LONDON — U.K. ministers are warning Elon Musk’s X it faces a ban if it doesn’t get its act together. But outlawing the social media platform is easier said than done.

The U.K.’s communications regulator Ofcom on Monday launched a formal investigation into a deluge of non-consensual sexualized deepfakes produced by X’s AI chatbot Grok amid growing calls for action from U.K. politicians.

It will determine whether the creation and distribution of deepfakes on the platform, which have targeted women and children, constitutes a breach of the company’s duties under the U.K.’s Online Safety Act (OSA).  

U.K. ministers have repeatedly called for Ofcom, the regulator tasked with policing social media platforms, to take urgent action over the deepfakes.

U.K. Technology Secretary Liz Kendall on Friday offered her “full support” to the U.K. regulator to block X from being accessed in the U.K., if it chooses to.

“I would remind xAI that the Online Safety Act Includes the power to block services from being accessed in the U.K., if they refuse to comply with U.K. law. If Ofcom decide to use those powers they will have our full support,” she said in a statement.

The suggestion has drawn Musk’s ire. The tech billionaire branded the British government “fascist” over the weekend, and accused it of “finding any excuse for censorship.”  

With Ofcom testing its new regulatory powers against one of the most high-profile tech giants for the first time, it is hard to predict what happens next.

Not going nuclear — for now  

Ofcom has so far avoided its smash-glass option.

Under the OSA it could seek a court order blocking “ancillary” services, like those those processing subscription payments on X’s behalf, and ask internet providers to block X from operating in the U.K.  

Taking that route would mean bypassing a formal investigation, but that is generally considered a last resort according to Ofcom’s guidance. To do so, Ofcom would need to prove that risk of harm to U.K. users is particularly great. 

Before launching its investigation Monday, the regulator made “urgent contact” with X on Jan. 5, giving the platform until last Friday to respond.

Ofcom stressed the importance of “due process” and of ensuring its investigations are “legally robust and fairly decided.”  

Limited reach  

The OSA only covers U.K. users. It’s a point ministers have been keen to stress amid concerns its interaction with the U.S. First Amendment, which guarantees free speech, could become a flashpoint in trade negotiations with Washington. It’s not enough for officials or ministers to believe X has failed to protect users generally.  

The most egregious material might not even be on X. Child sexual abuse charity the Internet Watch Foundation said last week that its analysts had found what appeared to be Grok-produced Child sexual abuse material (CSAM) on a dark web forum, rather than X itself — so it’s far from self-evident that Ofcom taking the nuclear option against X would ever have been legally justified.  

X did not comment on Ofcom’s investigation when contacted by POLITICO, but referred back to a statement issued on Jan. 4 about the issue of deepfakes on the platform.

“We take action against illegal content on X, including Child Sexual Abuse Material (CSAM), by removing it, permanently suspending accounts, and working with local governments and law enforcement as necessary. Anyone using or prompting Grok to make illegal content will suffer the same consequences as if they upload illegal content,” the statement said.

Big test  

The OSA came into force last summer, and until now Ofcom’s enforcement actions have focused on pornography site providers for not implementing age-checks. 

Online safety campaigners have argued this indicates Ofcom is more interested in going after low-hanging fruit than challenging more powerful tech companies. “It has been striking to many that of the 40+ investigations it has launched so far, not one has been directed at large … services,” the online safety campaign group the Molly Rose Foundation said in September.  

That means the X investigation is the OSA’s first big test, and it’s especially thorny because it involves an AI chatbot. The Science, Innovation and Technology committee wrote in a report published last summer that the legislation does not provide sufficient protections against generative AI, a point Technology Secretary Liz Kendall herself conceded in a recent evidence session. 

Political risks 

If Ofcom concludes X hasn’t broken the law there are likely to be calls from OSA critics, both inside and outside Parliament, to return to the drawing board.

It would also put the government, which has promised to act if Ofcom doesn’t, in a tricky spot.  The PM’s spokesperson on Monday described child sexual abuse imagery as “the worst crimes imaginable.”

Ofcom could also conclude X has broken the law, but decide against imposing sanctions, according to its enforcement guidance.

The outcome of Ofcom’s investigation will be watched closely by the White House and is fraught with diplomatic peril for the U.K. government, which has already been criticized for implementing the new online safety law by Donald Trump and his allies.

Foreign Secretary David Lammy raised the Grok issue with U.S. Vice President JD Vance last week, POLITICO reported. 

But other Republicans are readying for a geopolitical fight: GOP Congresswoman Anna Paulina Luna, a member of the U.S. House foreign affairs committee, said she was drafting legislation to sanction the U.K. if X does get blocked.