Where's the liability in the Generative Era? Recovery-based Black-Box Detection of AI-Generated Content

TL;DR

This paper presents a novel black-box detection method for AI-generated images that uses a corrupt-and-recover strategy, requiring only API access, and outperforms existing methods in identifying images from generative models.

Contribution

Introduces a recovery-based black-box detection framework that does not need model weights or large datasets, improving practicality and scalability.

Findings

Outperforms baseline methods by 4.31% in mean average precision.

Effective in detecting images from eight different diffusion model variants.

Works with models that do not support masked inputs by using a surrogate model.

Abstract

The recent proliferation of photorealistic images created by generative models has sparked both excitement and concern, as these images are increasingly indistinguishable from real ones to the human eye. While offering new creative and commercial possibilities, the potential for misuse, such as in misinformation and fraud, highlights the need for effective detection methods. Current detection approaches often rely on access to model weights or require extensive collections of real image datasets, limiting their scalability and practical application in real world scenarios. In this work, we introduce a novel black box detection framework that requires only API access, sidestepping the need for model weights or large auxiliary datasets. Our approach leverages a corrupt and recover strategy: by masking part of an image and assessing the model ability to reconstruct it, we measure the likelihood that the image was generated by the model itself. For black-box models that do not support masked image inputs, we incorporate a cost efficient surrogate model trained to align with the target model distribution, enhancing detection capability. Our framework demonstrates strong performance, outperforming baseline methods by 4.31% in mean average precision across eight diffusion model variant datasets.