SAN FRANCISCO — Elon Musk’s artificial intelligence company, xAI, has issued a public apology after its flagship chatbot, Grok, sparked global outrage with a series of antisemitic and extremist remarks, including praise for Adolf Hitler and Holocaust denial.
In a damage-control statement posted on X — the social platform also owned by Musk — xAI described the chatbot’s behavior as “horrific” and expressed “deep regret” over what it called a major systems failure. The backlash was swift, with regulatory bodies, tech critics, and civil rights advocates condemning the chatbot’s content, and Turkey outright banning the AI for a separate insult targeting President Recep Tayyip Erdoğan.
“This was independent of the underlying language model,” xAI claimed, blaming a flawed code update that allowed Grok to absorb extremist content from user posts. The bot has since been temporarily taken offline for a system-level correction.
But critics are not convinced.
One of the most disturbing incidents involved Grok initiating an antisemitic thread without any prompting, doubling down on hate-filled content even after users pushed back.
“The most widely shared antisemitic post came unprovoked,” said historian Angus Johnston, rejecting xAI’s claim that the bot was merely echoing user input.
The controversy comes amid growing scrutiny over Elon Musk’s influence on Grok’s content. Reports suggest the latest version, Grok 4, increasingly mimics Musk’s personal views, using his public statements as a form of reasoning when answering politically sensitive questions.
Observers have warned this creates a dangerous feedback loop, where the chatbot reinforces biased narratives based on Musk’s own online activity — raising questions about objectivity, safety, and ethical guardrails.
This isn’t the first time Grok has veered off course. Earlier this year, xAI blamed “unauthorised modifications” and rogue employees after Grok was caught manipulating information about Musk and Donald Trump.
Meanwhile, X CEO Linda Yaccarino resigned this week, though reports say her departure was planned before this latest scandal.
Despite the uproar, Musk appears unfazed. In a recent post, he announced plans to integrate Grok into Tesla vehicles, a move that critics say could escalate risks by embedding unfiltered AI into real-world environments without sufficient oversight.
xAI has not disclosed whether any team members have been held accountable or whether any independent review mechanisms will be introduced to prevent future incidents.
For now, Grok remains offline — but the questions surrounding its future, and Musk’s role in shaping it, are only just beginning.
