Meta’s new AI tool creates Deepfakes. Here’s how to protect yourself on Instagram

Meta has released a new AI model, and it’s a recipe for potential deep fake disaster and even more AI slope. If you have a public account, you should change your settings now to avoid becoming the unwitting subject of anyone’s AI creations.

The template, called Muse Image, is the first creative template in the new family of Muse Spark Templates made by Meta Superintelligence Labs. The company said in a blog post that it was designed to handle more complex requests, create composite photos and edit existing images. It is now available on the Meta AI app, Instagram and WhatsApp, with plans to eventually offer it on Facebook, Messenger and advertisers.

If you are over 18 and have a public account, anyone with a Meta AI account can “tag” you in their AI image prompts and create hyper-realistic AI images including your image – otherwise known as deep fakes. I tried out the new template this week and was able to create a deepfake of my CNET colleague Abrar Al-Heeti in less than a minute by including his Instagram username in my prompt.

The backlash against Meta’s new deepfaking AI was instantaneous and intense, ranging from regular Instagram users to advocacy and trade groups. Actors union SAG-AFTRA told the LA Times that automatic opt-in to the tool was “a complete miscalculation of public opinion regarding the obvious dangers and harms inherent in such use.” Haley McNamara, executive director of the National Sexual Exploitation Technology Monitoring Center, said: “How could this be green-lighted? [Meta is] creating obvious and predictable opportunities for exploitation, sexual abuse, harassment and identity theft.

This isn’t the first time an AI company has tried to entice people to use its creative AI by offering to place you and your friends in the AI ​​landscape. This was OpenAI’s speech at its launch its ill-fated Sora video app in 2024. But OpenAI always attracted anger from ordinary people and celebrities for its role in easily creating deepfakes. Meta’s new AI model poses the same risk.

These controls will be essential for professional creators and influencers, whose names and likenesses are their brand and therefore their livelihood. Meta claims that its models have built-in protections to prevent the model from creating illegal, abusive, or defamatory content. But as we saw with Soramotivated bad actors can circumvent a model’s safeguards. We’ll have to wait and see if the Meta are up to the challenge.

How to Disable Meta AI Deepfaking: Change These Instagram Settings

There are two main ways to unsubscribe. The first is to turn your public account into a private account. Here’s how.

1. Open Instagram and go to your profile.
2. Tap the three lines in the upper right corner to open your settings.
3. Faucet Account Privacy.
4. To light up Private account.

Private accounts don’t automatically make their content available to anyone who can remix or create it. But it will also make it harder for other people to discover and interact with your account. If you need to keep your account public (for example if you’re a creator or business), here’s what you can do to protect yourself.

1. Open Instagram and go to your profile.
2. Tap the three lines in the upper right corner to open your settings.
3. Scroll to a category called Sharing and reuse.
4. Disable setting “Let people reuse your content on Instagram and with AI features on Meta“.
5. You can do this to block Meta AI’s access to your posts and reels.

If you use the Meta AI app, you can also change your settings there. You can adjust how your image can be used in the app by going to Settings > Your image. You can choose who is allowed to use your image, including just yourself, followers you approve, mutuals, or everyone.