Logically, a company that uses AI and fact-checking methods to combat online harm, found that AI image generators accept more than 85% of prompts tailored for election manipulation.
The study also revealed significant gaps in platform moderation, with some platforms failing to reject violent or inappropriate content.