Facebook AI is now in your Gallery — and that’s kinda creepy
Turns out “cloud processing” is a polite way to say “we’re watching".
The thing with AI is that it can always bring your wildest imaginations to reality, and sometimes that's not necessarily a good thing.
Over the weekend, reports started coming in about Facebook asking users to grant it access to their camera roll so its AI can start experimenting with changes to your photos, even the ones you haven’t shared.
If you’re trying to post a new Facebook Story, you might get a pop-up asking if you’d like to turn on “cloud processing.” It sounds innocent, but by allowing cloud processing, you’re consenting to Facebook continuously uploading and scanning your camera roll to suggest things like AI-generated collages, themes, or stylized edits.
Meta has promised that the photos stay private and no one else sees them unless you decide to share. But underneath this agreement is consent to Meta AI Terms of Service, which allows it to analyze your media, including your face and the people, places, and things in each picture.
Of course, they're saying the feature isn’t for ad targeting, it’s opt-in, and you can switch it off anytime in your settings. Still, once you opt in, your photos start living on Meta’s servers, where they can be analyzed to generate creative ideas.
This is not Meta’s first time in this kind of privacy-gray-zone. Early this year, users in the EU were given until May 27 to opt out of their publicly shared posts and comments being used to train Meta AI. But this new feature takes things a step further, digging into content you may have never planned to post at all.
There is no mass uproar yet, but a few users have raised concerns. Reddit posts and anti-AI Facebook groups have surfaced screenshots of Meta suggesting AI-edited versions of older photos. One user found their image turned into an anime-style portrait without ever prompting it.
According to Meta, it’s all part of making content sharing easier. Right now, it’s a limited test in the U.S. and Canada, and only those who give explicit permission will see suggestions. But even so, the implications are a reminder that tech companies don’t always wait for your full understanding, just your tap on a "Allow" button.
Data privacy has been a major concern with the development of AI, and this incident only reflects how much it needs to be regulated.
Oluwaseun Bamisile
