Meta Pulls AI Image Generator After Three-Day Privacy Backlash

Meta has discontinued a new AI feature that allowed users to generate images from publicly posted Instagram photos—just 72 hours after introducing it.

The company launched Muse Image on Tuesday as part of Meta Superintelligence Labs' generative AI portfolio. By Friday, the feature was gone.

The core complaint

Muse Image automatically included all public Instagram accounts in its training data by default. Users had to actively opt out rather than opt in, a design choice that sparked immediate criticism over privacy and copyright concerns.

"Our intent was to provide a useful creative tool and to give people control over whether their public content could be referenced in this way," Meta said in a Friday statement, according to Inc. "We've heard the feedback that this feature missed the mark, so it's no longer available."

The rapid reversal highlights the tension between AI companies' appetite for training data and users' expectations around consent—even for content posted publicly.

Hollywood pushes back

Some of the strongest opposition came from the entertainment industry. Creative Artists Agency, one of Hollywood's most powerful talent agencies, called Meta's approach irresponsible.

"Artists deserve to decide if and how their likeness and work is used," CAA said in a statement released one day after Muse Image's debut.

SAG-AFTRA, the union representing more than 170,000 actors, journalists, and recording artists, urged its members to opt out and accused Meta of failing to grasp the implications for digital likeness rights.

The union's involvement reflects broader anxiety in creative industries about AI systems trained on copyrighted or personal material without compensation or permission.

Why it matters

The episode underscores a critical challenge for AI developers: the gap between what's technically legal (scraping public data) and what users consider ethical. Meta's default opt-in approach may have been within terms of service, but it violated user expectations around control over their images. For enterprise AI leaders, the lesson is clear—consent architecture matters as much as capability. As generative AI tools proliferate, companies that treat permission as an afterthought risk not just PR damage but regulatory scrutiny, especially as lawmakers worldwide draft AI governance frameworks.

What's next

Meta has not indicated whether it will reintroduce Muse Image with revised privacy controls or abandon the concept entirely. The company continues to invest heavily in generative AI across its product suite, including AI assistants and image generation tools within WhatsApp and Messenger.

The details were first reported by Inc.