In the latest development, Elon Musk’s AI chatbot Grok has restricted image generation and editing on X to paying subscribers after it was used to create non-consensual sexualized images, including deepfakes and child sexual abuse material, prompting widespread backlash. Critics say the change does not address core safety shortcomings in the model and may simply shift abuse behind a paywall rather than preventing it. The episode is intensifying regulatory scrutiny of X and xAI, raising legal, compliance, and brand-safety stakes for the business as governments weigh enforcement options.

Highlights:

• Safeguards questioned: Experts told the Financial Times that xAI lacked adequate safeguards for image generation, while noting a broader industry reality that many AI models are trained on datasets that can contain troubling material unless carefully filtered and governed.
• UK regulator action: Ofcom said it would accelerate an investigation into X as the UK government warned the platform to tackle indecent AI imagery or face severe consequences, according to The Guardian.
• Victims’ criticism: Lawmakers, regulators and victims argued that limiting access to paid accounts is “not effective,” with deepfake expert Henry Ajder telling Fortune that payment details are easy to fake and the approach is reactive rather than aligning the system itself.
• Product access gaps: WIRED reported that despite the “verified users only” messaging, people can still generate images via Grok’s app and website, which experts said risks enabling ongoing abuse.
• Enforcement pressure: UK officials said all options are on the table, including the possibility of restricting X if it cannot “get a grip” on AI sexualized images, CBS News reported.

Image generation and editing are currently limited to paying subscribers - Grok

Perspectives:

• xAI / Grok: The company implemented a restriction that limits image generation and editing to paying subscribers, framing it as a control on who can access the feature. (Financial Times)
• Victims and lawmakers: They argue the paywall approach is inadequate and disrespectful to those harmed because it does not prevent creation of unlawful images and may only make abuse harder to detect. (Fortune)
• UK government / regulators: UK officials warned X to take stronger action, while the regulator Ofcom moved to speed up scrutiny of the platform’s handling of indecent AI-generated imagery. (The Guardian)
• Independent researchers and media: Reporting describes a pattern of repeated Grok-related safety incidents on X, arguing that rapid scaling without robust controls creates recurring governance and reputational problems. (Quartz)

Sources:

• How Elon Musk’s Grok spread sexual deepfakes and child exploitation images - ft.com
• Lawmakers and victims criticize the choice to limit Grok’s AI image generation to paid users as ‘insulting’ and ‘not effective’ - fortune.com
• X Didn’t Fix Grok's ‘Undressing’ Problem. It Just Makes People Pay for It - wired.com
• Grok, Elon Musk’s A.I., Is Generating Sexualized Images of Real People, Fueling Outrage - nytimes.com
• Just like its dad, Grok keeps failing up - qz.com
• Musk’s Grok chatbot restricts image generation after global backlash to sexualized deepfakes - seattletimes.com
• Grok Is Being Used to Mock and Strip Women in Hijabs and Saris - WIRED - google.com
• Elon Musk's Grok bot restricts sexual image generation after global outcry - latimes.com
• U.K. could ban X if platform can't "get a grip" on AI sexualized images - cbsnews.com
• Elon Musk’s X threatened with UK ban over wave of indecent AI images - theguardian.com