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