Meta backed off using your Instagram photos for AI—here’s how to keep your stuff private

Meta recently pulled back on an AI tool that was automatically scraping public Instagram images to train its models. The move followed widespread criticism from users and privacy advocates who said the company used their content without asking. The company now says it will respect user privacy settings and limit future scraping. But if you’re concerned about your photos being used this way, there are steps you can take right now.

What happened

Meta had been testing a tool designed to collect publicly available Instagram images for training its AI systems. The tool automatically accessed photos and metadata from public accounts—those that anyone can view without following—and fed them into Meta’s AI training pipeline. This wasn’t limited to posts from users who opted in; the scraping was broad and automatic.

Privacy advocates and many users objected. The main complaint was that people didn’t know their content was being used, and they had no way to stop it. In response, Meta said it would rein in the tool, honor user privacy settings going forward, and avoid collecting images from accounts that later switched to private. The change does not retroactively remove images that were already scraped, but it does limit future use.

Multiple news outlets, including the Associated Press, reported on the story on July 11, 2026.

Why it matters

If your Instagram account is public, your photos—including ones you posted years ago—may have been used to train Meta’s AI models. You didn’t give explicit permission. You weren’t asked. And once data is used for training, it can’t easily be removed from the resulting model.

Beyond this specific incident, the episode highlights a broader issue: social media platforms treat public posts as a free resource for AI development. Even if a company like Meta promises to “respect settings” after backlash, the default assumption is often that anything public is fair game. That’s not the same as informed consent.

The change Meta announced does not undo what was already done, but it gives users more control over what happens from here.

What you can do

Here are practical steps to limit the chances that your Instagram content will be used for AI training in the future.

1. Check if your account is public.

  • Open Instagram, go to your profile, and tap the menu (three lines) in the top right.
  • Go to Settings and privacy → Account privacy.
  • If the switch next to Private account is off, your account is public. Tap it to turn it on. (A private account means only followers you approve can see your posts.)

2. Review your data sharing settings.

  • In the same Settings menu, go to Accounts Center (or Meta Accounts Center).
  • Look for Your information and permissions → Activity outside Meta technologies. This shows how Meta uses your data across its services. You can turn off future activity sharing, though it may limit some features.
  • Under Privacy, check Ad preferences and Data about your activity from partners. These settings don’t directly control AI training, but reducing data sharing makes it harder for Meta to use your content beyond what you post.

3. Remove third-party app access.

  • In Instagram Settings, go to Apps and websites.
  • You’ll see a list of apps and websites you’ve authorized. Remove any that aren’t essential. Even if those apps haven’t scraped your data, limiting access reduces risk.

4. Turn off facial recognition.

  • In Instagram Settings, under Privacy and security, go to Face recognition.
  • If it’s on, turn it off. This prevents Meta from using your face in photos for AI training or other purposes. (Note: Instagram may call this setting “Face recognition settings” or something similar depending on your region.)

5. Understand the limits.

  • Making your account private will protect future posts from public scraping. However, images already scraped while your account was public are likely already in Meta’s training data. There’s no way to extract them.
  • The same logic applies to old posts that were public at any time. The safest long-term approach is to keep your account private if you don’t want your content used without explicit permission.

What this means going forward

Meta’s retreat is a response to a specific outcry, not a permanent policy change. Similar tools from other platforms—such as X (formerly Twitter), Reddit, and Google—have sparked parallel debates. The legal landscape around AI training data is still unsettled. In the European Union, data protection authorities may push for stricter rules. In the United States, there is no federal law that clearly requires consent for scraping public social media content for AI training.

For now, the most reliable protection is controlling your own settings. If you want your photos to stay out of AI training pipelines, keep your account private and review your data-sharing preferences regularly. That won’t undo the past, but it gives you a better chance of controlling the future.

Sources: Associated Press (July 11, 2026), morning-times.com, The Tribune-Democrat. Multiple outlets reported on the same story with consistent details about Meta’s AI tool and its rollback.