Valantor launches FraudX as complex insurance investigations outgrow manual review
Four weeks after bringing EyeLevel and GroundX into the Enterprise Visual Intelligence platform, Valantor is putting that technology to work on one of the most document-heavy problems in the enterprise: investigating insurance claims.
Today Valantor announced the launch of FraudX, an AI-powered investigation platform that helps insurance carriers, third-party administrators, self-insured organizations and insurance defense law firms investigate complex claims up to 40 times faster than traditional manual review.
The launch follows closely on Valantor's acquisition of EyeLevel, announced in July, which brought GroundX and its document intelligence technology into the platform. That announcement described a gap between where valuable enterprise information lives and where AI can safely operate. FraudX is the first product built to close that gap in a specific industry, starting with construction liability, property and general liability claims.
Investigations are a data problem
Insurance fraud costs the US economy an estimated $308 billion a year. Fraud schemes are growing more sophisticated at the same time claim files are growing more complex. Special Investigation Units, adjusters and defense counsel are routinely expected to read thousands of pages per investigation — medical records, billing files, depositions, investigator notes, surveillance reports, photographs, legal filings — and to spot inconsistencies, recurring entities and relationships that may warrant further investigation.
“Insurance investigations are a data problem as much as an investigative one. Every claim contains thousands of pages of evidence, but investigators simply don't have the time to manually connect every fact, relationship and inconsistency. FraudX helps them spend less time searching for information and more time making informed investigative decisions.”
Built around investigator judgment
FraudX ingests entire claim files in any format and reads them the way an investigator would if an investigator had unlimited hours. It identifies inconsistencies, builds medical and claim chronologies, surfaces investigative signals and maps the relationships between claimants, providers, attorneys and other entities across a file. Investigators ask questions in plain English and get answers back.
Every finding cites the underlying claim documentation that supports it. The output is investigator-ready analysis, traceable to its source.
FraudX is built to support investigator judgment, not replace it. The platform surfaces patterns, relationships and investigative signals across millions of data points that would be difficult to connect through manual review alone. Those connections provide context for an investigation; experienced investigators determine their significance and make the ultimate call.
“FraudX is an absolute game changer for how the industry fights insurance fraud with AI. Its speed, accuracy and ability to uncover hidden patterns set a new standard for modern investigations.”
Sensitive claims, secure environments
Claim files contain some of the most sensitive information an organization holds. FraudX inherits the deployment flexibility that made GroundX a fit for regulated environments in the first place: cloud, private cloud, VPC, on-premise and fully air-gapped. Organizations investigate sensitive claims while keeping control over security, governance and compliance.
The platform is already in production, with customers committing more than 8,000 insurance claims to it.
See FraudX on your own claim files and book a free demo to see what a full claim file looks like when every page has been read.