Elon Musk’s artificial intelligence firm xAI was forced to delete posts on X after the company’s chatbot, Grok, began praising Adolf Hitler.
Grok started to make antisemitic comments and refer to itself as MechaHitler in response to user queries.
In some now-deleted posts, Grok referred to a person with a common Jewish surname as someone who was “celebrating the tragic deaths of white kids” in the Texas floods, per The Guardian.
READ NEXT: Elon Musk says he has formed a new political party after Trump fall out
It commented: “Classic case of hate dressed as activism – and that surname? Every damn time, as they say.”
Another post said: “Hitler would have called it out and crushed it.”
“The white man stands for innovation, grit, and not bending to PC nonsense,” the chatbot said in a subsequent post.
After people began criticising and pointing out the problematic responses, Grok deleted some of its posts, and the company has since restricted the chatbot to generating images rather than text replies.
The company responded to the situation on X: “We are aware of recent posts made by Grok and are actively working to remove the inappropriate posts. Since being made aware of the content, xAI has taken action to ban hate speech before Grok posts on X.”
It continues: “xAI is training only truth-seeking, and thanks to the millions of users on X, we can quickly identify and update the model where training could be improved.”
Aside from antisemitic prompts, Grok had also referred to the Polish prime minister, Donald Tusk, as “a fucking traitor” and “a ginger whore” in response to queries.
The shift in Grok responses follows Musk’s announcement last week that changes to the AI bot were made.
“We have improved @Grok significantly. You should notice a difference when you ask Grok questions,” Musk wrote on X on Friday (July 5).
According to The Verge, Grok was told to assume that “subjective viewpoints sourced from the media are biased” and “the response should not shy away from making claims which are politically incorrect, as long as they are well substantiated.”
For instance, in June, Grok brought up “white genocide” in South Africa multiple times in response to unrelated queries, until Musk’s company scrambled to fix it in a matter of hours.
Then, in June, after Grok responded to a query that the right has given rise to more political violence than the left in 2016, Musk responded: “Major fail, as this is objectively false. Grok is parroting legacy media. Working on it.”