A photo illustration from 2023 shows an advertisement to create AI girls reflected in a public service announcement issued by the FBI regarding malicious actors manipulating photos and videos to create explicit content and sextortion schemes.

As students nationwide return to the classroom for the 2025-26 school year, they’re experiencing a familiar sense of dread and anticipation. Yet, in addition to the all-too-familiar stress of exams, homework and social interaction, today’s young people must now deal with a new level of anxiety born from our technological age: fear of being the victims of AI-generated deepfake porn.  

Doctored images and videos aren’t new, of course. Celebrities like actress Scarlett Johansson were targeted back in 2018, and she has since become an outspoken advocate against nonconsensual deepfakes. 

What is new is the exploding pace of accessibility and realism of artificial intelligence image generation. In 2024, 34 million AI images were generated daily, and this number is dramatically accelerating.

Advertisement

Article continues below this ad

Explicit deepfakes have become ubiquitous. U.S. cybersecurity firm HomeSecurity Heroes reports that there were 95,820 deepfake videos online in 2023, a 550% increase from 2019. Of those deepfakes, 98% were pornographic in nature, with approximately 99% of nonconsensual sexual deepfakes targeting women and girls.

According to a 2025 study by sexual abuse prevention nonprofit Thorn, 1 in 8 people........

© San Francisco Chronicle