Instagram’s new AI image generator uses your photos: Here’s how to opt out

In early July 2026, Instagram activated an AI image generation tool that can pull from public photos on the platform to create new images. The feature went live quietly, and privacy experts quickly raised alarm. As reported by The Guardian and Yahoo Finance UK, the default setting for public accounts is to allow your photos to be used as training material for this AI. If you post publicly on Instagram, your images may now be fed into a system that can produce altered or entirely new visuals that resemble you or your content.

This isn’t a hypothetical risk. An analysis by Reuters, covered by Yahoo Finance, showed that Meta’s own AI detection tool fails to identify cropped AI-generated images, meaning it’s currently difficult to tell if an image is real or fabricated. The combination of automatic opt-in and weak detection makes the situation worth addressing promptly.

What happened

Instagram’s AI image generator allows users to type a prompt and receive a generated image built from patterns in existing public photos. According to inkl, most public accounts are included unless the account holder manually opts out. Meta has not asked for explicit consent before enabling this. The feature was introduced in the July 2026 app update, and many users first learned about it from news reports rather than an in-app notification.

Why it matters

The main concern is consent. When you upload a photo of your face, your home, or your children, you are not expecting it to be used as raw material for someone else’s AI prompt. With the generator, another user could create a deepfake-style image that looks like you—without your knowledge. Privacy experts have pointed out that this opens the door to harassment, impersonation, and misuse.

Additionally, Meta’s AI image detector is not reliable. The Reuters analysis found that cropping an AI-generated image was enough to fool the detection system. This means even if Meta wanted to flag abusive deepfakes, they currently lack a robust tool to do so.

What you can do about it

The good news is that you can opt out. The process takes about a minute, but you need to make sure your Instagram app is updated to the latest version (the setting may not appear on older builds).

Step-by-step to opt out:

  1. Open the Instagram app and go to your profile.
  2. Tap the three horizontal lines (menu) in the top right corner.
  3. Select Settings and privacy.
  4. Scroll down and tap Privacy.
  5. Look for AI and Data (it may be near the bottom).
  6. Under “AI generation from your photos,” you will see a toggle labeled Allow AI generation from your photos. It is likely turned on by default.
  7. Tap the toggle to turn it off.

After you turn it off, your photos should no longer be used as inputs for the AI image generator. Note that this does not remove any images already used before the opt-out, but it stops future use.

Additional steps to protect your images:

  • Consider switching your account to private. Private account photos are generally excluded from the AI training set (though confirm this in the AI and Data settings—some cross-app sharing with Facebook may still apply).
  • Remove geotags from posts you do not want to reveal location data.
  • Watermark your images with a subtle overlay. This won’t prevent the AI from using them, but it makes generated copies easier to trace.
  • Review your Facebook and Meta accounts. The AI generator may pull data across platforms if your accounts are linked. Check the same AI and Data settings on Facebook as well.

What to watch for

Privacy advocates have already called for regulatory scrutiny, and it’s possible that Meta will face pressure to change this to an opt-in model. In the meantime, keep an eye on your settings after app updates—defaults can sometimes revert. If you post publicly, consider whether the trade-off is worth it.

The deeper issue is that platform-level AI training is becoming the norm, and the burden is on users to opt out rather than on companies to ask first. Being aware of how your data is used is the first step; acting on that awareness is the second.

Sources

  • The Guardian, “Instagram’s AI image generator alarms privacy experts,” July 9, 2026
  • Yahoo Finance UK, “Instagram has just allowed anyone to alter your images with AI. Here’s how to opt out,” July 10, 2026
  • inkl, “Meta’s New AI Can Turn Instagram Photos Into Deepfakes. Most Public Users Are Included Unless They Opt Out,” July 9, 2026
  • Yahoo Finance UK (via Reuters), “Meta AI image detector fails to identify some of its own cropped AI images, Reuters analysis finds,” July 10, 2026