The field of multimodal safety and security is rapidly evolving, with a growing focus on developing robust defenses against increasingly sophisticated attacks. Recent research has emphasized the importance of considering multiple modalities, including text, images, and audio, in order to effectively mitigate potential threats. One notable trend is the development of unified frameworks that can handle diverse modalities and tasks, providing a more comprehensive approach to safety and security. Additionally, there is a growing recognition of the need for transparent and interpretable models, as well as the importance of addressing the limitations of language-specific guardrails. Noteworthy papers in this area include: DefenSee, which proposes a robust and lightweight multi-modal defense technique, and OmniGuard, which introduces a unified framework for omni-modal guardrails with deliberate reasoning ability. Other notable works include Aetheria, which presents a multimodal interpretable content safety framework, and CREST, which develops a parameter-efficient multilingual safety classification model.