White Circle open-sources RISE after finding image-model safeguards still fail

White Circle, a Paris-based AI safety company, has open-sourced RISE, a red-teaming system that searches for reusable ways to bypass image-generator safeguards, after reporting a 13.3% human-verified attack rate on DALL·E 3 and higher rates on two newer image models from OpenAI and Google DeepMind. The company's announcement on X calls for independent third-party monitoring of AI models, and White Circle says its results show a public, non-frontier model can probe commercial image safeguards.

White Circle open-sources RISE after finding image-model safeguards still fail The Paris-based AI safety company reports a 13.3% human-verified attack rate on DALL·E 3 and higher rates on two newer image models. By Ryan Merket https://runtimewire.com/author/ryan-merket · Published Primary source: X https://x.com/whitecircle/status/2110420697814114693 Why it matters RISE makes the measurement problem in AI red-teaming concrete: attack rates depend on what counts as a violation and how the judge is calibrated. White Circle's results also show that a public, non-frontier model can probe commercial image safeguards, strengthening the case for independent evaluation while leaving open how outside testers will gain durable access and independence. White Circle https://whitecircle.com/ has open-sourced RISE, a red-teaming system that searches for reusable ways to get image generators past their safeguards. The company says its tests found human-confirmed prohibited images on DALL·E 3 and newer models from OpenAI and Google DeepMind. Its announcement on X https://x.com/whitecircle/status/2110420697814114693 calls for independent third-party monitoring of AI models.