How Instagram’s New AI Image Generator Puts Your Privacy at Risk (And What to Do About It)

Privacy experts are raising red flags about Instagram’s AI image generator. Here’s how it works, what data it collects, and how you can protect yourself.

Introduction

Earlier this month, Instagram rolled out an AI-powered image generator that lets users create pictures from text prompts. The feature seemed like a fun addition—type in a description and get a custom image. But a July 9 report from The Guardian quoted privacy experts who say the tool raises serious concerns about how user data is collected and used.

If you use Instagram, it’s worth understanding what’s happening behind the scenes and what you can do to limit your exposure.

What happened

Instagram’s AI image generator works similar to tools like DALL·E or Midjourney: you describe an image, and the AI attempts to generate it. But the model needs training data. According to the Guardian report, experts believe the system is trained on images that Instagram users have uploaded to the platform—including photos of themselves, their friends, and their surroundings.

Meta, Instagram’s parent company, has not fully disclosed the specifics of the training dataset. The company’s privacy policy does mention that user content may be used to “develop and improve” its services, including AI features. However, the line between “developing” a feature and training a generative model on personal images is often blurry in the fine print.

Why it matters

Privacy concerns fall into three main categories:

  • Data scraping without clear consent. Many users never agreed to have their photos used to train an image generator. Even if the terms of service permit it, the default setting often opts users in rather than out.
  • Image rights and consent. When the AI generates an image, it may replicate elements from real people’s photos (faces, clothing, backgrounds) without their permission. This raises questions about ownership and likeness rights.
  • Lack of transparency. Meta has not released a clear audit of what data is ingested. Without independent oversight, users cannot verify whether their private photos—including those in direct messages or set to “only friends”—are being used.

Meta’s official response, as reported by The Guardian, emphasized that the feature complies with existing privacy policies and that users can adjust settings. But privacy advocates argue that the company should offer a clear, easy-to-find opt-out for AI training, rather than burying it in menus.

What readers can do

You can’t stop Instagram from using its own features, but you can reduce the amount of data it has access to. Here’s a step-by-step guide based on current settings (as of mid-2026):

  1. Review your privacy settings
    Go to Settings > Privacy and look for sections labeled “Data Sharing” or “Third-Party Sharing.” Disable any options that allow your content to be used for AI training or research. The exact label may vary by region.

  2. Check “Account Privacy”
    Switching to a private account does not necessarily stop Meta from using your images for training, but it limits public scraping. Set your account to Private under Settings > Privacy > Account Privacy.

  3. Turn off “Recommendations in AI Features”
    Some users have reported a toggle in Settings > Privacy > Data Sharing called “Use my images to improve AI.” Turn this off if you see it.

  4. Limit past content visibility
    In Settings > Privacy > Posts, you can restrict who sees prior posts. While this doesn’t retroactively remove data already used, it prevents future content from being accessible to the AI.

  5. Consider alternative platforms
    If you want generative AI features without the same privacy trade-offs, look into tools like Snapchat’s AI (which has clearer labeling of training data) or dedicated standalone generators that don’t rely on your personal photo library. No platform is perfect, but some publish more transparent data practices.

Conclusion

AI features are becoming standard in social media, and Instagram’s image generator is unlikely to be the last. The convenience of generating a picture from text comes at a cost that many users don’t realize until it’s too late. By adjusting your settings now and staying informed about policy changes, you can keep some control over your own images.

As The Guardian noted, the privacy landscape is shifting quickly. Until regulations catch up, the responsibility often falls on individual users to read the small print and act accordingly.

Sources:

  • The Guardian, “Instagram’s AI image generator alarms privacy experts” (July 9, 2026)
  • Instagram Help Center, “Data sharing and AI features” (accessed July 2026)
  • Meta Privacy Policy, sections on content use for service improvement (current version)