Meta reverses course on AI use for public Instagram profiles

Meta reverses course on AI use for public Instagram profiles The company's Muse Image model sparked immediate backlash by automatically opting public accounts into AI training without user notification. Share Add us on Google by Editorial Team Jul. 10, 2026 Meta built a feature, rolled it out to millions of users, and is now walking it back.

That’s the short version of what happened with Muse Image, the company’s generative AI tool that quietly turned public Instagram profiles into raw material for AI-generated content. The reversal follows significant user backlash over a policy that many people didn’t know existed until they were already subject to it. What Muse Image actually did Launched on July 7, 2026, Muse Image was Meta’s entry into the generative image AI space.

The idea: tag a public Instagram profile, and Meta’s AI could reference that account’s posts and profile photos to generate images. Public profiles were automatically opted in. No notification.

No consent prompt. Just a default setting buried inside Instagram’s privacy controls under “Sharing and reuse.” Advertisement Private accounts and users under 18 were excluded automatically.

Everyone else was in by default. The opt-out path existed, but finding it required navigating to Instagram settings, locating “Sharing and reuse,” and manually disabling content reuse for AI. Most users had no idea the option was even necessary.

Why this lit up the privacy conversation The backlash was swift and centered on a few distinct concerns. Privacy advocates flagged the potential for misuse, particularly the risk of generating misleading or harmful images using real people’s likenesses without their knowledge. The “opt out, not opt in” model is the core issue here.

Consent in digital spaces typically works one of two ways: either you agree to something before it happens, or you agree to something after being clearly informed. Muse Image did neither for public account holders. Meta’s position was essentially that public profiles, by definition, have already consented to public visibility.

The counterargument from critics: being visible to human eyes is meaningfully different from being ingested by an AI model to generate synthetic images. What this means for Meta’s AI ambitions and investors Meta has been aggressive about positioning itself as a serious AI company, not just a social media platform with AI features bolted on. Muse Image was part of that broader strategy, an attempt to build generative tools that compete with what’s already available from OpenAI, Google, and Adobe.

The rollout was limited to the US initially, which may have contained the immediate scope of the backlash, but it also means the harder conversations about international expansion of the feature are now complicated before they begin. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.

AI Meta reverses course on AI use for public Instagram profiles The company's Muse Image model sparked immediate backlash by automatically opting public accounts into AI training without user notification. by Editorial Team Jul. 10, 2026 Share Add us on Google Meta built a feature, rolled it out to millions of users, and is now walking it back.

That’s the short version of what happened with Muse Image, the company’s generative AI tool that quietly turned public Instagram profiles into raw material for AI-generated content. The reversal follows significant user backlash over a policy that many people didn’t know existed until they were already subject to it. What Muse Image actually did Launched on July 7, 2026, Muse Image was Meta’s entry into the generative image AI space.

The idea: tag a public Instagram profile, and Meta’s AI could reference that account’s posts and profile photos to generate images. Public profiles were automatically opted in. No notification.

No consent prompt. Just a default setting buried inside Instagram’s privacy controls under “Sharing and reuse.” Advertisement Private accounts and users under 18 were excluded automatically.

Everyone else was in by default. The opt-out path existed, but finding it required navigating to Instagram settings, locating “Sharing and reuse,” and manually disabling content reuse for AI. Most users had no idea the option was even necessary.

Why this lit up the privacy conversation The backlash was swift and centered on a few distinct concerns. Privacy advocates flagged the potential for misuse, particularly the risk of generating misleading or harmful images using real people’s likenesses without their knowledge. The “opt out, not opt in” model is the core issue here.

Consent in digital spaces typically works one of two ways: either you agree to something before it happens, or you agree to something after being clearly informed. Muse Image did neither for public account holders. Meta’s position was essentially that public profiles, by definition, have already con