
Reference fraud, whether in employment or income documents, used to require skilled fraudsters to spend hours or even days meticulously crafting fake documents using Photoshop or traditional print-based methods. Today, with the rise and accessibility of Generative AI, it's easier than ever to produce counterfeit documents. And now, it's no longer just career fraudsters - this capability is accessible to any applicant looking to exploit outdated referencing processes.
To test how easy it is to create fraudulent documents, we ran a prompt through OpenAI’s latest image generation model:
“Create a reference letter from ASDA confirming Jane Doe worked as a cashier from 2020 to 2024.”
The result? A surprisingly realistic letter on ASDA letterhead - complete with company branding, a plausible message, and a signature from “John Smith, Head of Human Resources.”
Sure, there are a few typos. But that’s beside the point. The takeaway is that this level of fraud-enabling technology is now accessible to anyone with an internet connection and a prompt - and it is only getting better.
And it’s not just employment letters at risk. A wide range of income-related documents, payslips, bank statements, and more, can be fabricated to build an entire arsenal of fraudulent but believable paperwork. These fakes can be used to gain employment, secure loans, or rent properties.
Fraud detection tools are getting smarter all the time. But so is the technology used by fraudsters. What we’re seeing now is an arms race between fraud-tech and the tools built to detect it.
The danger comes when fraud-tech gains the upper hand. And that happening more and more often with the pace at which AI is advancing. Every month brings new advancements in AI-powered tools, making it easier than ever to create forged documents that look convincingly real, including:
Fraudsters no longer need templates, design software, or access to the dark web. They just need ChatGPT. And that’s a serious problem.
If your verification process still relies on trusting a PDF, a signature, or an email from a supposed “HR contact,” it's only a matter of time before something fake gets through, leaving your organisation exposed.
At Konfir, we work with businesses in employment screening, lending, and tenant referencing. We've already heard reports from these businesses of fake documents generated with AI, including entire employer identities created from scratch to support false claims.
These AI-powered fakes are faster, cheaper, and more convincing than anything we’ve seen before. Even vigilant teams are struggling to detect them.
Konfir verifies employment and income directly from the sources - no PDFs, no references, and no guesswork.
We connect to and provide a waterfall of verification sources, including:
If you’re a lender, employer, or referencing provider, now’s the time to ask yourself:
Are you still solely relying on documents?
Are you verifying the truth directly from the source?
Konfir makes it simple to transition to true, source-based verification. Get in touch to learn how we can help you detect and prevent fraud before it slips through the cracks.