Did you know Meta's AI Feature Sparks Concern Over Fake User Images Without Consent

 

A user on Reddit is questioning Meta’s motives after sharing screenshots of a bizarre feature that the tech giant is testing.

The feature fills users’ feeds with plenty of AI pictures in fictional scenarios. It arose after the Reddit user was trying to use Meta AI to make edits to images of themselves. As per journalists from 404 Media, one representative from the company shared that this was not an ad.

It’s a feature Meta unveiled in September but that left the user confused. The company never spoke about a feature that suddenly allocates users’ images made using AI into the feed. At the time, from what we recall, it did share how the goal was to imagine yourself as some kind of superhero in the feed or other places like your profile picture through Facebook.

It’s a feature that enables users to share AI images so friends can see them. As explained, pictures made through Meta AI are designed for you, the company shared. This will all depend on what your current likes and dislikes might be.

At first, many people got skeptical. Users shared inside comments how the matter was bizarre as to have themselves put into a scenario all of a sudden didn’t make too much sense. But a spokesperson for the firm says it’s nothing to worry about. The feature is still in the experimental phase and comes under the Imagine Yourself offering. Thankfully, Instagram was nice enough to share that this was only visible to the user.

One thing is for sure, it’s not ads and we feel that’s a huge relief. Images provided by the user who experienced this showed screengrabs of landscape-like pictures but there’s no indicator from Meta that the user would get inserted here without a warning.

The experiment arose after a small share of AI personalities started to appear on the platform to cause concern. We’ve seen the company rolled out chatbots in the start of December which got less criticism. Thankfully, they didn’t entail any visuals made using AI that we can recall.

The world of generative AI is trying to insert itself into every inch of the web. However, this is the first time that an app went to the extent of serving users fake pictures of themselves without any type of mention or warning/labels beforehand.

Generative AI is spreading across the internet, but this may be the first instance of a social media platform showing users fake images of themselves without clear consent. According to 404 Media, Snapchat might use AI-generated images of users' faces in ads. Jason Koebler from 404 Media, tested this by uploading his selfie to Meta AI during the "Imagine Yourself" onboarding. As of now, he hasn't seen any images of himself in a virtual room of endless mirrors.

 

 

 Image: Green_Video_9831 / Reddit