<p>Generative artificial intelligence is capable of generating high-quality, fine-grained 3D assets with realistic visual effects. These 3D assets are often utilized in various scenarios that demand different levels of security permissions. As such, they need to display tiered visual effects while ensuring that the content remains protected. Recently, several researchers have proposed that traditional security encryption methods might not be sufficient to fulfill these requirements. We conduct an in-depth analysis of the structure of generative 3D assets and their visually salient features. Based on this analysis, we propose an encryption scheme for generative 3D assets that achieves tiered visual effects upon decryption. Specifically, a novel coordinate fine-grained regularization method is employed to partition different sub blocks, and ring encryption is implemented, followed by generating keys at different levels, ultimately constructing an encryption scheme with tiered visual effects. Experimental results demonstrate that the proposed scheme not only meets diverse security requirements but also effectively addresses the complexity issues in processing large-scale 3D assets.</p>