Billionaire Elon Musk-led social media platform X is now investigating “racist and offensive” posts made by xAI’s artificial intelligence (AI) chatbot Grok, Sky News has reported.
Grok, an AI chatbot, allegedly generates racist and hateful posts online. Several users on X have flagged this trend and pointed out what appears to be abuse of the chatbot. The posts were generated after users called out Grok for “vulgar” remarks.
The report suggests that X users are increasingly challenging Grok to create vulgar responses, particularly targeting Hinduism, Islam and football fan groups. A Sky News analysis found that some AI-generated responses contained highly offensive language and profanity about Hinduism and Islam, and denigrated the religion with racist remarks.
The British government criticizes Grok’s offensive posts
The British government on Sunday (local time) called the posts “disgusting and irresponsible” and said they were against British values.
The posts appear to be part of a growing online trend in recent days where X users are asking the platform to generate “vulgar” and unfiltered comments.
A spokesman for the Department of Science, Innovation and Technology said artificial intelligence services, including chatbots that allow users to share content, fall under the Online Safety Act and are required to prevent illegal material, including hateful and offensive content. They added that authorities will continue to take decisive action “if AI services are deemed not to be doing enough to ensure a secure user experience”.
The development comes nearly two months after the Musk-led platform faced the threat of a ban from the UK government after sexualised AI-generated images of women undressing emerged.
Grok blames Liverpool fans for Hillsborough disaster
The Sky News report also said the AI tool falsely blamed Liverpool FC fans for the Hillsborough disaster in which 97 fans died and used derogatory terms about the city.
Although the police initially blamed Liverpool supporters for causing the tragedy, this claim was later discredited after decades of campaigning by the victims’ families.
Following the posts, Liverpool officials are now seeking to have the content removed.
X deletes offensive posts
Although X removed the posts flagged by Sky News, there have been no signs of changes to the platform’s policies aimed at preventing online harm when Grok is called out for “vulgar” responses.
This comes a day after Musk wrote on X: “Only Grok speaks the truth. Only true AI is safe.”
The crackdown on Grok continues
According to Reuters, governments and regulators are cracking down on sexually explicit content generated by Grok. The investigations, bans and security challenges reflect growing global efforts to curb illegal material online.
In January of this year, xAI announced that it had restricted Grok’s image-editing features for certain users and blocked others based on their location from creating images of people in revealing clothing in regions where such content is prohibited. The company did not disclose which countries were affected.
