META SPARKS OUTRAGE BY ALLOWING USERS TO GENERATE PERSONALISED AI IMAGES OF PUBLIC INSTAGRAM USERS BY DEFAULT

Meta has quietly rolled out a controversial new feature for its generative artificial intelligence tools that allows anyone to generate highly realistic AI images of another person simply by tagging their public Instagram handle.

The tool utilizes Meta's advanced "Muse Image" model. It automatically scrapes public profile photos to use as a visual reference for the AI generator. The update has sparked immediate backlash from privacy advocates, who argue the feature essentially functions as an in-platform "deepfake" generator that compromises personal identity security by default.

Default Opt-In and No Notifications

The most contentious aspect of the rollout is Meta’s implementation strategy. The feature is enabled by default for all public Instagram accounts. Furthermore, Meta will not notify users when a stranger uses their likeness to generate a new image.

Privacy experts have expressed deep concern over the retroactive loophole built into the system. If a user discovers the feature and decides to opt out later, that choice is not retroactive. Any AI-generated images of that user created before the opt-out will remain permanently accessible on Meta's platform.

While the AI generation tool is initially rolling out to users within the United States, the implications are global. Anyone with a public profile anywhere in the world can have their likeness scraped if a US-based user inputs their handle.

The Mechanism: From Tag to AI Clone

The system operates via simple user prompts. A user typing into the Meta AI interface can include a prompt such as, "Create an image of @username as [something]."

The AI model then instantly references the public photos and reels associated with that specific Instagram handle to clone facial features and posture, embedding the targeted individual into the newly fabricated scene.

Meta has long maintained that public data uploaded to its platforms is fair game for training its artificial intelligence models. However, this feature marks a significant shift. It moves from using public data for backend algorithmic training to actively weaponizing individual user data into user-facing generation tools that any member of the public can manipulate.

How to Protect Your Likeness

Users who wish to block Meta from utilizing their photos for personalized AI generations must manually opt out.

To disable the feature, users must navigate to their Instagram Settings, select "Sharing and reuse," and locate the section regarding content use with AI features. Toggling OFF the permissions for both Posts and Reels will immediately revoke the platform's ability to scrape the profile for real-time prompt generation.

Alternatively, switching an account from Public to Private automatically excludes the profile from the Muse Image ecosystem. However, digital safety experts warn that switching off Meta's native tool does not stop bad actors from manually downloading or screenshotting public images to feed into third-party, external deepfake software.