mlsanityData to 7 Oct 2026

JPEG Compressed Images Can Bypass Protections Against AI Editing

arXiv 2304.02234 ↗First posted 5 Apr 2023cs.LG cs.CR cs.CVcited by 14

On the map: Diffusion models → Robustness, privacy and adversarial aspects

Abstract (arXiv)

Recently developed text-to-image diffusion models make it easy to edit or create high-quality images. Their ease of use has raised concerns about the potential for malicious editing or deepfake creation. Imperceptible perturbations have been proposed as a means of protecting images from malicious editing by preventing diffusion models from generating realistic images. However, we find that the aforementioned perturbations are not robust to JPEG compression, which poses a major weakness because of the common usage and availability of JPEG. We discuss the importance of robustness for additive imperceptible perturbations and encourage alternative approaches to protect images against editing.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 7 of its 10 references that are in the map. First posted on arXiv 5 Apr 2023.

Builds on

7 of 10 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Raising the Cost of Malicious AI-Powered Image Editing13 Feb

Other references · 6

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  4. 2023Glaze: Protecting Artists from Style Mimicry by Text-to-Image Modelscited by 200
  5. 2017Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compressioncited by 86
  6. 2021Data Poisoning Won't Save You From Facial Recognitioncited by 45

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 5; 2025: 4; 2026: 3.

14 papers.

Sort:

By area:

  1. 2023Toward effective protection against diffusion based mimicry through score distillationcited by 64
  2. 2023IMPRESS: Evaluating the Resilience of Imperceptible Perturbations Against Unauthorized Data Usage in Diffusion-Based Generative AIcited by 32
  3. 2024Pixel is a Barrier: Diffusion Models Are More Adversarially Robust Than We Thinkcited by 10
  4. 2024DiffVax: Optimization-Free Image Immunization Against Diffusion-Based Editingcited by 10
  5. 2025AdLift: Lifting Adversarial Perturbations to Safeguard 3D Gaussian Splatting Assets Against Instruction-Driven Editingcited by 9
  6. 2025Silence is Golden: Leveraging Adversarial Examples to Nullify Audio Control in LDM-based Talking-Head Generationcited by 7
  7. 2024Exploring Adversarial Attacks against Latent Diffusion Model from the Perspective of Adversarial Transferabilitycited by 5
  8. 2024Pixel Is Not a Barrier: An Effective Evasion Attack for Pixel-Domain Diffusion Modelscited by 3
  9. 2024Protective Perturbations against Unauthorized Data Usage in Diffusion-based Image Generationcited by 2
  10. 2025EditTrack: Detecting and Attributing AI-assisted Image Editingcited by 2
  11. 2026SyncBreaker:Stage-Aware Multimodal Adversarial Attacks on Audio-Driven Talking Head Generationcited by 2
  12. 2025Efficient and Robust Video Defense Framework against 3D-field Personalized Talking Facecited by 1
2 more
  1. 2026Purify Once, Edit Freely: Breaking Image Protections under Model Mismatchcited by 1
  2. 2026A Survey on Adversarial Attacks and Defenses for Diffusion Models Across Multiple Modalitiescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Is Perturbation-Based Image Protection Disruptive to Image Editing?
  2. 2023EditShield: Protecting Unauthorized Image Editing by Instruction-guided Diffusion Models
  3. 2025DCT-Shield: A Robust Frequency Domain Defense against Malicious Image Editing
  4. 2023Can Protective Perturbation Safeguard Personal Data from Being Exploited by Stable Diffusion?
  5. 2025Towards Robust Defense against Customization via Protective Perturbation Resistant to Diffusion-based Purification
  6. 2024DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing
  7. 2025GuardDoor: Safeguarding Against Malicious Diffusion Editing via Protective Backdoors
  8. 2023Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation