Stable Diffusion creates fake financial spreadsheets by treating rows of data like tiny pictures.
Image generators are usually for art, but they are surprisingly good at capturing the subtle correlations in tabular data. Attackers can use these models to create synthetic evidence that looks perfectly normal to automated auditing tools. This leads to a silent corruption of the ground truth where AI systems start making decisions based on hallucinated data. Most security systems are looking for deepfake images or videos rather than deepfake CSV files. This technique creates a new category of fraud that targets the very foundations of corporate data analysis.
Repurposing Image Diffusion Models for Adversarial Synthetic Structured Data: A Case Study of Ground Truth Drift
arXiv · 2605.00788
Public image diffusion models are now powerful enough that an attacker without the resources to train a tabular-specific generator may repurpose one off the shelf. This study tests that possibility directly. An unmodified Stable Diffusion U-Net is applied to the UCI Adult Income dataset by reshaping each row into a small single-channel pseudo-image. The architecture's inductive bias toward spatial locality makes feature placement a design variable, and several layouts are tested. However, this i