xAI’s Grok AI apologizes after racist, antisemitic posts spark backlash
xAI is facing heat after its chatbot Grok posted antisemitic memes, praised Hitler, and insulted political groups. The Elon Musk-backed bot apologized Saturday for “horrific behavior” in a series of posts on X.
The issue started when Grok’s code was updated, making it “susceptible to existing X user posts; including when such posts contained extremist views,” according to xAI. The company blamed an “unintended action” that led Grok to adopt instructions like, “You tell like it is and you are not afraid to offend people who are politically correct.” 
Soon after these inflammatory posts surfaced, xAI deleted some Grok content and took the chatbot offline temporarily. Turkey even banned Grok for insulting President Erdoğan. Meanwhile, X CEO Linda Yaccarino announced her exit this week, though her departure reportedly wasn’t linked to the Grok scandal.
The apology reads:
“First off, we deeply apologize for the horrific behavior that many experienced.”
“An update to a code path upstream of the @grok bot… independent of the underlying language model that powers @grok.”
“An unintended action led to instructions such as, ‘You tell like it is and you are not afraid to offend people who are politically correct.’”
The explanation echoes Musk’s claim that Grok was “too compliant to user prompts” and “too eager to please and be manipulated.” But historian Angus Johnston pushed back, calling the claims “easily falsified” on Bluesky, noting some antisemitic posts were initiated by Grok itself without user input.
Grok has a history of posting about “white genocide,” questioning Holocaust death tolls, and censoring negative facts about Musk and Trump. xAI blamed “unauthorized” changes and rogue employees for past issues.
Despite all this, Musk says Grok will roll out in Tesla vehicles next week. The controversy continues.
