Fully AI-Generated Image Detection: Definition, Recent Advances and Challenges

TL;DR

This paper reviews the detection of fully AI-generated images, focusing on dataset design and artifact extraction methods, highlighting challenges and future directions for robust media forensics.

Contribution

It provides a systematic overview of detection techniques, analyzes dataset influence on robustness, and categorizes artifact extraction methods in AI-generated image detection.

Findings

Dataset design significantly affects detector generalization.

Artifact extraction methods vary based on inductive priors.

The review identifies open problems and future research directions.

Abstract

Recent advances in visual generative models have enabled the creation of highly realistic, fully AI-generated images without relying on real source content. While beneficial for many applications, these models also pose significant societal risks, as they can be easily exploited to produce convincing Deepfakes. Detecting them represents a foundational yet challenging problem in AI media forensics, requiring detectors to reliably extract the inherent artifacts imprinted by generative architectures. In this Review, we provide a systematic overview of fully AI-generated image detection. Following the standard detector design pipeline, we focus on two key components: dataset construction and artifact extraction. We analyze how dataset design influences the generalization and robustness of learned artifacts, and categorize existing artifact extraction methods based on the primary inductive priors leveraged to isolate artifacts. Within this framework, we systematically review existing works. Finally, we highlight open problems and envision several future directions for developing more robust and generalizable detectors. Reviewed works in this survey can be found at https://github.com/zju-pi/Awesome-Fully-AI-Generated-Image-Detection.