How to Stop Meta AI From Processing Your Instagram Content

With the new Muse Image model, Instagram users can tag public profiles in AI prompts—allowing the creation of images using others’ photos, Reels, or audio. For public accounts, this reuse is enabled by default. If you do not want this, you must take action yourself. Here is the guide—and an explanation of Meta’s objectives.

Since the launch of Muse Image (more on that here), a public Instagram account can be tagged in a Meta AI prompt via an @-mention. Meta AI then accesses the publicly visible content of that profile and uses it to build a new image. The problem for those affected: for public accounts, this option is active in the default settings, and according to Meta’s Help Center, you are not notified when others use your content in this way. Those who want to prevent this without making their profile entirely private can manually disable the corresponding switches.

Step-by-step: How to turn off reuse

1. Open the Instagram app and switch to your profile.
2. Tap the menu (three lines) in the top right and open “Settings and activity.”
3. Scroll down to the “Sharing and reuse” section.
4. Deactivate the individual switches—depending on your version, you will find separate options for posts, Reels, and original audio recordings.

Regarding the audio option, the current phrasing is approximately: “Allow others to get creative with your original audio recordings on Meta AI and reuse them.” The description states: “If this option is enabled, anyone can create content in apps and on websites with Meta AI using your entire original audio or an excerpt of it. You can disable this option at any time.” Similarly, the switches for posts and Reels control whether your photos and videos may serve as templates for others’ AI generations.

Note: Meta is rolling out these new settings and phrasing gradually. It is possible that you do not yet see all options or these exact labels. In that case, it is worth checking again in a few days.

What disabling does NOT do

It is important to understand the limits of this setting:

• Not retroactive: Deactivating only stops future generations. Images that others have already created from your content will remain and will not be deleted.
• Privacy only protects moving forward: Even if you switch your account to private retroactively, it does not remove AI images that have already been generated. However, a private profile is more effective protection for the future because the function is designed for public content.
• Meta AI remains in the app: The AI assistant cannot be completely removed from Instagram—it is integrated into Search and Chat, among other things. You can only disable the reuse of your own content by others.

As a self-protection measure, Meta points out that all images created with Muse Image carry an invisible watermark called “Content Seal” and that filters are in place against clearly harmful content. Critics argue that while such a watermark proves AI origin, it does not prevent an unwanted image from being created in the first place.

Meta’s objectives

The fact that Meta releases public content by default, rather than making it available only after explicit consent, is no coincidence; it follows a clear business logic.

A unique selling point in the AI race. Meta’s most important advantage over competitors like OpenAI or Google is its own social graph: billions of public photos, videos, and the knowledge of who is who. This exact social anchoring—Meta markets Muse Image as a model “that knows your world”—is the point where OpenAI and Google cannot easily compete with their models. Incorporating real people from one’s own circle into images thus becomes a differentiating feature.

More content, more usage. The easier it is to create personal, shareable images with friends or creators, the more content is produced that is shared back on the platforms. This increases dwell time and reach—a self-reinforcing cycle from which Meta’s ad-funded business model directly benefits.

Tailwinds for the advertising business. The same technology flows into Meta’s Advantage+ advertising tools, which allow brands to generate personalized ad creatives. Ultimately, more AI content means more material and more opportunities for Meta in its core business: advertising.

Independence from third-party providers. Previously, image and video functions at Meta ran via external models like Midjourney and Black Forest Labs. With its own model, the corporation brings this value creation in-house—part of Mark Zuckerberg’s overarching goal of a “personal superintelligence.”

Opt-out instead of Opt-in—deliberately. If Meta only unlocked the function after active consent, the pool of usable content would remain small. By including public accounts by default, the corporation maximizes the available image, video, and audio base. This exact pattern—the broad use of data unless one actively objects—is something users and privacy advocates have criticized Meta for in the past. In Europe, the function is likely to be closely monitored with regard to the GDPR and the protection of biometric data.

For users, this remains a trade-off between convenience and control: those who want to use the creative functions benefit from the openness of the system—those who want to limit their reach in others’ AI images should check their settings now.