business
business

Researchers discover ChatGPT can be tricked into generating sexualised, violent images

OpenAI’s ChatGPT can be manipulated into generating sexualised images and scenes of graphic violence through a slightly altered version of a widely shared prompt, the BBC reports, citing findings from British AI security firm Mindgard.

Sade Ogunleye
BySADE OGUNLEYE-senior correspondent
Last Updated: 2026-06-18T07:46:57.184000
3 Min Read
Researchers discover ChatGPT can be tricked into generating sexualised, violent images
Share This Article

Stay Ahead With AMX News

Get breaking news, in-depth analysis, and expert commentary delivered straight to your inbox. From politics and business to technology and culture — never miss a story that matters.

Explore Now