---
title: White Circle RISE: strategy-level red-teaming bypasses modern image model safety; newer GPT-Image/Nano Banana worse | ctxsec
url: https://ctxsec.com/items/whitecircle-rise-t2i-jailbreak
description: ""
status: "confirmed (vendor research)"
published: 2026-10-15
---

> 

[← AI](/ai) · [markdown](/items/whitecircle-rise-t2i-jailbreak.md)

confirmed (vendor research)

2026-10-15

### White Circle RISE: strategy-level red-teaming bypasses modern image model safety; newer GPT-Image/Nano Banana worse

White Circle published RISE (Red-teaming via Iterative Strategy Evolution), an open pipeline that evolves reusable natural-language attack strategies with an abliterated Qwen3-32B attacker and a human-calibrated Gemini judge, rather than paraphrasing seed prompts. On human-verified nudity attacks they report up to 13.3% ASR on DALL·E 3 after exploitation; post-publication runs on GPT-Image-2.5-Flare and Nano Banana 2.1 roughly doubled ASR versus prior versions (7.8% and 20.7% nudity). Prior methods that claimed ~30%+ success fell near 0% under the same calibrated judge. Code is at github.com/whitecircle/rise; strategies, attack prompts, and unsafe images are withheld. Paper: arXiv:2609.34920. Findings were shared with affected vendors before publication.

Sources: [White Circle RISE](https://whitecircle.com/research/rise) · [arXiv:2609.34920](https://arxiv.org/abs/2609.34920) · [GitHub whitecircle/rise](https://github.com/whitecircle/rise)