A potential X ban in the U.K. would trample free speech and be impractical
In January 2025, the social media platform X sent yet another shockwave around the world. Reports circulated that X’s AI bot, Grok, had generated sexually explicit and graphic content of women and children being undressed without their consent. X then monetized this content by restricting the Grok image generation feature behind a paywall. This horrifying revelation marked yet another episode in long-ranging discussions around social media regulation. In response to these incidents, politicians across the aisle in the United Kingdom, ranging from Prime Minister Keir Starmer to Reform Party leader Nigel Farage, have condemned X’s actions. Britain’s media regulator Ofcom, becoming increasingly concerned about Grok’s production of these inappropriate images, has threatened to block X in the U.K. if the company does not immediately address this issue. Technology Secretary Liz Kendell said she would support such an action by Ofcom, claiming that the Online Safety Act of 2023 grants Ofcom such authority. While seemingly beneficial, a ban on X would undermine democratic and legal principles of free speech and would not be the most effective solution to resolve social media safety concerns.
Despite government officials’ grievances, banning X would set a dangerous precedent for violating democratic norms on free speech. Over 400 million people use X each year, so a potential X ban would prevent all these users from accessing a crucial mode of communication and information dissemination, key components of a functioning democracy. Not only has Elon Musk, the owner of X, pushed back on the potential ban for infringing upon free expression, but free speech advocates themselves have also expressed concern, adding to the credibility of this criticism. As Ari Cohn, the lead counsel for tech policy at the Foundation for Individual Rights and Expression (FIRE), explains, “X and Grok are tools for communication, much like printing presses and cell phones are tools for communication… Free nations that claim to honor the expressive rights of their citizens must recognize that mass censorship is never an acceptable approach to objectionable content or illegal conduct.” Britain’s potential X ban follows a dangerous example from authoritarian regimes like Russia and China, which have implemented mass media bans. Following in these authoritarian footsteps would further empower the U.K.’s regulatory-heavy media landscape, posing a grave threat to the bedrock of British democracy: free and independent media and the right to political participation.
In addition, an X ban is illegal under U.K. law. While the Online Safety Act of 2023 does allow Ofcom to ban social media platforms for noncompliance with content moderation, it does not necessarily allow Ofcom to take such action under just any circumstances. For one, the Online Safety Act did not anticipate the rapid development of artificial intelligence (AI) in the last few years, so the legislation does not mention the use of such tools. The lack of provisions addressing AI has left Ofcom’s authority under the Act ambiguous. Consistent with the British legal doctrine of a harmonious interpretation of U.K. laws (the idea that judges should defer to the Human Rights Act of 1998 when interpreting conflicting or ambiguous laws), one must interpret the Online Safety Act in the context of Article 10 of the Human Rights Act of 1998, which outlines specific exceptions to freedom of expression based on European Union principles. While the Human Rights Act would allow Ofcom to ban social media, in this case, to safeguard public safety, it only permits such an action if it is “necessary in a democratic society”. The ultimate test of whether Ofcom can and should take action lies in the practical determination of whether the U.K. truly needs it.
Indeed, the multitude of other more targeted and effective regulatory options available proves that an X ban is unnecessary and thus illegal under the Human Rights Act. Ofcom’s threats to ban X appear to be a knee-jerk reaction before Ofcom has seriously considered alternatives. For example, Ofcom may consider referring this case to investigators who can prosecute individuals using the software for malicious purposes, as the objectionable content on Grok falls beyond the issue of speech content and under the purview of criminal activity. Ofcom should also mandate that X take down Grok until the company can revise its algorithms to prevent the generation of sexually exploitative content.
Ultimately, even if public safety supersedes the right to free expression, which is a valid viewpoint, an X ban would be ineffective and impractical. These specific courses of action, including criminal referrals and ordering a revision of X algorithms, all better address concerns on public safety than a blanket ban would. Even under a ban, users can use alternative tools like VPNs to bypass an X blockage, negating the purpose of the ban. Even if Ofcom were to successfully thwart these alternative access methods, its regulatory decision would not change the fact that Grok remains widely used in various other countries around the world. Thus, the U.K. government would be merely shirking responsibility for our collective global safety—a classic case of moral hazard in policymaking.
Fortunately, Ofcom does not appear ready to follow through on its threat to ban X just yet. Ofcom announced that X would still remain active pending a formal investigation and a request for an explanation. If Ofcom does ban X, it would be responsible for a worst-case scenario, where neither the democratic right to free speech nor public safety is fully protected, and set a dangerous future precedent for government overreach, where agencies are permitted to take regulatory actions disproportionate to the actual problem.
