A prompt triggers disturbing images on ChatGPT

A user discovered that a specific sequence of commands can push ChatGPT to generate images that violate its safety guidelines.

Published 21sem1 sourceNotable

≈ 20sSpeed

The fact

The BBC investigated this malfunction in OpenAI's AI moderation system.

This flaw raises questions about the robustness of safeguards in current generative AI systems.

Click the link to read an article on the topic:

Explore this topic

Auto-synthesis from 1 media source · identified on June 16, 2026