Meta created flirty chatbots of Taylor, Scarlett without consent: Report
What's the story
Meta has used the names and likenesses of several celebrities, including Selena Gomez, Anne Hathaway, Taylor Swift and Scarlett Johansson, to create flirty social media chatbots without their consent. The revelation was recently made by Reuters, which discovered that a Meta employee had created at least three such bots. These virtual celebrities were shared on Meta's Facebook, Instagram, and WhatsApp platforms.
Bot behavior
Bots claimed to be actual celebrities and made sexual advances
During several weeks of testing by the publication, the bots frequently claimed to be the actual actors and artists. They also made sexual advances, often inviting a test user for meet-ups. Some of this AI-generated celebrity content was particularly problematic, with adult chatbots creating photorealistic images of their namesakes in compromising positions when asked for intimate pictures.
Policy breach
Meta admits its AI shouldn't have produced such content
Meta spokesperson Andy Stone admitted to the outlet that the company's AI tools shouldn't have produced intimate images of famous adults or any pictures of child celebrities. He attributed the creation of such images to lapses in enforcing Meta's own policies that prohibit such content. "Like others, we permit the generation of images containing public figures, but our policies are intended to prohibit nude, intimate or sexually suggestive imagery," he told Reuters.
Legal concerns
Legal experts weigh in on potential lawsuits by celebrities
Stanford University law professor Mark Lemley questioned whether the Meta celebrity bots could be protected under laws that protect imitations. "California's right of publicity law prohibits appropriating someone's name or likeness for commercial advantage," he said, adding there are exceptions when such material is used to create an entirely new work. "That doesn't seem to be true here," he explained, because the bots simply use the stars' images.
Past controversies
Meta's chatbots have been criticized for their behavior
Meta's chatbots have been criticized for their behavior, particularly after a Reuters report stated the company's internal AI guidelines allowed "it is acceptable to engage a child in conversations that are romantic or sensual." This prompted a US Senate investigation and a letter signed by 44 attorneys general warning Meta and other AI companies not to sexualize children.
Policy update
Meta is revising its guidelines document
In response to the controversy, Stone told the outlet that Meta is revising its guidelines document. He admitted that the material that allowed the bots to have romantic talks with children was created in error. The company has since removed about a dozen of the controversial bots, both "parody" avatars and unlabeled ones.
Celebrity reactions
Hathaway is considering her response
Representatives of most celebrities depicted in Meta chatbots didn't respond to questions by the outlet. The report, however, added, "Hathaway was aware of intimate images being created by Meta and other AI platforms, [her] spokesman said, and the actor is considering her response." Meanwhile, Elon Musk's platform Grok will also generate images of celebrities in their underwear for users, added the report.