Meta Pushed to Better Protect Users from Deepfake Abuse

When a friend tried to help someone targeted by a deepfaked video on Instagram, Meta ignored the reports five times before finally acting. Now the company's independent Oversight Board is pushing for changes that could protect millions of users from similar abuse.

The case started when someone noticed an AI-generated video on Instagram showing their friend in a sexualized situation. The person depicted in the fake video had already closed her account. Despite multiple reports to Meta, the company refused to remove the content, saying it couldn't confirm the person was real or hadn't consented.

Only after the Oversight Board got involved did Meta respond, but even then, the company just marked the video as adults-only instead of removing it. Meta told investigators that if the person in the video had reported it themselves, they would have taken it down under their Adult Sexual Exploitation policy.

That response revealed a troubling gap in Meta's system. The Board found that ordinary people have almost no way to prove they didn't consent to deepfakes unless they report the content themselves, which means facing the traumatic experience directly. Public figures can get help from law enforcement or media outlets, but everyday users are largely on their own.

The Ripple Effect

The Board's three recommendations could change how all social media platforms handle AI-generated abuse. They want Meta to treat all AI impersonations as non-consensual by default, let users designate trusted friends or family to report violations on their behalf, and create a specialized reporting form available worldwide.

Currently, only people in Texas and Florida have access to a form specifically for reporting deepfake intimate imagery. The Board argues this protection should extend globally since AI abuse affects people everywhere, disproportionately harming women and girls.

Meta must respond to these recommendations but isn't required to implement them. The Board has already overturned Meta's decision in this specific case and required the company to remove the video.

The changes matter because deepfake technology keeps getting faster, cheaper, and more convincing. Making it easier to remove harmful content without forcing victims to relive their trauma represents real progress in protecting people's dignity online.