$Z_T$: Simplifying Data Distributions for Effective AI-Generated Image Detection

The rapid progress of image generation technology has heightened concerns about misuse, including misinformation and deepfakes, creating an urgent need for reliable AI-generated image (AIGI) detection. Although state-of-the-art detectors based on inversion and reconstruction achieve strong performance in the ideal setting, where the inversion and reconstruction models are the same, their performance drops significantly when these two models differ. All previous works on diffusion-based AIGI detection overlook an important prior, that the noise of generated images from various models is drawn from a standard Gaussian distribution $\mathcal{N}(0,I)$. We propose to exploit this prior in order to improve the generalization and robustness of AIGI detection. We first perform a series of analyses on the distributional dynamics of the diffusion inversion process. Our analysis shows that the prediction errors of the denoising model accumulate at each inversion step, preventing the real image distribution converging to $\mathcal{N}(0,I)$ in the noise space. To further deepen our analysis, we employ a Chi-squared test and a Bayes classifier to identify the distributions to which real and generated image patches converge. Our results show that the distributions of generated and real image patches converge to two very simple forms: the former approximates $\mathcal{N}(0,I)$, while the latter becomes a mixture of a non-standard Gaussian distribution and a distribution with small $\Vert z_T\Vert$. Motivated by these findings, we propose Inversion Noise Classification (INC), a straightforward classification method that operates directly on $z_T$. Extensive experiments across diverse diffusion models and settings show that INC consistently outperforms state-of-the-art AIGI detectors, with notable superiority in robustness to input distortions and generalizability across inversion methods.