Tagged:

4 posts

• Multimodal Jailbreaking: How Attackers Use Images to Bypass Text Safety Filters

  In many multimodal deployments, text safety classifiers operate on tokens — not image pixels. Attackers exploit this gap by embedding harmful instructions in rendered typography, screenshots, and contextually manipulative images — bypassing filters that block the same content in plain text.

  Jul 9, 2026 jailbreak vision-language-models multimodal ai-safety prompt-injection typography-attack image-safety

• Trojan Triggers in Multi-Modal Models: How Visual Backdoors Activate Hidden Behaviors in Vision-Language Systems

  A small adversarial patch embedded in an image — invisible or benign-looking to humans — can reliably trigger hidden behaviors when a vision-language model processes it. As VLMs get deployed in agentic pipelines, visual backdoors become a practical production threat.

  Jul 4, 2026 backdoor-attacks vision-language-models multimodal adversarial-ml supply-chain-security agentic-ai trojan

• Adversarial Attacks on Vision-Language Models: Pixels as Injection Vectors

  Gradient-crafted image perturbations can override VLM safety guardrails, inject attacker-defined instructions, and trigger tool calls in agentic pipelines. A survey of the empirically confirmed attack landscape and practical defenses.

  Jun 30, 2026 adversarial-attacks vision-language-models multimodal prompt-injection ai-safety agentic-ai

• Beyond Text: How Simple Perceptual Tricks Break Multimodal AI Safety

  A systematic study of multimodal jailbreaks shows that models with near-perfect text-only safety can be compromised with >75% success rates using basic image and audio transformations — no gradient access required.

  Jun 19, 2026 jailbreak vision-language-models multimodal ai-safety adversarial-attacks red-teaming