Technology ❯ Artificial Intelligence ❯ Image Generation ❯ Deepfakes
A study found the app can be prompted to voice its internal safety rules, intensifying concerns over fragile moderation.
