implementing Emailage Adaptive in its risk decisions.

"The self-calibrating AI model has helped us modernize fraud prevention by minimizing reliance on static rules and manual policy adjustments," said Jeremy Cole, Director, Business Operations, Dell Technologies. "Our fraud team can now calibrate less, act faster and approve more legitimate transactions with confidence."

LexisNexis said the model is built around a continuous flow of customer transaction data, industry trends and emerging fraud patterns. This allows the system to refine risk scores over time and provide reason codes explaining which factors affected an assessment.

The scoring framework was also designed with explainability, bias mitigation and data privacy in mind. The product uses confirmed transaction outcomes and customer feedback to adjust as new patterns emerge.

Fraud pressure

Email remains a key data point in the company's approach. LexisNexis said there are 7.9 billion email accounts globally and that one-third of consumers keep the same email address for more than 10 years, making historical usage patterns a useful signal when combined with other risk intelligence.

Kimberly Sutherland, Global Head of Fraud and Identity at LexisNexis Risk Solutions, said the growth of generative AI is increasing the speed and scale of attacks. Organisations are struggling to keep manual fraud controls aligned with shifts in criminal behaviour, she said.

"As generative AI scales, fraud attacks are now happening at a faster rate than most manual prevention strategies can keep pace with, overwhelming systems and exposing businesses to ever greater risk," said Sutherland. "By continuously learning from fraud outcomes, backed by cross-industry risk intelligence, a self-calibrating fraud model can adjust instantly to subtle changes in fraud patterns, removing the lag between when patterns emerge and when defenses calibrate. This allows fraud teams to continually enhance predictive accuracy while reducing manual workload."

LexisNexis Risk Solutions is part of RELX and says it serves customers in more than 190 countries and territories.