Meta shut down a feature of its new image-generation AI tool on Friday after public outcry over the automatic use of Instagram users' photos without explicit consent. The company had launched Muse Image less than a week earlier as part of Meta AI, its artificial intelligence assistant, but the feature that scraped all public Instagram accounts for reference material proved immediately controversial.
The tool worked like other AI image generators—users typed prompts and the system created images. But there was a crucial difference: Muse Image automatically made every public Instagram photo available as training material, giving users no way to opt in or out before the feature went live. That gap between what Meta called "control" and what users actually experienced sparked immediate backlash across social media, with people sharing instructions on how to restrict access to their accounts.
Who Bears the Risk
Actors and performers faced particular exposure. The Screen Actors Guild-American Federation of Television and Radio Artists moved quickly to warn its members about the risks of having their likenesses captured and used to generate synthetic images without their knowledge. The union's concern wasn't abstract: nonconsensual digital replicas have become a growing threat to performers' livelihoods and reputations. SAG-AFTRA urged members to change their Instagram privacy settings immediately, treating the feature as a privacy emergency rather than a useful creative tool.
Meta's own framing of the rollout revealed the disconnect between corporate intent and user reality. "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," the company said in its statement. But the feature didn't give people that control upfront—it took their images first, then let them opt out afterward. That's a fundamentally different proposition.
The Response
SAG-AFTRA's statement after the shutdown carried a pointed message: "With the dangers of nonconsensual digital replicas well known to all, a feature that encouraged that behavior is unwise. We appreciate its discontinuance. It is the right thing to do." The union's language suggested this wasn't a technical misunderstanding but a choice that prioritized speed and scale over consent and protection.
Meta acknowledged in its own statement that "this feature missed the mark," a euphemism for a privacy violation that affected millions of Instagram users without their knowledge or agreement. The company didn't explain why the feature was designed to scrape all public accounts by default rather than requiring explicit opt-in before launch—a choice that would have been straightforward to implement.
The rapid reversal shows that public pressure still matters. Within days of launch, users, advocacy groups, and organized labor forced a major tech company to abandon a feature it had invested resources to build. But it also highlights the pattern: companies launch invasive features, face criticism, then claim surprise at the reaction.
Why This Matters:
This incident reveals how tech companies often treat user data and likeness as resources to be extracted unless users actively resist. The default should be consent, not extraction followed by opt-out options. For workers in entertainment and beyond, the stakes are concrete: unauthorized digital replicas can be used to undermine earnings, damage reputations, and displace human labor. Meta's decision to kill the feature is a win for privacy advocates, but it came only after users discovered the threat themselves. Stronger regulations requiring genuine informed consent before any personal data or likeness is used in AI systems could prevent these crises from happening in the first place. The current model—launch, face backlash, retreat—leaves ordinary people responsible for policing corporate behavior rather than building systems designed with consent and protection as foundational principles.