Meta's latest foray into generative artificial intelligence has sparked fresh concerns about digital privacy and the unauthorized use of personal images. The company announced the release of Muse Image, an AI-powered image generator that integrates directly with Instagram, and a forthcoming video generator called Muse Video. While Meta positions these tools as creative enhancements, the default settings allow anyone to generate images of public Instagram users without their explicit consent.
The feature works by scanning an Instagram account's public photos to learn a user's appearance. Once activated, individuals can prompt the AI to create new images of themselves or others. In tests conducted by journalists, the system successfully generated images of both high-profile figures like Mark Zuckerberg and complete strangers—people with public accounts who had no prior connection to the prompter. The only limitation appears to be for private accounts, where the AI may struggle to access sufficient visual data.
How the AI Uses Your Instagram Photos
Muse Image is accessible through Meta's AI web app after logging in with an Instagram account. Users can type a prompt such as 'generate an image of me' or 'generate an image of [username]' and the system will attempt to fulfill it. For private account holders, the AI claimed it could not gather enough data and instead produced a placeholder image of a random person, labeling it as a simulated version. This workaround, while humorous to some, reflects the extent to which Meta is mining visual data.
For public accounts, the process is seamless. The AI pulls from the user's published photos, reels, and stories to construct a digital likeness. This capability was first reported by Wired, which noted that allowing others to use your likeness in this way is the default setting for public Instagram profiles. Meta updated its help page to disclose this, stating that 'people may be able to create content with your Instagram content using AI features at Meta.' The phrasing is broad, covering not only image generation but also remixes, sequences, templates, and stickers.
Opting Out: A Step-by-Step Guide
Concerned users can take immediate steps to protect their images. The quickest method is to set an Instagram account to private using a desktop browser. For those who prefer to keep their account public, the mobile app offers a more granular option. Navigate to your profile, tap the hamburger menu in the upper-right corner, then select 'Sharing and reuse.' There, you will find a toggle for 'Allow people to use your content on Instagram and with AI features on Meta.' Switching off both 'Posts' and 'Reels' will prevent the AI from accessing your photos and videos.
Meta has stated that these controls are now available to U.S. users on Instagram Stories, with WhatsApp integration coming to certain territories. Facebook is expected to receive similar functionality soon. The company also promises that Muse Video, the upcoming generative video tool, will operate under the same privacy framework.
Historical Context and Industry Comparisons
Meta's move is hardly unprecedented. Two years ago, Elon Musk's X (formerly Twitter) rolled out a similar image generator integrated with social media, powered by Grok. That tool quickly became controversial when users generated sexualized images of real people, including minors, during the 2025 holiday season. OpenAI's Sora, which launched as a dedicated app in September 2025, also faced backlash before being shut down in March 2026 as the company pivoted its focus. Both platforms struggled to balance creative freedom with user safety.
Meta appears to be taking a similar risk. While the company claims that 'Muse Image has built-in protections to help prevent the generation of policy-violating content, including violent, sexual, or defamatory imagery of real people,' the effectiveness of these safeguards remains untested at scale. The statement from Meta emphasizes that content violating community standards will be subject to enforcement, but critics argue that reactive moderation is insufficient for a feature that enables deepfakes at such a low barrier to entry.
Implications for Digital Identity
The ability to generate realistic images of strangers without their consent raises profound ethical questions. Digital identity is increasingly vulnerable in an era of advanced generative AI. Public Instagram accounts are often used by creators, journalists, activists, and ordinary individuals who may not anticipate their photos being repurposed for AI training or generation. Unlike a casual remix or sticker, an AI-generated image can replicate a person's likeness with uncanny accuracy, potentially leading to misrepresentation, harassment, or identity theft.
Legal frameworks around the world are still catching up. The European Union's AI Act imposes transparency obligations on providers of general-purpose AI models, but it does not specifically address the use of personal images from social media for generation. In the United States, state-level laws like the Illinois Biometric Information Privacy Act (BIPA) govern the collection of biometric data, but courts have yet to rule on whether AI-generated likenesses qualify as such. Meta's terms of service may grant them broad rights to use uploaded content, but users often skim these agreements without understanding the implications.
For now, the onus is on individuals to navigate the opt-out process. Privacy advocates recommend that anyone with a public Instagram account review their settings immediately. Even those who trust Meta's moderation policies may wish to prevent their face from being used in AI experiments without their knowledge. The company's low—profile rollout of Muse Image has allowed many users to remain unaware of the change until they encounter an AI-generated image of themselves sent by a friend—or a stranger.
As Muse Video prepares to launch, the potential for misuse will only grow. Video generation requires even more data and raises additional concerns about non—consensual deepfakes in motion. Meta has not specified how it will moderate video content differently from images, but the company's statement suggests a commitment to the same enforcement mechanisms. Whether that will be enough remains an open question.
The conversation around AI and privacy is far from over. With each new product release, companies like Meta push the boundaries of what is technically possible while leaving users to fend for themselves. In the meantime, a simple toggle in the settings menu might be the only barrier between your public Instagram photos and an AI system that can clone your appearance on demand.
Source:Gizmodo News