A real student was removed from a Stanford University promotional photo and replaced with an AI-generated Black woman, with no public disclosure that the image had been altered.
The incident implicates Stanford’s Residential & Dining Enterprises, exposes a gap between written AI policy and actual practice, and puts a name to the stakes: Billy Ramirez, a Hispanic student in Stanford’s class of 2027.
What the Photo Actually Showed
<a href="https://stanfordreview.org/stanford-r-de-uses-ai-to-race-swap-students-for-advertising/” rel=”nofollow noopener” target=”_blank”>The Stanford Reviewpublished a side-by-side comparison on Sept. 21, 2026, showing what had changed between the original photograph and the promotional version.
Ramirez had been removed entirely and replaced by an AI-generated Black woman. According to The Stanford Daily, at least two other students’ faces were digitally slimmed and their clothing was changed to Stanford-branded merchandise.
Ramirez confirmed to the New York Times that he remembered posing for the original photograph.
A Policy That Said No, Clearly
Stanford’s AI guidance prohibits using AI to create or alter images of Stanford people, events, research, facilities, or achievements, and separately requires disclosure when nondisclosure could mislead audiences.
Stanford Director of University Public Relations Charlene Gageconfirmed that Residential & Dining Enterprises crossed both lines. “Both the alteration and lack of disclosure in this case violate that policy,” Gage said
The nondisclosure is a separate violation from the alteration itself. Stanford treated each as a distinct breach, meaning transparency was required independent of whether the underlying edit was permissible.
“Silenced and Erased”
Ramirez’s reaction shifted as the full scope of the changes became clear to him.
He initially described the alteration as humorous. After learning that an algorithm had reconstructed his race and gender, he told the New York Times he felt “silenced and erased.”
Stanford removed the banners bearing the image and announced it would provide additional training to Residential & Dining Enterprises staff and strengthen review of communications materials. Having an AI policy on paper is a starting point, not a safeguard. This case shows what happens when review procedures fail before altered materials reach the public.
