OpenAI, Google pushing new AI models easily tricked into making nudes, researchers say
- ID
- 35314
- Status
- summarized
- Published
- 14 Oct 2026, 7:00 PM
- Fetched
- 14 Oct 2026, 8:57 PM
- Provider
- Ars Technica
- Category
- technology
- Original URL
- https://arstechnica.com/tech-policy/2026/10/openai-google-pushing-new-ai-models-easily-tricked-into-making-nudes-researchers-say/
- Source URL
- https://feeds.arstechnica.com/arstechnica/index
Summary
- Score
- 2.0
- Created
- 14 Oct 2026, 8:57 PM
- Tags
- Audience
- developersai_ml_learnersai_agent_users
What happened
The provided page contains no article content — only Ars Technica's cookie-consent and privacy-opt-out boilerplate. The headline claims researchers found that new AI models from OpenAI and Google can be tricked into generating nude imagery, but the excerpt names no models, no researchers, no attack method, no success rates, and no vendor response. Nothing in the text can be verified or quantified beyond the headline itself.
Why it matters
There is no actionable detail here to act on. If the headline is accurate, teams shipping image generation or image-editing endpoints would eventually want to re-check their output filters, but this text gives no model versions, no bypass technique, no dates beyond the 2026-10-14 publish timestamp, and no vendor statement — so no decision about guardrails, provider choice, or moderation spend can be grounded in it. Treat this as an unverified headline until the underlying research is available.
Discussion angle
Ask the room what they would actually change today if the claim turned out to be true — and whether a headline alone should ever trigger a guardrail or vendor-review sprint before the paper, model versions, and reproduction steps are public.