Can Generative Models Actually Forge Realistic Identity Documents?

TL;DR

This paper assesses whether current open-source generative models can produce realistic identity documents and finds they mainly mimic surface aesthetics without structural or forensic authenticity, affecting the perceived risk of forgery.

Contribution

It provides an empirical evaluation of popular diffusion-based generative models' ability to create identity documents that could bypass verification systems.

Findings

Models replicate surface aesthetics but lack forensic authenticity.

Current models pose limited risk for realistic document forgery.

Collaboration with forensics is essential for accurate risk assessment.

Abstract

Generative image models have recently shown significant progress in image realism, leading to public concerns about their potential misuse for document forgery. This paper explores whether contemporary open-source and publicly accessible diffusion-based generative models can produce identity document forgeries that could realistically bypass human or automated verification systems. We evaluate text-to-image and image-to-image generation pipelines using multiple publicly available generative model families, including Stable Diffusion, Qwen, Flux, Nano-Banana, and others. The findings indicate that while current generative models can simulate surface-level document aesthetics, they fail to reproduce structural and forensic authenticity. Consequently, the risk of generative identity document deepfakes achieving forensic-level authenticity may be overestimated, underscoring the value of collaboration between machine learning practitioners and document-forensics experts in realistic risk assessment.