Meta has introduced a new image generator called Muse Image, available within its apps like Instagram and WhatsApp. This tool allows users to create custom digital pictures just by describing what they want to see. However, the update includes a controversial setting that automatically uses public Instagram photos to build these images. If your profile is public, other users can tag your handle in their prompts, and the AI will pull your pictures to incorporate them into new AI-generated visuals. You will not receive a notification when this happens, and Meta does not require your explicit permission for each specific instance.
Protecting your photos from AI
To see if you are opted in, go to your Instagram profile, open the menu via the three lines in the corner, and look for the Sharing and reuse section. There, you should find toggles for allowing others to use your posts and reels with Meta’s AI features. While you can switch these off to prevent future use of your images, turning them off will not delete AI-generated content that was already created using your photos before you changed the setting.
WHAT'S HAPPENING, A tool like this functions as a digital collage artist. Normally, when you use an AI image generator, it relies on a massive file of data—known as a dataset—that it studied during its training phase. Training is the process of showing the computer an enormous number of images until it learns to recognize patterns, such as what a "couch" or a "historical landmark" looks like. However, Muse Image is designed to look beyond that initial static file. When someone tags your profile, the system acts like a real-time intern. It reaches out to your public Instagram gallery, pulls your photos, and uses them as a direct reference point to include your likeness in the new image being crafted.
HOW IT WORKS, To understand this, imagine being in a library where one librarian focuses only on the books they were given on day one, and another librarian is allowed to browse the live internet to find updated information. Most AI systems are like the first librarian; they are limited to what they learned during training. Meta’s new approach links the generative tool to your live social media account. By letting the AI scan your current, public images, the system can combine your physical appearance with the user's custom prompt. This makes the generated images feel more personalized, but it changes the expectation of privacy. You are essentially allowing a third party to borrow your personal photos as ingredients for their own AI creations.
WHY IT MATTERS, This launch highlights the friction between how companies want to use our personal content and how we expect that content to be handled. When features are turned on by default, the burden falls on you to be aware of the change and take action to stop it. It brings up an important question: do you own the digital version of yourself, or is it merely raw material for larger systems to use? While Meta sees this as an easy way for people to have fun with their friends, many users are rightfully concerned about their images appearing in AI content they never consented to be a part of.
Liked this one? The next lands at breakfast.
Every story in tomorrow's AI news, rebuilt in plain English — five minutes, sources linked, free forever.
By joining you agree to receive Article's daily newsletter — unsubscribe in one click. Privacy