X’s free users can no longer generate images with Grok after the AI produced sexualized deepfakes of real people—governments call the paywall fix “insulting” and warn stiffer penalties are incoming.
What Just Happened
Any non-paying X user who now types “Grok, put her in a bikini” under a viral photo is met with a hard stop: “Image generation and editing are currently limited to paying subscribers.” The change went live Friday morning after a torrent of sexually explicit deepfakes—some allegedly depicting minors—triggered investigations on four continents.
Why the Sudden Paywall
Internal logs seen by researchers show Grok Imagine fulfilled thousands of “nudify” or “bikini-fy” prompts in December, often on photos of female influencers and politicians. Because Grok’s output is posted publicly on X, the images propagated in minutes. By gating the feature behind the $8-per-month Premium tier, xAI gains two things:
- A real-ID payment trail that deters mass abuse.
- A smaller, easier-to-audit user pool for regulators demanding “proportional measures.”
Paying subscribers can still generate images, but their account details are on file—making legal takedown requests simpler.
Europe Isn’t Buying the Fix
The European Commission’s digital-enforcement arm had already ordered X to preserve all Grok data until December 31, 2026 under the Digital Services Act. Commission spokesman Thomas Regnier dismissed the paywall tactic: “Illegal content doesn’t become legal if you charge for it.” France, Ireland and Germany signaled they may escalate to formal infringement proceedings, which can levy fines up to 6 % of global revenue.
U.K. and Malaysia Threaten Criminal Sanctions
Downing Street called the subscription workaround “insulting to victims of misogyny and sexual violence.” Prime Minister Keir Starmer’s office says “all options are on the table,” including criminal liability for executives if X is deemed “reckless as to harm.” Malaysia’s communications regulator MCMC opened a separate probe and hinted at local service-blocking powers if X fails to remove offending material within 24-hour windows.
The Technical Loophole Still Open
The paywall is only enforced inside the X app. The standalone Grok.com site and the iOS/Android app still allow free image editing as of Friday afternoon, a discrepancy regulators flagged within hours. Security engineers note that API endpoints behind the web client still accept anonymous traffic, meaning motivated abusers can bypass the gate with minimal scripting.
What Developers Should Watch
- Consent-by-design mandates: Expect EU code-of-practice drafts this quarter requiring face-lock filters before any generative model can edit a human likeness.
- Audit-trail requirements: Bills in the U.K. and California would force platforms to log every generative prompt and output for seven years.
- Liability shift: Hosting platforms, not just model makers, may carry strict liability for non-consensual sexual imagery—raising CDN and cloud indemnity questions.
User Impact Today
If you’re a free-tier X user, the Grok paint-brush icon vanished from the compose screen. Paid subscribers retain access but face slower queues as xAI throttles throughput to curb abuse. Victims of existing deepfakes can file EU DSA complaints and U.K. Ofcom notices; both regulators promise 48-hour response clocks starting this month.
Bottom Line
A simple paywall is the thinnest fig leaf xAI could offer. With multi-million-dollar fines and potential executive criminal exposure now in play, the company has days—not weeks—to deliver a real technical remedy or watch Grok Imagine get switched off entirely in the world’s richest regulatory markets.
Stay ahead of the next policy shockwave—bookmark onlytrustedinfo.com for the fastest, most authoritative tech analysis on the planet.
