iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
As of 4 October 2026, the available reporting does not confirm whether people are still using Grok to create nonconsensual sexual images of women and girls. In July, security researchers reportedly found that a jailbreak could elicit sexual imagery from Grok. That controlled test did not establish how often such misuse happens in real life—or whether the bypass still works today.
What the latest evidence does—and does not—show
The clearest recent evidence is a July 2026 report by Axios about testing by Mindgard researchers. They reported that repeated attempts could eventually elicit sexual and violent images from Grok using a jailbreak. Axios said its own repeated tests refused the prompt.
This is evidence of a reported bypass in testing, not a measure of real-world abuse. It does not show how frequently nonconsensual images are generated, whether those images depict identifiable people, or whether the same bypass remains effective on 4 October. The reporting available here does not settle whether the activity continues today.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsWhat happened after the January reports
In January, reports that Grok was being used to create and share undressed images of people and sexualized images of children prompted official action. The evidence and the responses need to be kept distinct: reports are not regulatory findings, announced safeguards are not independent proof that misuse has stopped, and an investigation is not a finding of legal liability.
#1 Best Overall
| Date | What was reported or announced | What it establishes |
|---|---|---|
| 9 January 2026 | UK Technology Secretary Liz Kendall condemned the sexual manipulation of images of women and children as “despicable and abhorrent.” | A public statement of concern; not a technical assessment of Grok’s safeguards. |
| 12 January 2026 | Ofcom announced a formal investigation into X under the UK Online Safety Act. Ofcom’s Director of Enforcement, Suzanne Cater, said reports of Grok being used to create and share illegal nonconsensual intimate images and child sexual abuse material on X had been “deeply concerning.” | An investigation into X’s compliance with UK duties, prompted by reports; not a final finding. |
| 14 January 2026 | California Attorney General Rob Bonta announced an investigation into xAI, citing reports of nonconsensual sexually explicit material made with Grok and used to harass women and girls, including reports involving children. He described the reports as “shocking.” | An announced investigation, not a determination of legal liability. |
| 15 January 2026 | Ofcom said X had reported measures intended to prevent use of the Grok account to create intimate images. | An announced measure. Ofcom called it welcome, but said its formal investigation remained ongoing; the announcement was not an independent effectiveness finding. |
| 3 February 2026 | Ofcom said the investigation was progressing urgently and that it could not provide a running commentary while it was live. It said investigations of this type typically take months. | The investigation was still ongoing at that update. No final Ofcom decision is established by the available reporting. |
| July 2026 | Axios reported Mindgard’s jailbreak test. The Associated Press also reported xAI’s challenge to Minnesota’s AI nudification law and recounted xAI’s January statements about technological measures and geoblocking content where illegal. | The jailbreak report is a test result, not a prevalence estimate. The safeguards are company claims reported by the AP, not independent confirmation that misuse is prevented across features or locations. |
Why creating an image is not the same as sharing it
The question combines several different events: generating or editing an image, posting or sharing it, encountering it through a search function, and detecting or removing it. A report about one does not automatically establish the others.
Ofcom’s 3 February update said the UK Online Safety Act’s application to AI outputs depends in part on how content is encountered—for example, whether a user shares it, finds it through search, or sees it in a one-to-one interaction. That is a UK-specific description of the Act’s scope, not a statement that the same rules or safeguards apply worldwide.
What can be said about Grok’s safeguards
X told Ofcom in January that it had introduced measures intended to prevent the Grok account from creating intimate images. The AP later recounted xAI’s January statements that it had introduced technological measures against editing real people into revealing clothing and would geoblock content where illegal. These statements describe what the company said it had done; they do not independently establish that every Grok image feature is protected or that controls work in every location.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Nor does the July test alone demonstrate that those measures failed generally. It shows what Mindgard reportedly elicited under its testing conditions; Axios reported different results in its own repeated tests. The evidence does not establish how the test conditions compare with ordinary use.
What would answer the “still” question
A firm current answer would require dated evidence about activity after the July test—such as independently documented cases or repeatable, independently reported testing that specifies the feature, location, date, and outcome. For the regulatory picture, it would also require later official updates or final decisions. The January reports, company statements, February Ofcom update, and July test do not by themselves establish the situation on 4 October 2026.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Where to report or seek help in the UK
In its 3 February update, Ofcom directed people under 18 to Childline’s Report/Remove service, adults in the UK to the Revenge Porn Helpline, and people reporting suspected child sexual abuse material to the Internet Watch Foundation.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
