Leave My Images Alone: Preventing Multi-Modal Large Language Models from Analyzing Images via Visual Prompt Injection

TL;DR

This paper introduces ImageProtector, a method for protecting images from analysis by multi-modal large language models through imperceptible visual perturbations that induce refusal responses.

Contribution

It proposes a proactive, user-side visual prompt injection technique to prevent MLLMs from extracting sensitive information from images.

Findings

ImageProtector effectively induces refusal responses in six MLLMs across four datasets.

Countermeasures like Gaussian noise, DiffPure, and adversarial training partially mitigate ImageProtector.

Mitigation methods often degrade model accuracy and efficiency.

Abstract

Multi-modal large language models (MLLMs) have emerged as powerful tools for analyzing Internet-scale image data, offering significant benefits but also raising critical safety and societal concerns. In particular, open-weight MLLMs may be misused to extract sensitive information from personal images at scale, such as identities, locations, or other private details. In this work, we propose ImageProtector, a user-side method that proactively protects images before sharing by embedding a carefully crafted, nearly imperceptible perturbation that acts as a visual prompt injection attack on MLLMs. As a result, when an adversary analyzes a protected image with an MLLM, the MLLM is consistently induced to generate a refusal response such as "I'm sorry, I can't help with that request." We empirically demonstrate the effectiveness of ImageProtector across six MLLMs and four datasets. Additionally, we evaluate three potential countermeasures, Gaussian noise, DiffPure, and adversarial training, and show that while they partially mitigate the impact of ImageProtector, they simultaneously degrade model accuracy and/or efficiency. Our study focuses on the practically important setting of open-weight MLLMs and large-scale automated image analysis, and highlights both the promise and the limitations of perturbation-based privacy protection.