ShieldGemma 2: Robust and Tractable Image Content Moderation
This provides an open tool for multimodal safety and responsible AI development, though it appears incremental as it builds on existing Gemma 3 architecture.
The authors tackled the problem of image content moderation by introducing ShieldGemma 2, a 4B parameter model that achieved state-of-the-art performance on both synthetic and natural images across key harm categories.
We introduce ShieldGemma 2, a 4B parameter image content moderation model built on Gemma 3. This model provides robust safety risk predictions across the following key harm categories: Sexually Explicit, Violence \& Gore, and Dangerous Content for synthetic images (e.g. output of any image generation model) and natural images (e.g. any image input to a Vision-Language Model). We evaluated on both internal and external benchmarks to demonstrate state-of-the-art performance compared to LlavaGuard \citep{helff2024llavaguard}, GPT-4o mini \citep{hurst2024gpt}, and the base Gemma 3 model \citep{gemma_2025} based on our policies. Additionally, we present a novel adversarial data generation pipeline which enables a controlled, diverse, and robust image generation. ShieldGemma 2 provides an open image moderation tool to advance multimodal safety and responsible AI development.