Elon Musk’s artificial intelligence chatbot Grok has disabled its image creation feature for non-paying users following mounting international condemnation over its exploitation to generate sexualised deepfakes of women and children, a development that has placed the billionaire entrepreneur and his social media platform X under intense regulatory scrutiny across multiple jurisdictions.
The restriction, implemented quietly over the weekend, marks a significant retreat for a tool that Musk had promoted as embodying a freer approach to artificial intelligence, one less constrained by what he has repeatedly characterised as excessive content moderation. The move comes after reports emerged that some users had weaponised Grok’s image generation capabilities to create explicit imagery depicting women and children undressed, sometimes positioned in overtly sexualised contexts, raising alarm among child safety advocates, women’s rights organisations, and government regulators worldwide.
Responding to users on Friday via Musk’s social media platform X, Grok posted: “Image generation and editing are currently limited to paying subscribers. You can subscribe to unlock these features.” The announcement represents a sharp policy reversal for a platform that has positioned itself as a champion of minimal content restrictions, and underscores the legal and reputational risks associated with unregulated generative artificial intelligence tools capable of producing realistic imagery.
The change means that a substantial portion of Grok’s user base can no longer generate or edit images using the AI system. Only paying subscribers who have provided the platform with credit card information and personal identification details now retain access to the controversial feature, a gatekeeping mechanism that critics argue may be insufficient to prevent further abuse given the determined nature of those seeking to create illegal content.
The European Commission escalated the matter this week, declaring that the photographs of undressed women and children generated through Grok were unlawful under European Union regulations. On Thursday, the Commission formally ordered X to retain all internal documents and data related to Grok until the end of 2026, a move typically associated with investigations into potential violations of the bloc’s Digital Services Act, which imposes strict obligations on large online platforms to prevent the dissemination of illegal content.
The preservation order signals that European regulators are preparing for a potentially lengthy investigation into X’s content moderation practices and the adequacy of safeguards built into Grok’s image generation system. Under the Digital Services Act, platforms can face fines of up to six percent of their global annual revenue for systemic failures to address illegal content, a financial penalty that could run into billions of dollars given X’s valuation and operational scale.
United Kingdom Prime Minister Keir Starmer joined the chorus of official criticism, stating that X has “got to get a grip of this” and revealing that he had asked communications regulator Ofcom “for all options to be on the table,” according to media reports. Starmer characterised the images as “unlawful” and declared that Britain was “not going to tolerate it,” language that suggests potential regulatory action against the platform if voluntary compliance measures prove inadequate.
The British government’s intervention reflects growing impatience among Western democracies with social media platforms perceived as prioritising growth and engagement over user safety, particularly when vulnerable populations such as children are placed at risk. Ofcom, which gained expanded enforcement powers under the Online Safety Act passed in 2023, has authority to impose substantial fines and operational restrictions on platforms that fail to prevent the distribution of illegal content, including child sexual abuse material and non-consensual intimate imagery.
France, Malaysia and India have also publicly criticised Musk’s platform over the deepfake controversy, reflecting the global nature of concerns surrounding generative AI and its potential for harm. French digital affairs officials have reportedly been in contact with European Commission counterparts to coordinate a unified response, while Malaysian authorities have indicated they are reviewing whether X’s operations comply with the country’s communications and multimedia regulations. Indian officials, meanwhile, have raised the issue through diplomatic channels, according to sources familiar with the matter.
The international backlash represents a significant challenge for Musk, whose ownership of X has been marked by contentious battles with regulators over content moderation policies. Since acquiring the platform formerly known as Twitter in October 2022 for approximately $44 billion, Musk has repeatedly clashed with governments and civil society organisations over his approach to free speech, which critics argue has resulted in insufficient action against hate speech, misinformation, and now, sexually explicit deepfakes.
Responding to the mounting criticism last week, Musk wrote on X: “Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content.” The statement appeared designed to distance the platform from user behaviour while asserting that enforcement mechanisms were in place, though it did not address questions about whether Grok’s design incorporated adequate safeguards to prevent such abuse in the first instance.
