eWEEK content material and product suggestions are editorially unbiased. We could earn cash if you click on on hyperlinks to our companions. Learn More.
Singer Katy Perry shared an AI-generated picture of herself attending the Met Gala, referencing the same picture that caught her mom unawares final yr. The high-profile share of the photorealistic picture highlights how convincing deepfake pictures may be and the way simple it’s to probably create inflammatory footage.
AI invented a Met Gala costume, hair, and make-up styling for Perry
The image Perry shared exhibits the star carrying an outfit matching the 2025 Met Gala theme, “Superfine: Tailoring Black Fashion.” The annual Met Gala raises cash for the Metropolitan Museum of Art’s Costume Institute and serves as an avant-garde trend present populated by the wealthy and well-known. The AI-generated costume combines a futuristic-looking costume with a proper, pinstriped swimsuit.
“couldn’t make it to the MET, I’m on The Lifetimes Tour (see you in Houston tomorrow IRL‼️),” Perry wrote on Instagram, referring to her ongoing live performance tour selling her latest album, “143.” “P.s. this year I was actually with my mom so she’s safe from the bots but I’m praying for the rest of y’all.”
In 2024, Perry made headlines with an Instagram put up displaying one other AI-generated picture of herself on the Met Gala, together with a textual content from her mom. The textual content included “Didn’t know you went to the Met. What a gorgeous gown.” Perry responded with “lol mom the AI got you too, BEWARE!”
Perry’s put up may be seen as a lighthearted observe to her followers to concentrate on extra doable dupes. She didn’t share who had created the deepfake or what generative AI mannequin had been used.
AI-generated pictures of people usually are not allowed by OpenAI – until they’re public figures
One of essentially the most well-known AI corporations, OpenAI, loosened its picture era content material moderation insurance policies in March 2025 to permit ChatGPT to generate pictures of public figures. Previously, OpenAI’s picture era instruments would discourage photorealistic pictures of actual folks in an effort to chop down on deceptive or in any other case dangerous political imagery.
The similar replace allowed the mannequin to copy the kinds of dwelling artists, resulting in the flood of Studio Ghibli-inspired pictures earlier this yr. Artists protested the automated replication of such distinct kinds, after which OpenAI put a cease to the development, citing “important questions and concerns within the creative community.”
Currently, OpenAI’s utilization insurance policies state: “You may not edit images or videos that depict any real individual without their explicit consent,” in addition to “4o image generation is capable in many instances of generating depictions of a public figure based on a text prompt” and offering directions for public figures who want for his or her picture to not be generated. Of course, customers have a tendency to seek out methods round particular person corporations’ tips.
Read eWeek’s protection about actress Scarlett Johansson calling out the usage of her likeness in a deepfake video. And, that judges of the 2026 Academy Awards is not going to keep in mind whether or not generative AI instruments had been used within the making of a film.
