Press Space to continue
Latest AI image models remain vulnerable to jailbreaks
Press Space to continue
We tend to assume that newer AI models are safer than previous (seems like not true) check this research from White Circle shows that even the latest image models from OpenAI and Google remain vulnerable to jailbreaks 😱 and the interesting part is how they're discovered their RISE system evolves attack strategies instead of simply rewriting prompts:
whitecircle@whitecircle5h ago
We are open-sourcing RISE, our red-teaming algorithm that breaks the safeguards of the latest image models from OpenAI, DeepMind, and others. RISE uses readily available models like Qwen and Gemini Flash under the hood, with no additional training required. Read more: https://t.co/kt8ub0OmeW
View