Meta has rolled back part of its new Muse Image feature just three days after launch. The company removed the @-mention capability that allowed users to generate AI images based on public Instagram accounts.
The feature debuted on July 7 as part of the broader Muse Image rollout but disappeared by Friday evening after widespread criticism. In a blog post, Meta acknowledged that the feature had “missed the mark.” Puck News’ Dylan Byers first reported the reversal on X.
What Meta Removed and Why Users Objected
Meta removed only the Instagram-reference feature. Muse Image itself remains available across Meta AI, Instagram, and WhatsApp for standard AI image generation and editing.
The removed capability allowed anyone to mention a public Instagram account belonging to an adult and generate AI-created images based on that person’s public photos.
Meta automatically included public accounts but excluded private accounts and accounts belonging to minors.
Users had to manually find and enable an opt-out setting to prevent others from using their images. Meta also did not notify people when someone generated AI images using their public content.
The design quickly drew criticism because it relied on automatic participation rather than explicit consent.
Meta said the feature would give users more control over their content. However, critics argued that the default settings offered little meaningful choice.
Industry Backlash and the Broader AI Consent Debate
The backlash grew rapidly after organizations including CAA and SAG-AFTRA criticized the rollout. SAG-AFTRA described anything short of clear opt-in consent as unacceptable, while several technology publications published guides explaining how users could disable the feature.
After Meta withdrew the capability, CAA welcomed the decision but urged the company to go further by making consent opt-in by default. The agency also argued that people should have the ability to monitor and remove unauthorized AI-generated uses of their likeness rather than relying solely on opt-out controls.
The debate also exposed broader industry tensions. Deadline noted that CAA operates its own AI Vault program to archive clients’ digital likenesses, highlighting the complex balance between commercial AI use and personal control.
Meta is not the first AI company to reverse course after introducing default likeness-generation features. OpenAI faced similar criticism after launching comparable image-generation capabilities with Sora in 2025 and later scaled them back following public backlash.
The repeated reversals suggest that consent has become one of the biggest challenges in consumer AI.
As companies release increasingly powerful image-generation tools, users and creators continue to demand stronger privacy protections, clearer consent mechanisms, and greater control over how AI systems use their digital identities.
Source: TechCrunch, "Meta Removes Controversial AI Feature on Instagram After Backlash"
