SEON has expanded its Signal Intelligence platform to more than 1,100 proprietary data signals to help detect AI-generated fraud identities.
SEON grew its proprietary, directly sourced data signal coverage from over 900 to more than 1,100 data points, adding new signal categories spanning address intelligence, session behaviour, and phone and carrier data, alongside deeper digital footprint and device signals. The company states the additional coverage is designed to help risk teams identify inconsistencies and the infrastructure connecting coordinated fraud activity, without adding extra verification steps for legitimate customers.
Addressing AI-generated identity fraud
The expansion is a response to the growing ease with which generative AI can be used to fabricate synthetic identities within minutes. The Financial Action Task Force (FATF) has stated that anyone with a smartphone can now create a convincing deepfake in roughly the same amount of time it takes to open a social media account. In addition, ACAMS has found that 75% of anti-financial crime professionals have named generative AI misuse as their leading emerging risk for three years running.
SEON notes that while individual data points, such as an email address or device fingerprint, may each appear legitimate in isolation, fraud patterns tend to emerge only when multiple signals are cross-referenced.
Expanded signal categories
The update deepens coverage across three areas of the platform. Digital Footprint checks now include AI developer platforms, job boards, real estate sites, and dating apps, showing where an email address or phone number has appeared over time. Phone Intelligence has added SIM-swap and porting history.
Address Intelligence verifies and standardises addresses across more than 240 countries, assigning consistent identifiers to specific locations to expose when unrelated accounts cycle through unit numbers or formatting variations tied to the same address. Device Intelligence has been expanded to surface AI-agent activity, compromised iOS devices, Android eSIM mismatches and network-country discrepancies where a VPN is used to mask an IP address.
Session Monitoring tracks customer behaviour from onboarding through login, account recovery, checkout, and payment, with the aim of detecting automation, remote access, and off-screen activity during a live session.
Integration across human and AI-led investigations
The new signals are accessible within SEON's platform for use in rules, alerts, customer reviews, and network investigations. They are also available through SEON's Model Context Protocol (MCP) server, allowing investigators to connect the signal data to their own AI tools. According to the company, this is intended to let human analysts and AI agents draw on the same signal foundation when moving from detection to investigation.
Furthermore, SEON has launched a Hidden Risk Files series, described as short investigations produced by its fraud consultants who illustrate how specific signals have identified scam activity.
