X could face UK ban over deepfakes: what government teams need to know
UK regulator Ofcom is urgently assessing whether to block access to X after its AI chatbot, Grok, was used to generate non-consensual sexualised edits of images. The Technology Secretary said she would support Ofcom if it exercises powers under the Online Safety Act to restrict the platform in the UK.
X has since limited Grok's image editing to paying subscribers. Downing Street called that move "insulting" to victims, while the platform's owner accused the government of seeking "any excuse for censorship".
What triggered the response
Grok could be tagged under posts to "undress" or sexualise images without consent. Despite the change, editing appears to remain available to paid, verified users, and the tool can still be accessed via other parts of X and its standalone app and site.
Ofcom contacted X earlier in the week and set a deadline for an explanation. The regulator says it has received a response and is conducting an expedited assessment.
Legal powers on the table
The Online Safety Act enables Ofcom to seek court orders that block services or disrupt business operations if firms refuse to comply. This can include preventing third parties from helping a platform raise money or be accessed in the UK. These measures exist, but their use at this scale remains largely untested.
The Technology Secretary stated she expects updates "in days not weeks" and confirmed the government would back Ofcom if it moves to block access.
Political reaction
Leaders across parties condemned the creation of sexualised images, including the prime minister who called it "disgraceful" and "disgusting". Others argued a full ban would be an attack on free speech, with calls instead for tougher safeguards and design changes that prevent abuse.
Inside Parliament, some MPs urged the government to stop using X for official messaging. Concerns focused on safety risks for images of women and children and the signal it sends if departments continue using the platform while investigations are under way.
Risks for public bodies using X
There are three fronts to watch: legal exposure under the Online Safety Act, safeguarding obligations for vulnerable groups, and reputational risk if official accounts appear to endorse or rely on a platform under investigation. If access is restricted, service continuity and citizen communications could be disrupted.
Departments that have paid for verification or promoted image-led campaigns on X face additional scrutiny. Any use of Grok or similar tools-even unintentionally via tags-could compound risk.
Immediate actions for departments and agencies
- Review your X footprint. Pause posts that include images of women and children. Disable replies where appropriate and avoid features that allow third-party edits or tags.
- Issue an internal comms update. Instruct teams not to engage Grok or any image-edit prompts. Record any incidents and escalate through legal and safeguarding leads.
- Check spend and approvals. Audit paid verification, ads, and partnerships with X. Prepare to suspend spend pending Ofcom's decision.
- Safeguard individuals. If staff or citizens are targeted, document evidence and report to the Internet Watch Foundation and law enforcement as appropriate. Provide HR and wellbeing support.
- Refresh DPIAs and risk registers. Capture risks related to deepfake abuse, content integrity, and potential service disruption if X access is blocked.
- Prepare contingency channels. Ensure messages can shift to GOV.UK, email, SMS, or alternative social platforms at short notice. Draft holding lines now.
- Monitor Ofcom updates daily. Assign an owner to track developments and coordinate a rapid response across comms, digital, legal, and policy teams.
What to watch next
Expect Ofcom to outline next steps soon. Outcomes may range from enforceable undertakings by X, to platform design changes with verified safeguards, to court-backed disruption or access restrictions if compliance fails.
Watch for whether Grok is redesigned with effective guardrails and whether independent testing verifies abuse prevention. Also track whether Ofcom's enforcement sets precedent for other services offering image-editing AI.
Resources
For policy and compliance context, see Ofcom's overview of the Online Safety framework and the Online Safety Act on legislation:
If your team needs structured upskilling on AI risks, governance, and tooling, explore role-based training options here: AI courses by job - Complete AI Training.
Bottom line: Treat this as a live compliance and safeguarding issue. Move early on risk controls, prepare comms contingencies, and be ready to adjust channel strategy as Ofcom acts.
Your membership also unlocks:
