Researchers are developing new artificial intelligence technology that can "quarantine" hate speech in a manner akin to malware filters. The AI, which is currently 80% accurate at identifying hate speech, will offer a new way to control exposure to negative content without resorting to censorship.