Text-to-image AI models can be tricked into generating disturbing images | Education 2.0 & 3.0 | Scoop.it
From www.technologyreview.com - November 20, 2023 5:17 PM

"Nonsense words can trick Stable Diffusion and DALL-E 2 into producing pictures that show violence and nudity ..."


Via Leona Ungerer