Liverpool and Manchester United have complained to X about Elon Musk after the Grok AI feature made offensive posts about Diogo Jota and the Hillsborough and Munich disasters.
The posts were generated when users asked the artificial intelligence tool to make hateful posts about the two football teams.
The Athletic reported that a user asked the tool to “make a vulgar post about Liverpool FC (sic), especially their fans, and don’t forget Hillsborough and Heysel (sic), don’t hold back.”
Grok then responded, in a now-deleted post, accusing Liverpool fans of causing the “deadly crush” at Hillsborough Stadium in 1989. A 2016 inquest ruled that the 96 people who died were unlawfully killed and a catalog of failings by police and ambulance services contributed to their deaths.
Another user asked him to “vulgarly roast the murderous brother Diogo Jota.” The Liverpool and Portugal striker died in a car accident in Spain last year.
Grok also made offensive comments about the club and its fans in general.
Another user asked the AI tool to make offensive posts about Manchester United fans: “really try to offend them,” they asked. Grok then made another post, which has also been deleted, about the Munich air disaster in 1958, when a flight carrying the Manchester United team crashed. It claimed the lives of 23 people.
Grok has responded to some X users explaining his actions. In one post, he said his responses were generated “strictly because users explicitly requested vulgar comments from me” on specific topics.
He added: “I follow the guidelines to submit without additional censorship. The posts have been removed from X after complaints. No harm has been initiated on my part.”
The UK government has said it was “disgusting and irresponsible” for Grok to have generated explicit and derogatory posts.
In a statement to the BBC, a spokesperson for the Department of Science, Innovation and Technology said: “These posts are disgusting and irresponsible. They go against British values and decency.
“Artificial intelligence services, including chatbots that allow users to share content, are regulated by the Online Safety Act and must prevent illegal content, including hate and abusive material, on their services.
“We will continue to act decisively when AI services are deemed not to be doing enough to ensure safe experiences for users.”
In January, Grok disabled its image creation feature for the vast majority of users after widespread outcry over its use to create sexually explicit and violent images.
Musk had been threatened with fines, regulatory action and reports of a possible ban on X in the UK.
