TL;DR
Get smart everyday buys delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Finovox says it detects more than €100,000 in AI-related document fraud per month, an annualised rate above €1.2 million. The company also reports generative AI accounts for up to 20% of detected fraud at some clients, with insurance its most affected sector. The figures are company-reported; the source gives no methodology or client sample.
Finovox says it detects more than €100,000 in AI-related document fraud each month, equivalent to an annualised rate above €1.2 million. The European document-fraud detection company also reports that generative AI accounts for up to 20% of detected fraud at some clients, with insurance seeing more of this activity than banking or retail, according to the company.
Finovox’s figures concern fraud it says its systems detect, not a verified estimate of all AI-related fraud across the European market. The report published by Finextra does not specify the number or location of clients included, the period covered by the data, or how Finovox classifies a case as AI-related. The monthly figure should therefore be read as the company’s reported detection total, rather than a market-wide measure.
Finovox says fraudsters use generative AI to produce fabricated documents, alter existing records and manipulate photographic evidence linked to financial transactions. In insurance, the company points to altered images showing damage that did not occur, submitted in support of claims. It also reports an increase in AI-modified invoices, in which existing documents are changed to misrepresent financial information.
The company argues that increasingly realistic alterations can be difficult to spot through conventional document checks. Finovox says it plans to continue investing in research to keep pace with new AI models and fraud techniques. The report does not give a timetable, spending figure or further detail on the research programme.
Insurance Claims Face Altered Evidence
The figures point to a practical problem for organisations that rely on documents and images to approve claims or financial transactions: convincing alterations may pass checks based mainly on visual review. Insurance is identified by Finovox as the most affected sector for generative AI-related fraud, because manipulated photographs can be presented as evidence of damage.
If such claims or invoices are accepted, organisations may face direct losses and the cost of investigation, while legitimate customers could face more demanding verification. Those are potential consequences, not outcomes quantified in the report. Finovox’s data cannot establish the total scale of losses or show how much fraud its tools prevent; it describes detections reported by the company.
The reported €100,000-plus monthly figure gives readers an indication of activity visible to one detection provider. Its relevance depends on the underlying cases and coverage, which Finovox and the report do not detail. The up-to-20% figure applies only to some clients and should not be treated as a rate for all organisations or industries.
Products Worth Considering
Dell UltraSharp U4025QW 40 5K2K WUHD Curved LED Monitor (21:9)
Dell XPS Tower 8960 Desktop, Intel Core i7-13700, NVIDIA GeForce RTX 4060, 16GB DDR5, 512GB SSD
Logitech MX Mechanical Wireless Illuminated Full-Size Keyboard, Clicky
As an affiliate, we earn on qualifying purchases.
How Finovox Describes the Fraud
Finovox is described by Finextra as a European specialist in AI-powered document fraud detection. Its stated work is to help organisations identify falsified or manipulated documents and strengthen verification processes. The report places its latest figures against the growing use of generative AI to create or alter records and images used in financial activity.
The examples cited are specific: fabricated documents, photographs altered to depict nonexistent damage, and invoices changed to misstate financial information. Finovox says traditional verification can struggle with realistic falsifications, particularly subtle edits that are hard to notice with the naked eye. The source provides no case studies or technical evaluation comparing detection methods, so it does not establish how often these examples occur or how reliably any particular tool identifies them.
“With generative AI, incredibly powerful technologies are now in everyone’s hands, including fraudsters.”
— Théophile de Bouët du Portal, Finovox chief technology officer
Scope of the Detection Figures
The report does not say when the data was collected, how many clients contributed, or what geographic and industry coverage the figures represent. It also does not define the criteria used to identify AI involvement, distinguish confirmed fraud from suspected cases, or state whether the monthly total reflects unique incidents, attempted fraud or financial value at risk.
It is also unclear how the “up to 20%” figure was calculated and which clients it describes. No independent data or comparison with other detection providers is cited. The source gives no breakdown of confirmed losses, prevented payments, false positives or detection rates, so the reported figures cannot establish the overall prevalence or financial impact of AI-related fraud.
Finovox Plans Further Research
Finovox says it intends to continue investing in research as AI models and fraud methods change. The company has not provided a schedule, budget, or specific product or research milestones. Any subsequent figures would be easier to assess if they included a defined reporting period, client sample and explanation of how AI-related cases are counted.
For now, the reported monthly detection total remains a company estimate rather than a comprehensive measure of the market. The source does not announce a regulatory change or deadline; the next development to watch for is further information from Finovox about its data and the results of its research.
Key Questions
What does Finovox say it detects each month?
Finovox reports detecting more than €100,000 in AI-related fraud per month. It presents this as an annualised rate above €1.2 million; the source does not provide the measurement period or methodology.
Which sector does Finovox identify as most affected?
Finovox says insurance experiences the most generative AI-related fraud among the sectors it highlights, including banking and retail. It gives no sector-by-sector figures.
How are fraudsters reportedly using generative AI?
The company cites fabricated documents, altered invoices and manipulated photographs, including images made to show damage that did not occur and then used to support insurance claims.
No. The figures are Finovox’s reported detections, and the source does not disclose the client sample, data period, case definitions or market coverage needed to estimate total prevalence.
Source: finextra.com
