Meta has decided to withdraw one of the key features of its new image generation model, Muse Image, which allowed the use of public Instagram profiles as a source for creating content. This decision came just three days after launch, indicating serious miscalculations in assessing user expectations and regulatory risks.
As a reminder, on July 7, Meta announced Muse Image as the first image generation model from Meta Superintelligence Labs, integrated into Meta AI. Among its capabilities was a unique option: users could specify a public Instagram account in a prompt, and the model used the open photos from that profile to generate a new image. This was intended to simplify the creation of personalized content but sparked a wave of criticism.
By July 10, the company updated the official announcement, stating that the feature was no longer available. The reason given was that the option "did not meet user expectations" in terms of privacy. This soft explanation actually conceals a deeper issue: the automatic activation of the feature without explicit user consent caused sharp discontent.
Industry representatives expressed particular concern. Actress Hannah Einbinder publicly called for disabling this option, and the SAG-AFTRA union (Screen Actors Guild‐American Federation of Television and Radio Artists) described the approach as a "complete miscalculation in assessing public opinion." The organization emphasized that any use of Instagram user images without clear and conspicuous consent is unacceptable and carries obvious risks.
It is important to note that the Muse Image model itself remains available as a product within Meta AI. It continues to work for generating and editing images based on text prompts and uploaded photos. Additionally, Meta reported that Muse Image is already being used in over 30 new AI effects for Instagram Stories and in WhatsApp chats with Meta AI in a limited number of countries. Thus, the withdrawal of the feature is not a failure of the entire line, but a targeted correction of a dangerous precedent.
Expert opinion: This case is a vivid example of how tech giants, chasing innovation, forget basic principles of privacy. The automatic activation of a feature using public data to generate content was a ticking time bomb. Meta did the right thing by reacting quickly, but the very fact of such a launch indicates that internal mechanisms for checking alignment with user expectations are still lacking. For the industry, this is a signal: even "public" data does not mean automatic consent for its use in generative models.