Stanford's dining and housing office took a photo of a real student and ran it through an AI editing tool.
That tool did not just retouch the picture, it swapped his race and his gender before anyone outside the building saw it.
What Stanford published instead of him is why students are now calling the ad something else entirely.
Stanford Review Exposes the Edited Ad
Billy Ramirez is a sophomore at Stanford.
He was one of the students photographed for a promotional image put out by Stanford Residential & Dining Enterprises, the department that runs student housing and meal plans.
Somewhere between the camera and the finished ad, Ramirez disappeared.
The Stanford Review reported on September 21, 2026 that R&DE ran the photo through artificial intelligence and replaced Ramirez with an AI-generated Black woman.
Two other students standing next to him in the same photo were digitally slimmed down to look thinner than they actually are.
Ramirez told the Review he was "immediately baffled" when he saw what Stanford had done with his picture.
"I don't agree with Stanford making those choices about how we were represented," he said.
He described watching his own identity get replaced as feeling "silenced and erased from a representation that was supposed to include me."
Stanford has not issued a public statement addressing the Review's reporting.
A University That Preaches Authenticity Got Caught Faking It
Stanford did not run out of Black students to photograph, it had a real one standing in the shot and erased him with software instead.
This is not the first time a school has manufactured diversity for a brochure instead of showing the real thing.
The University of Wisconsin was caught in 2000 Photoshopping a Black student into a crowd photo on an admissions booklet cover, inserting him into a scene he was never actually part of.
Stanford did the reverse and arguably the uglier version, taking a student who was actually there and digitally deleting him to make the picture look the way the school wanted it to look.
Critics of this trend call it digital blackface, using AI or Photoshop to manufacture Black or brown faces for a company's image instead of hiring, admitting, or simply photographing real ones.
https://x.com/Polymarket/status/2102106441100267704“>https://x.com/Polymarket/status/2102106441100267704
A university that lectures the country about representation and authenticity got caught doing the opposite, using a computer to decide which races look good in a brochure.
The Diversity Office Treated a Real Student Like a Placeholder
Stanford spends enormous money selling itself as a campus where every student's identity is honored.
The department running that campaign never asked Ramirez before swapping his identity for someone else's, and it did not ask him after the fact either, based on what he told the Review.
That is not representation, that is a marketing department deciding a white male student was the wrong look for the brochure and fixing it with a keyboard.
The same administrators who require diversity statements from faculty and mandatory training for staff apparently saw nothing wrong with digitally replacing a student's race for an ad campaign.
If a fraternity had used AI to lighten a Black student's skin in a flyer, Stanford would have opened an investigation within the hour.
Because the swap went the other direction, and because R&DE has offered no public response since the Review's story ran, Stanford is betting this quietly disappears the way the student in the photo did.
Sources:
- The Stanford Review Staff, "Stanford R&DE Uses AI to Race Swap Students for Advertising," The Stanford Review, September 21, 2026.
- The Daily Caller Staff, "Elite University Used AI To Make Students Hotter, More Diverse In Ad, Student Publication Alleges," The Daily Caller, September 21, 2026.
- Staff, "Stanford Student Says Ad Left Him Feeling 'Erased' After AI Edits Altered Students' Race and Gender," IBTimes UK, September 2026.
- Snopes Staff, "University of Wisconsin Booklet Photoshopped to Add Black Student," Snopes, 2000.
