Block News International

@2026 Block News International. All Rights Reserved.

Blends Media
A Blends Media Group Production

Meta Pulls Instagram AI Photo Generator After Privacy Backlash

Arry Hashemi
Arry Hashemi
Jul. 13, 2026
SunriseMeta has removed a feature that allowed users to generate AI images using photos from public Instagram accounts, just days after Muse Image launched. (Unsplash)

Meta has withdrawn a newly launched artificial intelligence feature that allowed people to generate images by referencing public Instagram accounts, reversing course only days after introducing its first image-generation model from Meta Superintelligence Labs.

The feature arrived alongside Muse Image and allowed a user to add a public Instagram account to a Meta AI prompt through an @-mention. Meta AI could then reference publicly available photos associated with that account when producing a new image. Public account holders were initially included unless they changed their Instagram content-reuse settings.

The company has now removed the capability following criticism over consent and the potential misuse of people’s images. Meta acknowledged the reversal, saying it had heard feedback that the feature had “missed the mark.”

Meta Launches Muse Image

Muse Image is the first image-generation model released by Meta Superintelligence Labs, the artificial intelligence organization responsible for the company’s latest generation of models. Meta launched the system through its Meta AI app and website, while also introducing related functions on Instagram Stories in the United States and WhatsApp in a limited number of countries.

Muse Image can follow written instructions, edit existing images and compose a new picture from multiple references. The model can also use search and coding tools while working on a request, capabilities Meta says can improve results involving real-world information, diagrams and functional elements such as QR codes.

The company is positioning Muse Image as more than a conventional text-to-image generator. Meta says the model can assess its initial output and revise parts of an image when it detects a problem, including by editing a detail, beginning again or seeking additional reference material. Those descriptions are based on Meta’s own evaluations and have not been presented as independently verified benchmark results.

Muse Image also includes an invisible watermarking system called Content Seal on images created through the Meta AI app and Meta’s website. Meta says the signal is designed to remain detectable after an image is cropped, compressed, resized or captured in a screenshot. A detection tool is being previewed to help users determine whether supported images contain the watermark.

InstagramPrivacy concerns overshadowed Meta's Muse Image debut, prompting the company to remove a feature tied to public Instagram accounts. (Unsplash)

The most contentious part of the rollout was not Muse Image’s underlying generation technology, but its connection to public Instagram accounts. At launch, Meta allowed people to reference a public account in a prompt and use its publicly posted photos as visual input for an AI-generated creation.

Instagram users could disable the reuse of their content through the platform’s “Sharing and reuse” settings. The arrangement nevertheless drew criticism because the restriction depended on account holders finding and changing a setting rather than affirmatively agreeing to participate.

The issue was especially significant for actors, creators and other professionals whose appearance is closely tied to their work. The controversy also raised broader concerns about whether public visibility should be treated as permission for personal images to be reused in AI-generated content.

Concerns extended beyond celebrities. A public Instagram account can belong to a photographer, small-business owner, journalist or ordinary user who made posts visible for social sharing without anticipating that another person could invoke the account inside an AI prompt. The rollout exposed the widening gap between content being publicly viewable and users consenting to new forms of automated reuse.

Meta Reverses Course

Meta has now removed the public-account reference capability after Muse Image was announced. The company said its intention had been to provide a creative tool while giving users control over whether their content could be referenced but accepted that the implementation had failed to meet public expectations.

The reversal did not amount to the withdrawal of Muse Image itself. The image model remains available through Meta AI, and its other generation and editing functions continue to operate. Creative tools powered by the model also remain part of Meta’s broader plan to integrate generative media across Instagram, WhatsApp, Facebook, Messenger and its advertising products.

Meta’s rapid retreat illustrates the difficulty technology companies face when inserting generative AI into platforms built around personal identity and user-created material. A capability may rely on information that is already public, but the social expectations attached to viewing a photograph are not necessarily the same as those attached to generating a new likeness from it. Muse Image still gives Meta a new foundation for competing in a crowded image-generation market.