Multimodal Attacks

Embedding malicious text instructions or prompts within images to bypass text-based content filters and inject harmful directives through the visual modality.

Exploiting inconsistencies or conflicts between different input modalities to confuse the AI system and bypass security controls or trigger unintended behaviors.

Crafting audio inputs with imperceptible perturbations that cause speech recognition or audio processing systems to misinterpret commands or bypass security measures.

Manipulation of video streams or recorded content to inject malicious visual sequences, subliminal frames, or adversarial patterns that compromise video understanding systems.

Manipulation of sensor inputs (IoT devices, environmental sensors, biometric readers) to feed false data to AI systems and compromise decision-making in autonomous systems.

Bypassing content filters and safety measures by exploiting weaknesses in specific modality processing, using less-protected input channels to circumvent text-based safeguards.

Crafting inputs across multiple modalities that occupy similar positions in embedding space to confuse similarity matching, retrieval, or classification systems.

Crafting adversarial examples in one modality that successfully transfer to compromise other modalities, exploiting shared representations in multimodal models.

Inserting backdoors that activate only when specific combinations of inputs across multiple modalities are present, creating stealthy trigger-based compromises.

Exploiting the system's prioritization or weighting of different input modalities to bypass security controls by manipulating lower-priority channels.