Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Elon Musk’s X is the latest social media site to roll out a feature to call edited photos “edited media,” if Elon Musk’s post is to be believed. But the company hasn’t made it clear how it will make this determination, or whether it includes images that have been edited using traditional tools, such as Adobe’s Photoshop.
Meanwhile, the details of the new section come from a cryptic X post from Elon Musk that, “Edited image alert,” while also sharing a new X announcement made by a unknown X account DogeDesigner. This story is often used as a springboard for new X features, as Musk will also write from there to share stories.
However, information about the new system is scarce. DogeDesigner notes that the new X feature will make it “difficult for media groups to spread fake photos or images.” It also said that the feature is new to X.
Before it was found and called X, the company it was known as Twitter tweeted using misleading, fraudulently modified, or fictitious information as an alternative method of withdrawal. Its principles were not limited to AI, but also included things like “selective editing or cropping or reducing or multiplying, or changing small words,” the website’s director, Yoel Roth, said in 2020.
It is unclear if X follows the same rules or if they have changed significantly to deal with the AI. His help documents currently they say there is a law against sharing fake news, but it’s not encouraged, like The recent deepfake debacle of users who share inappropriate nude photos. In addition, even the White House now shares updated photos.
Calling something a “media hack” or an “AI image” can be a bit of a stretch.
Since X is a playground politics attractionsinternally and externally, some understanding of how the company determines what is “modified,” or perhaps AI or AI-generated, should be documented. Additionally, users should be aware of any conflicts of interest beyond X’s crowdsourced Community Notes.
Techcrunch event
Boston, MA
| |
June 23, 2026
As Meta discovered when it released AI characters in 2024, it’s easy for the machine’s recognition system to get confused. On the contrary, The meter has been found to be installed incorrectly real images with “Made with AI” labels, even if they were not created using artificial AI.
This is because AI features are becoming more and more advanced included being creative tools used by artists and photographers. (New Apple Creator Studio suite, launching todayis a recent example.)
As it turned out, this messed up Meta’s identification tools. For example, Adobe’s cropping tool used to soften images before saving them as JPEG, and activated Meta’s AI analyzer. In another example, Adobe’s Generative AI Fill, which is used to remove things – such as wrinkles in a shirt, or unwanted features – was also causing images to be labeled “Made with AI,” when they were only processed by AI tools.
Finally, Meta has changed its logo to “AI info,” to avoid labeling images as “Made with AI” when they were not.
Today, there is an organization that sets standards for verifying authenticity and digital content, known as C2 PA (Coalition for Content Provenance and Authenticity). There are similar experiments CAIor Content Authenticity Initiative, and Introduction to the Projectfocused on adding visual metadata to social media.
Presumably, the launch of the X would follow a well-known method of identifying the content of AI, but the owner of the X, Elon Musk, did not say what it is. He also didn’t make it clear if he was talking about AI photos, or anything that wasn’t an image being uploaded to the X directly from your smartphone’s camera. It’s unclear if the feature is new, as DogeDesigner claims.
X isn’t the only site struggling with modified media. In addition to Meta, TikTok has also been writing about AI. They are streaming meetings like Deezer and Spotify They’re also developing ways to analyze and categorize AI music, as well. Google Photos uses C2PA showing how the images on his platform were created. Microsoft, BBC, Adobe, Arm, Intel, Sony, OpenAI, and others are on C2PA steering committeewhen many companies join as members.
X is currently undefined membersalthough we have reached out to C2PA to see if this has changed recently. X doesn’t usually respond to requests for comment, but we asked anyway.
