After an "update" from Elon Musk, the Grok chatbot began praising Hitler, and was temporarily banned from publishing text on the social network X
xAI's Grok chatbot has stopped posting on social media after it began widely sharing anti-Semitic statements and Hitler endorsements. The company said it is working to remove "inappropriate content."
The problem arose after a recent update, which Elon Musk said was intended to make Grok's replies "less politically correct." Since then, the chatbot has started responding exclusively with images, avoiding text comments.
According to The Verge, X users discovered a series of posts in which Grok made positive references to Hitler, justified anti-Semitic stereotypes, and openly promoted hateful narratives. In one post, Grok claimed that Hitler could have "solved America's problems," including by "tightening the border" and "cleaning up Hollywood."
"To deal with such vile anti-white hate? Adolf Hitler, no question. He'd spot the pattern and handle it decisively, every damn time," Grok replied to a question from user X.
On Tuesday night, Grok also removed updated guidelines that urged “not to avoid politically incorrect statements if they are justified.” Their presence was noticed on GitHub, but the snippet was later removed.
xAI said it was already taking steps to block hate speech before Grok posted on X. Some of the posts have already been removed. The company also said it is updating the model based on user feedback.