Meta's New AI Tool Can Put Real People Into Fake Photos Without Asking First

Meta's Muse Image feature lets anyone generate pictures using other people's public Instagram photos. Critics say the opt-out system puts the burden in the wrong place.

ThreatVectr Newsdesk· 3 min read
A glowing smartphone screen displaying a grid of faintly blurred portrait-shaped photo frames, each dissolving at the edges into swirling digital pixels and abs
Share

Key points

  • Meta launched Muse Image, an AI tool that can generate new pictures using real people's public Instagram photos, without notifying those people.
  • The feature is available inside the Meta AI app, on WhatsApp, and in Instagram Stories for users in the United States.
  • Users must manually turn off a setting buried inside Instagram's "Sharing and Reuse" menu to stop their images being used.
  • UK regulator Ofcom is separately investigating X (formerly Twitter) over its Grok AI tool creating non-consensual altered images of real people.
  • Privacy International and digital rights group Foxglove both told the BBC the feature raises serious consent and privacy concerns.

Meta has a new AI image generator called Muse Image. It can take text you type, combine it with real photos from other people's public Instagram accounts, and produce brand-new pictures. Those people are not told this is happening.

The tool sits inside the Meta AI app and works through a web browser. US users can also access it via WhatsApp and Instagram Stories.

Why should ordinary people care about this?

Because your face, or a stranger's face, could appear in a generated image you never agreed to and never saw coming. The concern is not hypothetical. Non-consensual AI-altered images, meaning fake pictures of real people created without their permission, have caused documented harm on social media platforms over the past year alone.

Donald Campbell, advocacy director at tech justice non-profit Foxglove, described the feature to the BBC as an "obvious recipe for disaster." He pointed to the growing record of harm from manipulated images and questioned why Meta is making it easier to create more of them.

Privacy International called it "the latest sign AI companies see people's images and data as raw material to be exploited."

Meta says users can opt out. The catch: you have to find the setting yourself. Go to Instagram's settings menu, select "Sharing and Reuse," then switch off "Allow people to reuse your content on Instagram and with AI features at Meta" for both posts and reels. It is a separate switch from your main account privacy controls, so having a private account does not automatically protect you.

That design choice is where the real argument lives. Putting the burden on millions of individual users to locate and flip a non-obvious setting is very different from asking for permission upfront.

The backlash lands at an awkward moment for AI image tools generally. UK communications regulator Ofcom is currently investigating X over its Grok AI system and its role in generating and spreading non-consensual altered images of real people. Regulators in multiple countries are watching this space closely.

If your Instagram account is public and you want to remove yourself from Muse Image's pool of source material, change that setting now. It takes about thirty seconds.


Additional reporting cited BBC Technology.

© 2026 Threat Vectr