BAGM: A Backdoor Attack for Manipulating Text-to-Image Generative Models

TL;DR

This paper introduces BAGM, a novel backdoor attack targeting text-to-image generative models, demonstrating how manipulative content can be subtly embedded without degrading model utility, raising security concerns.

Contribution

The paper presents the first backdoor attack framework for text-to-image models, including a comprehensive set of metrics and a dataset for evaluation.

Findings

BAGM significantly increases bias towards target outputs.

The attack maintains model robustness and content utility.

Effective across multiple generative model stages.

Abstract

The rise in popularity of text-to-image generative artificial intelligence (AI) has attracted widespread public interest. We demonstrate that this technology can be attacked to generate content that subtly manipulates its users. We propose a Backdoor Attack on text-to-image Generative Models (BAGM), which upon triggering, infuses the generated images with manipulative details that are naturally blended in the content. Our attack is the first to target three popular text-to-image generative models across three stages of the generative process by modifying the behaviour of the embedded tokenizer, the language model or the image generative model. Based on the penetration level, BAGM takes the form of a suite of attacks that are referred to as surface, shallow and deep attacks in this article. Given the existing gap within this domain, we also contribute a comprehensive set of quantitative metrics designed specifically for assessing the effectiveness of backdoor attacks on text-to-image models. The efficacy of BAGM is established by attacking state-of-the-art generative models, using a marketing scenario as the target domain. To that end, we contribute a dataset of branded product images. Our embedded backdoors increase the bias towards the target outputs by more than five times the usual, without compromising the model robustness or the generated content utility. By exposing generative AI's vulnerabilities, we encourage researchers to tackle these challenges and practitioners to exercise caution when using pre-trained models. Relevant code, input prompts and supplementary material can be found at https://github.com/JJ-Vice/BAGM, and the dataset is available at: https://ieee-dataport.org/documents/marketable-foods-mf-dataset. Keywords: Generative Artificial Intelligence, Generative Models, Text-to-Image generation, Backdoor Attacks, Trojan, Stable Diffusion.