The Adversarial AI-Art: Understanding, Generation, Detection, and Benchmarking

TL;DR

This paper introduces ARIA, a comprehensive dataset of real and AI-generated images across various categories, and evaluates methods for detecting AI-generated images to address security and authenticity concerns.

Contribution

It provides the ARIA dataset for adversarial AI-art research and benchmarks multiple detection techniques, including a new ResNet-50 classifier.

Findings

Users struggle to distinguish AI-generated images without references.

State-of-the-art detectors have limited effectiveness on the ARIA dataset.

The ResNet-50 classifier achieves promising accuracy and transferability.

Abstract

Generative AI models can produce high-quality images based on text prompts. The generated images often appear indistinguishable from images generated by conventional optical photography devices or created by human artists (i.e., real images). While the outstanding performance of such generative models is generally well received, security concerns arise. For instance, such image generators could be used to facilitate fraud or scam schemes, generate and spread misinformation, or produce fabricated artworks. In this paper, we present a systematic attempt at understanding and detecting AI-generated images (AI-art) in adversarial scenarios. First, we collect and share a dataset of real images and their corresponding artificial counterparts generated by four popular AI image generators. The dataset, named ARIA, contains over 140K images in five categories: artworks (painting), social media images, news photos, disaster scenes, and anime pictures. This dataset can be used as a foundation to support future research on adversarial AI-art. Next, we present a user study that employs the ARIA dataset to evaluate if real-world users can distinguish with or without reference images. In a benchmarking study, we further evaluate if state-of-the-art open-source and commercial AI image detectors can effectively identify the images in the ARIA dataset. Finally, we present a ResNet-50 classifier and evaluate its accuracy and transferability on the ARIA dataset.