Have you ever wanted to animate your profile picture on Facebook? Or turn your latest Instagram upload into a cartoon-inspired moment? Well, you’re in luck. Meta launched an AI photo app that allows users to “create AI images in seconds,” according to the Meta website.
The tool allows users to create AI images from a prompt or use the tool to edit their own Facebook or Instagram photos by implementing features like photo animations or photo background changes.
The tool isn’t brand-new; it started popping up on Facebook last year. Yet, many platform users are just recently being met with a notification alerting them to the new product. The notification reads: “Your camera just got smarter. The new Meta AI app understands and interprets your photos.”
While some people aren’t too concerned with this notification, some social media users are freaked out by the idea of AI and the social media apps “understanding” and “interpreting” their photos. And cybersecurity experts told HuffPost that people aren’t wrong for being concerned.
Here’s what to know:
The photo app allows for AI editing on your Facebook and Instagram pictures and AI photo creation.
“Meta is trying to catch up with what Google and Apple has already done with their phones and their photo programs,” said JP Castellanos, the director of threat intelligence at Binary Defense. This is Meta’s way of offering AI photo editing.
“It’s not just a normal Instagram filter … it’s going to be AI,” Castellanos said. “Your data, your photos and your videos are basically taken from your camera roll, and then they’re going to be uploaded into Meta servers so then meta AI can then start analyzing them and making suggestions.”
This could also include photos that are on your camera roll but you’ve never posted publicly, Castellanos added.
If you’re thinking, “Doesn’t Meta already have all of my photos and videos?” ― that is a good question. But Castellanos noted that the Meta AI servers are different than the Meta servers, so your data would now be in two different places. That is, if you opt in to the feature.
“This new feature is not going to be used for any ad targeting or it’s not going to be used to improve its AI models unless you wish to share that information,” Castellanos said. So you have to opt in for your photos and videos to be used in this manner.
“It’s like any new feature,” said Sean Gorman, the CEO of Zephr.xyz. “You want to look at the terms of service and understand how they’re using the data.”
Gorman said he doesn’t think this one AI photo feature poses any additional privacy threats.
“The challenge is everybody ― whether it’s Meta, Google, OpenAI, Anthropic ― are trying to figure this out. I don’t know if we necessarily know what the repercussions are going to be,” Gorman said.
It all depends on how comfortable you are as a user; the trade-off is using these free tools in exchange for the data you provide. The difference here, though, is the trade-offs — the ability for targeted ads, the understanding of your habits — are accelerating more rapidly with AI, Gorman said.
If you are OK with that and like the personalization that comes with these social media apps having more and more of your information, you’ll probably be excited about this new Meta feature, according to Gorman. But if you don’t like the idea of these companies having a large amount of information about you, you probably will want to opt out.
SOPA Images via Getty Images
Whether or not you use this photo feature is up to you, but it could come with security risks, experts say.
“It’s not just a fun photo filter, Castellanos said, it’s also a potential privacy issue in regard to your camera roll.
You’re giving separate Meta servers access to your photos and videos. But it is worth noting that some Meta servers may already have access to this data if you have a Facebook or Instagram account. So, opting out of the Meta AI photo app won’t limit permissions everywhere.
“From a cybersecurity perspective, I would urge caution because your camera rolls can contain more than just your selfies or your vacation photos,” Castellanos said. It may include screenshots of private conversations, photos of your kids, images of receipts or medical paperwork. This is all sensitive material that you’d be sharing with Meta AI, he noted.
Unless you specifically want Meta AI to use your photos and have access to your camera roll, Castellanos recommends turning off the AI photo feature and “limiting Instagram and Facebook to just selected photo access only, and going from there.” You can alter your photo permissions in the app settings.
“Not to say that Facebook probably doesn’t already have your data, but now you’re just giving them more information,” Castellanos said.
Since so much of this data is already out there, specifically for those who already use Meta social media apps, Gorman isn’t too concerned with the privacy aspect of the new Meta AI photo feature.
“For people that are already on these services, I don’t think it’s a watershed type of event,” Gorman noted. “There’s a bigger question that we all need to ask ourselves at a social level, political level, personal level, as to how you want social media integrated into society.”
Does this new Meta AI photo feature change the debate around social media?
“I don’t think that it does,” Gorman said.
“It’s fascinating technology, and there’s some cool stuff, and maybe, as it evolves and these models get smarter and more capable, there is a future where that poses new questions that we need to address. But at least with what this is specifically, I think it doesn’t reframe the debate,” Gorman said. “It just is another signpost of where these systems are going.”
It’s up to you if you want to use Meta’s AI photo feature. While experts are torn on whether it comes with inherent security risks, it is up to you if you’re comfortable with sharing more and more data with these companies and their emerging AI.
