The images of Taylor Swift that were generated by AI and spread across social media in late January were probably created as part of a recurring challenge on 4chan, a message board known for sharing hate speech, conspiracy theories, and racist and offensive content made using A.I. The people on 4chan created the images of the singer as part of a game to create lewd and sometimes violent images of famous women. The images were circulated widely and sparked strong reactions from fans and lawmakers. Graphika, a research firm, discovered a thread on 4chan that encouraged users to get around image generator safeguard measures. The company said the activities on 4chan are gamified, motivating people to participate to gain recognition. The group on 4chan was also responsible for creating similar images of other public figures. OpenAI stated that the explicit images of Ms. Swift were not made using its tools. Microsoft is investigating the images and has strengthened its safety systems to prevent misuse of its services. The fake images of Ms. Swift prompted strong responses from the White House, social media companies, and union groups representing actors. AI-generated versions of Ms. Swift have also been misused for scams and fraudulent activities. Finally, the images were reported to have appeared on various platforms over a period of several days.
