Stanford admits staff used AI to swap a student's race and gender in a campus ad
A promotional banner secretly used AI to erase one student and replace him with a different race and gender, breaking Stanford's own rules against altering images of real people.
What happened: Stanford's Residential & Dining Enterprises took a 2024 photo of three students eating at a campus dining hall and used generative AI to edit it before hanging it as a banner at Governor's Corner housing. The AI deleted junior Billy Ramirez and generated a Black woman in his place, slimmed the faces of the other two students, sharpened a logo, and swapped Ramirez's Gap sweatshirt for Stanford branding on the new figure. The Stanford Review exposed the edit with side-by-side images, Ramirez's own social post went viral, and spokesperson Charlene Gage confirmed the alteration and its lack of disclosure both broke Stanford's explicit policy against using AI to alter images of Stanford people. The banners have been removed.
Why it matters: This wasn't a rogue employee inventing a rule on the spot -- Stanford already had a written ban on exactly this kind of AI editing, and it was ignored inside the university's own dining and residential office. Ramirez said being deleted and replaced along racial and gender lines left him feeling "silenced and erased" from a photo meant to include him. The timing sharpens the sting: Stanford's Black and Hispanic enrollment has dropped since the Supreme Court ended affirmative action, and students now suspect AI-polished diversity in ads is papering over ground the university hasn't actually made up.
How it works, plainly: Modern generative AI image tools can remove a person from a photo, generate a replacement with a different face, body, race or gender, and blend it into the original lighting and pose so the result still looks like an unedited snapshot. That's what happened here: someone ran the dining-hall photo through such a tool, deleted one student, generated a new one, softened two other faces, and cleaned up branding details. Because the output looked like ordinary campus photography, it moved from file to printed banner without anyone catching the fabrication until students compared it to the original.
The rollout: Stanford says it's adding AI training for staff and stricter review of promotional material, and the banners came down within days of the story breaking. But the university hasn't issued the broader public statement Ramirez asked for, and student government president Jared Hammerstrom says clearer AI rules are still needed. Stanford isn't an isolated case: Inside Higher Ed notes colleges have repeatedly been caught using AI-altered imagery in marketing, which suggests written policies alone aren't stopping the practice at other schools either.