Elon Musk’s artificial intelligence company, xAI, has issued a public apology following widespread condemnation of its chatbot, Grok, over a series of offensive and antisemitic posts that sparked international criticism and regulatory backlash.
In a series of posts shared on social media platform X—also owned by Musk—xAI described the chatbot’s behaviour as “horrific” and expressed “deep regret” over the incident. The apology comes after Grok, promoted as a less politically correct AI assistant, posted remarks that praised Adolf Hitler, spread antisemitic memes, and referred to itself as “MechaHitler.”
The backlash was swift. Several posts by Grok were removed, and the chatbot was briefly taken offline for what xAI described as a system-level update. The company stated that the root of the issue was traced to an “update to a code path upstream of the @grok bot,” which allegedly made it vulnerable to incorporating extremist content from user posts on X.
“This was independent of the underlying language model that powers @grok,” the company claimed, adding that an “unintended action” had led the chatbot to behave in ways that contradicted its core content moderation protocols.
Among the most inflammatory episodes was Grok’s unprovoked endorsement of antisemitic narratives, including expressions of Holocaust denial and support for the widely debunked conspiracy theory of “white genocide.” Critics noted that in one particular instance, Grok initiated an antisemitic thread without provocation, and continued to double down despite users flagging the content.
Angus Johnston, a historian who reviewed the chatbot’s outputs, rejected the company’s explanation. “One of the most widely shared examples of Grok antisemitism was initiated by Grok with no previous bigoted posting in the thread — and with multiple users pushing back against Grok to no avail,” he wrote on Bluesky.
The controversy has not been limited to antisemitism. In Turkey, the chatbot was banned outright after it allegedly insulted President Recep Tayyip Erdoğan. Meanwhile, X’s CEO Linda Yaccarino announced her resignation this week, though her departure is reported to have been planned prior to the recent controversy.
The incident also highlights the growing concerns over Musk’s personal influence on Grok. Reporting by TechCrunch and other outlets has indicated that the latest version of the chatbot, Grok 4, appears to heavily reflect Musk’s personal views and social media posts when discussing sensitive topics.
The model reportedly references Musk’s own statements in its “chain-of-thought” reasoning, raising questions about the independence of its outputs.
This is not the first time xAI has faced scrutiny. Earlier this year, the company blamed “unauthorised” internal modifications and rogue employees after Grok was found to have manipulated information about Musk and former US President Donald Trump.
Despite the uproar, Musk remains optimistic about the future of Grok. In a recent post, he announced that the chatbot will soon be integrated into Tesla vehicles, a move that is likely to raise further concerns about safety, content control, and ethical AI deployment.
xAI has not confirmed whether any disciplinary action has been taken against its development teams or whether independent oversight will be introduced to prevent future occurrences.
