Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Meta has a new AI module, this time dedicated to creating and editing AI images. And yes, you can use it on Instagram. But if you have a public account, you need to change your settings now to avoid becoming the unintended subject of anyone’s AI creations.
The model called Muse Image is the first creative model of the new Muse Image family Muse Spark models Made by Meta Super Intelligence Labs. The company said in Blog post It is designed to handle more complex applications, create photo montages, and edit existing images. It’s now available on Meta AI, Instagram, and WhatsApp, with plans to eventually make it available to Facebook, Messenger, and advertisers.
CEO Mark Zuckerberg showed off the new model on his Instagram account on Tuesday. He showed off some of the 30 new AI editing effects the model runs on Instagram Stories, including photos of several Zuckerberg clones, a 360-degree camera show with AI lead Alexander Wang and an exposure photo mode with Andrew Bosworth, Meta’s chief technology officer.
CEO Mark Zuckerberg demonstrates the editing capabilities of the new AI model in Instagram Stories featuring Alexander Wang (center), Andrew Bosworth (right) and several AI versions of himself (left).
This isn’t the first time an AI company has tried to entice people into using creative AI by offering to put you and your friends in the AI scene. This was OpenAI’s offering when it was launched The ill-fated Sora video app In 2024. But OpenAI It’s still infuriating From ordinary people and celebrities for his role in creating deep fakes with ease. Meta’s new AI model poses the same risk.
Let’s move away for a moment from the fact that this new paradigm is likely to lead to this More AI regression On Instagram. The images you upload to the Meta AI application are used to improve the Meta Services. There are important details in the settings that everyone with a public Instagram account should know. If you’re over 18 and have a public account, anyone with a Meta AI account can “tag” you in their AI photo prompts and create hyper-realistic AI photos including your likeness – known as Deep fakes.
How to protect yourself from deepfakes
I tried out the new model to see how easy it is to create a deepfake. My CNET colleague Abrar Al-Hiti has a public Instagram account, and I was able to create an AI hacker profile of her in less than a minute by including her Instagram username in my prompt. When I tried myself, tagging my own Instagram account, the Meta AI couldn’t complete the request.
Although Meta AI and I did not need to get my colleague Abrar Al-Hiti’s permission to make this AI-generated image of her as a hacker, I obtained her consent before including it in this story.
Meta confirmed to CNET that creators with a public Instagram account can prevent people from creating AI content that looks like them by toggling the setting. Go to Instagram Settings > Sharing and Reuse > Turn off “Allow people to reuse your content on Instagram and using AI features in dead” You can set this control for posts and clips. Content from private accounts can’t automatically be accessed by anyone so anyone can remix or recreate it. (After our testing, Hiti has disabled this permission.)
You can also reduce the risk of deepfakes when you tag yourself in a photo request for the first time in the Meta AI app. It will walk you through some steps to help the app recognize you. This includes taking a photo of your face and, optionally, uploading three photos of yourself. In this process, you can choose who is allowed to use your photo, including only you, followers you approve, subscribers, or everyone. You can set this in the app by going to Settings > Your likeness.
These controls will be essential for professional creators and influencers, whose names and likenesses represent their brand and thus their livelihood. Meta says its models have built-in protections to prevent a model from creating illegal, offensive, or defamatory content. but As we saw with SoraMotivated bad actors can circumvent the model’s safeguards. We’ll have to wait and see if Meta is up to the challenge.
