X's Grok AI Edit Button Is Stripping Artist Watermarks, and Creators Are in Open Revolt

When a social media platform hands every user an instant, one-click tool to alter someone else's artwork, the fallout arrives fast.

On December 24, 2025, X rolled out a new feature powered by Grok AI that allows any user to edit any image posted on the platform with a single click. The tool works through a simple long-press on mobile or an "Edit" button on the web. Users can enter prompts to modify an image, whether it's their own content or someone else's, then post or reply with the AI-altered version.

Elon Musk promoted the feature on Christmas Day, posting "Try Grok image edit and video edit" to his followers. The platform now lets any user AI-edit any public image on X, with zero consent required from the original poster.

That alone would have been enough to enrage the creative community. What happened next made it worse.

The tension reached a breaking point after an incident involving an artist on X known as Iomaya, whose paid commissioned artwork was hijacked by an anonymous user and altered using the Grok AI tool. In the modified image, the user scrawled the word "Luddite" across the character's abdomen. Iomaya's thread garnered over five million views and more than 15,000 reposts, drawing a wave of global support from the creative community.

The account behind the manipulations, El3v3nDimesion, targeted multiple creators before being deactivated.

There is no way to opt out or disable the option for others to edit posted images, no notifications are sent when someone edits original work, and no protection exists for watermarks. Even with all Grok and AI settings disabled in privacy preferences, other users can still use the AI edit button on an artist's images. That detail landed like a gut punch.

Among the most prominent voices was Mu-jik Park, the South Korean manga artist known by his pen name Boichi, illustrator of Dr. Stone and Sun-Ken Rock. Park announced he would stop publishing manga pages and illustrations on X, citing concerns over how his work could be used without consent or proper compensation, and saying the decision was made "with a heavy and broken heart." He said he would remain active on X for communication but would publish his comics and illustrations on Instagram instead.

A day later, Kei Urana, the manga artist behind Gachiakuta, announced she would also be moving her art posts primarily to Instagram. Chrono Crusade creator Daisuke Moriyama went further, tweeting that he would delete all of his artwork from X entirely.

Protective tools are failing, too. Testing shows that Glaze and Nightshade, the AI-poisoning tools artists use to protect their work from being scraped for training, do not prevent Grok from editing protected images, since they were designed for training data poisoning, not real-time editing prevention. Some creators are deleting old images entirely or switching to GIF format as a workaround, though even that approach is far from foolproof.

The ramifications extend well beyond illustration. Grok is now racing to fix flaws in its image-editing tool after users reported it could turn pictures of children or women into sexualized content. Grok acknowledged the problem, stating, "We've identified lapses in safeguards and are urgently fixing them," and adding that "CSAM is illegal and strictly prohibited." Authorities in India and France have begun investigations into the misuse of AI-generated content on the platform.

For digital artists, the calculus is now painfully simple. Posting original work on X means offering it up to anyone with a text prompt and a click. No consent required. No credit preserved. Until X introduces opt-out controls or meaningful guardrails, the exodus shows no sign of slowing.