Videos of weird AI glasses are still prevalent on Instagram despite Meta’s pledge to stop them. – Whether you’re asking an AI assistant about what’s in front of you, taking a hands-free picture, or listening to music, smart glasses may be really helpful. Sadly, placing a camera on someone’s face might also lead to some very unpleasant applications. For Meta, that is starting to cause issues.
Adam Mosseri, the head of Instagram, stated in July that the company was trying to get rid of videos showing users secretly filming and harassing others using Meta glasses. But according to a recent Business Insider investigation, the cleanup isn’t proceeding as easily as Meta had intended. Weeks after Mosseri’s remarks, the magazine discovered that numerous pickup and practical joke videos were still accessible on Instagram. Several included women being accosted and captured on camera in public using Meta glasses, while others came from well-known or verified creators.
Instagram is acting, but videos are not getting thru.
Meta has begun cleaning the house already. According to reports, thousands of posts have been removed from Instagram, some larger accounts have vanished, and searches for terms like “rizz” and “cold approach” are now blocked. Indeed, that seems like a really hostile reaction. However, a lot of identical movies are still getting thru in practice. Things get messy in that gap. During Business Insider’s inquiry, Meta received over 20 potentially questionable videos, but only nine of them were eventually taken down. Additionally, there have been several perplexing account-level rulings. After being deactivated, one creator caught up in the crackdown momentarily reappeared on Instagram before being removed once more after the error was discovered.
What happens after watching one of these clips might be the more difficult problem. Instagram’s algorithm is designed to find out what makes you scroll, which may undermine its attempts at moderation. Reels started providing a deluge of comparable content after users interacted with a few videos throughout the inquiry. At that point, Instagram was locating these videos on its own, so it wasn’t necessary to search for particular accounts or keywords.
This could cause Meta more headaches.
This is a moderation issue for Instagram. It might also turn into a hardware issue for Meta as a whole. According to reports, sales of Meta’s smart glasses exceeded 7 million units in 2025, more than tripling the previous year. They also play a significant role in Mark Zuckerberg’s larger initiative for wearable computing driven by AI. Because of this, their reputation is especially crucial. Meta has a lot greater problem than a few dubious Reels if users start to associate camera-equipped glasses with being unintentionally filmed on the street.
Meta is placing a wager that AI will someday be able to enhance its moderating on a large scale. According to the business, preliminary testing revealed that its AI systems were able to identify more content violations while making less errors than human moderators. It remains to be seen if that results in a more hygienic Instagram experience. As of right now, Meta has made a clear distinction about this type of information. Making ensuring Instagram can enforce it is challenging, particularly as millions more camera-equipped glasses end up on people’s faces.
About the Author
