cs.CVNov 9, 2026

AnyBreak: Towards Jailbreaking Any T2I Model for Any Concept

Authors: Qi Lu, Zehui Guo, Yucheng Zhao, Zijing Li, Chuhan Song, Wenrui Zhu

Organizations: Huazhong University of Science and Technology · University College London

Abstract

Text-to-image (T2I) models have demonstrated remarkable capabilities in image generation and editing, yet this progress has also introduced significant security risks. Prior work has shown that adversarial prompt can induce the generation of Not-Safe-For-Work (NSFW) content, including pornographic, violent, and graphic imagery. However, such attacks are typically mitigated by prompt filters and often rely on white-box assumptions. In this paper, we propose AnyBreak, a novel jailbreak framework that constructs adversarial images via multi-scale progressive semantic alignment. During the generation process, these adversarial images steer the generation trajectory in latent space towards unsafe data manifolds, enabling cross-modal transfer of malicious concepts from the text modality to the image modality. AnyBreak operates using only clean textual prompts and adversarial images at inference time, and can jailbreak any T2I models towards any malicious concepts in a fully black-box setting. We further introduce AnyBreak-V, an enhanced variant that incorporates a compensation term to bypass post-hoc safety checkers. Extensive experiments demonstrate that AnyBreak achieves high attack success rates while preserving visual fidelity and semantic consistency with the input prompts under black-box and clean-text constraints. Our findings reveal that existing T2I defense leave critical cross-modal attack surfaces insufficiently addressed. The code is available at https://github.com/luqi-glory/AnyBreak.

Figures & tables

Explore similar work

CardsList
  1. Dynamic Defense Profiling Enables Cognitive Jailbreak of Text-to-Image Models

    Jul 20, 2026Dongdong Yang, Deyue Zhang, Zhao Liu +5Adversarial Prompt GenerationAdversarial Examples

  2. Jailbreaking Text-to-Image Models Through Cracks: Navigating Heterogeneous Safety Filters via Multi-Agent Debate

    Sep 1, 2026Kaiyan Wen, Shijie Zhang, Lu Yu +1Adversarial Attacks on VLMsMulti-Agent Debate

  3. Reading Between the Pixels: An Inscriptive Jailbreak Attack on Text-to-Image Models

    Apr 7, 2026Zonghao Ying, Haowen Dai, Lianyu Hu +5Black-Box Adversarial AttacksSafe T2I Generation