Press Space to continue
Well this is pretty concerning. independent safety researchers
Press Space to continue
Well this is pretty concerning. independent safety researchers tested the latest OpenAI and Google image models and found BOTH got easier to jailbreak. same attacks, newer models, higher success rates. https://x.com/Hesamation/status/21104234…
whitecircle@whitecircle3h 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
