OnlyFake and the Expanding Risk of AI-Generated Fraud
AI-generated identification documents have moved from a speculative threat to an active criminal business model. Yuriy Nazarenko created OnlyFake, a service that allowed users to produce realistic fake IDs, according to 404 Media.
What makes the case notable is the reported arrival of North Korean actors. Their interest suggests that services such as OnlyFake can supply forged identity material to actors seeking to conceal who they are, operate online accounts, or facilitate fraud. It also highlights the dual-use nature of generative AI: the same systems that can help legitimate users create images and documents can make convincing impersonation easier and more scalable.
The episode illustrates why synthetic-document detection and identity verification need to evolve alongside image-generation technology. Traditional checks based on the appearance of a document may be less reliable when forgeries can be produced quickly and at scale. Defenders may need to combine document analysis with independent identity verification and signals from account behavior.
OnlyFake’s story is ultimately a warning about criminal infrastructure around generative AI. Tools do not need to automate an entire fraudulent operation to be consequential; they can simply remove one important bottleneck. In this case, that bottleneck appears to have been the production of realistic fake identification.