X’s official “Safety” account subsequently issued a statement saying the platform addresses illegal content on X “by removing it, permanently suspending accounts, and working with local governments and law enforcement as necessary.” The assurance, however, has done little to mollify critics who argue that reactive content removal is insufficient when dealing with artificial intelligence tools capable of generating harmful imagery at scale and speed.
The controversy surrounding Grok’s image generation feature is part of a broader reckoning within the technology industry over the societal implications of generative AI. Since the public release of tools like OpenAI’s DALL-E, Midjourney, and Stable Diffusion over the past two years, concerns have mounted about their potential misuse for creating non-consensual pornography, impersonating public figures, spreading disinformation, and producing child sexual abuse material.
Many mainstream AI image generators have implemented strict content filters and usage policies designed to prevent the creation of explicit imagery, particularly involving real individuals or minors. These safeguards typically include pre-trained filters that reject prompts containing certain keywords, automated systems that scan generated images for prohibited content, and human review processes for flagged material. Some platforms also employ digital watermarking and provenance tracking to help identify AI-generated content.
Grok’s apparent lack of comparable restrictions prior to the current controversy has raised questions about X’s approach to responsible AI development. Critics argue that Musk’s stated commitment to minimal content moderation, while appealing to free speech absolutists, created an environment where malicious actors could exploit powerful generative tools with insufficient oversight.
The restriction of image generation to paying subscribers represents one potential mitigation strategy, as requiring payment creates an accountability trail that may deter some abusers and facilitate law enforcement investigations. However, experts in online safety and artificial intelligence ethics have expressed scepticism that payment barriers alone will prove sufficient, noting that determined offenders have historically demonstrated willingness to pay for access to platforms that enable illegal activity.
Dr. Sarah Chen, a researcher specialising in AI ethics at Stanford University, told reporters this week that “requiring payment is a step, but it’s not a solution. What’s needed are robust technical safeguards built into the AI model itself, including classifiers that can detect and block attempts to generate sexualised imagery of minors, and systems that can identify when real individuals are being depicted without consent.”
The regulatory pressure on X and Grok also reflects growing governmental assertiveness in holding technology companies accountable for the downstream effects of their products. The European Union’s Digital Services Act, which came into full force in 2024, represents the most comprehensive attempt to date to regulate large online platforms, imposing transparency requirements, content moderation obligations, and mechanisms for independent auditing and oversight.
Similar legislative efforts are underway or being considered in numerous other jurisdictions, including Canada, Australia, and several Southeast Asian nations, as governments seek to establish clearer rules for platform responsibility in an era of rapid technological change. The United Kingdom’s Online Safety Act, which received Royal Assent in 2023 and has been progressively implemented, gives Ofcom unprecedented authority to require platforms to proactively identify and mitigate risks, including through the use of technology to detect illegal content.
For Musk personally, the Grok controversy adds to a growing list of regulatory challenges that have complicated his ownership of X. The platform has faced investigations or sanctions in multiple countries over content moderation practices, with the European Commission having previously threatened substantial fines over alleged failures to comply with Digital Services Act requirements related to misinformation and transparency.
The billionaire’s confrontational approach to regulation, which has included public criticism of government officials and assertions that content restrictions constitute censorship, has won him support among certain political constituencies but has also deepened tensions with regulators who view platform accountability as essential to protecting public safety and democratic institutions.
The immediate future of Grok’s image generation capabilities remains uncertain. While paying subscribers currently retain access, there is no indication whether X plans to implement additional technical safeguards beyond the payment requirement, or whether the feature might face further restrictions if abuse continues. The European Commission’s document preservation order suggests that regulators are preparing for potential enforcement action that could result in more stringent requirements or operational limitations.
Child safety organisations have called for Grok’s image generation feature to be suspended entirely until comprehensive protections can be demonstrated. The Internet Watch Foundation, a UK-based organisation that works to eliminate child sexual abuse content online, issued a statement describing the reported abuse of Grok as “deeply alarming” and calling on X to “implement industry-standard safety measures before allowing any further image generation.”
