META'S MUSE IMAGE DEFAULTS TO USING YOUR PUBLIC INSTAGRAM FOR AIGC

Meta launched a new AI image generator called Muse Image on Tuesday, and the way it works is already drawing sharp criticism. The model lets anyone generate AI images that incorporate another person's public Instagram posts, including their profile picture, by simply @mentioning that account in a prompt. Public profiles are opted in by default.

Muse Image is available through Meta's AI app, WhatsApp, and Instagram Stories. Meta says the model "uses advanced reasoning to understand complex prompts, seamlessly blending multiple photos into high-quality creations you can download and share anywhere." It can also browse the web for context and generate infographics with legible text. A companion tool called Muse Spark, which debuted in April, provides the underlying reasoning capabilities. Muse Image brings 30 new AI effects to Instagram Stories and can even help users redesign rooms using products from the web and Facebook Marketplace.

you may like

META PULLS MUSE AFTER 3 DAYS, PRIVACY DEBACLE

GOOGLE IS NOW TRAINING ITS AI ON YOUR UPLOADS, YOU'RE OPTED IN ALREADY

META'S BIG BET: LET USERS CONTROL THEIR ALGORITHMS

THE DEFAULT TRADE

The controversy centres on consent. If your Instagram account is public, any of your photos, reels, and profile pictures can be used by anyone who tags your username in Muse Image. There is no notification when it happens, and images already generated with your content will not be removed even if you later opt out. Meta's help page makes this explicit: users will not be informed when their content is used.

Privacy advocates were quick to react. Public Citizen called the feature "an egregious invasion of user privacy." J.B. Branch, a critic cited in coverage of the launch, put it bluntly: "People should not wake up to discover their face has become raw material for someone else's AI experiment." Branch added that Meta's playbook of default opt-in and buried settings is familiar: "This is another invasion of consumers' privacy. Instead of asking for meaningful consent, Meta quietly defaults users into the system and buries the opt-out in account settings. It’s a playbook we've come to expect from a company with a long history of putting its business interests ahead of the public".

The model is part of a broader push by Meta to compete with generative AI tools from OpenAI, Google, Midjourney, and Adobe. But the rollout has reopened long-running privacy battles. Meta has faced years of scrutiny for using public posts to train AI models by default, requiring users to opt out rather than opting in proactively. This feature extends that logic into the generation phase, where another user actively reuses your content.

HOW TO PROTECT YOUR IMAGES

For now, there are exactly two ways to keep your public Instagram content out of Muse Image. The easier one is to set your account to private. If your profile is private, the model cannot access your photos because it requires public visibility.

The second path is for those who want to keep a public profile but disable reuse. Inside the Instagram app, navigate to Settings, then the tab. Two separate toggles control whether your posts and your reels can be used. Both default to on. Switch them off and your content will no longer be available for AI creations going forward.

The controls are only available in the app, not on the web version. That matters because users who primarily access Instagram via desktop may not realise these toggles exist, and the default setting means their content is available from the moment Muse Image went live.

WHAT THE MODEL ACTUALLY DOES

Beyond the privacy question, Muse Image is a capable image editor. It can edit photos directly in Meta AI, create images from simple language descriptions, and even generate infographics with rendered text that Meta says will be styled to match user preferences. The model suggests presets when users lack ideas and can provide suggestions for what to do with an uploaded photo.

Meta describes Muse Image as "rooted in your world". The company says the model uses the same reasoning technology from Muse Spark, which employs multiple AI agents to handle tasks with speed and accuracy. Muse Spark was pitched as bringing us into an era. Muse Image can handle complex prompts, understand conversational language, and blend photos seamlessly.

The model is rolling out "in limited countries with more on the way", but Meta plans to expand to more countries soon. It will eventually reach Facebook and Messenger, with additional updates for Instagram and WhatsApp. The company is also exploring health assistance features and has worked with over 1,000 physicians to provide curated training data for related AI capabilities.

IS THIS META'S BIG PLAY?

This launch is the latest flash point in a pattern that privacy advocates have been calling out for years. Meta has built a business model that depends on massive data collection, and its approach to AI tools has consistently defaulted to maximum access with minimum friction for consumption, while offloading the burden of privacy protection onto users. The opt-out mechanism exists, but it is hidden several taps deep in a mobile settings menu, and no one will tell you that your face has been used in someone else's creation.

what to read next

GOOGLE ADDS AIGC LABELS ACROSS SEARCH AND YOUTUBE, BUT THERE'S A CATCH

META'S AI CHATBOT LETS HACKERS HIJACK 20,000 INSTAGRAM ACCOUNTS

GEMINI LETS ANYONE* GENERATE PERSONALISED IMAGES FOR FREE

The company clearly sees Muse Image as a way to make its platforms more engaging and to compete with fast-moving rivals. But the cost in user trust may be rising faster than the feature's appeal. For now, any public Instagram user is effectively a contributor to Meta's AI training and generation pipeline, unless they take the time to find the right switch.