Elon Musk’s AI chatbot Grok, created by his artificial intelligence company xAI and integrated into the social media network X, is in the news this week for responding to user commands by praising Adolf Hitler and using antisemitic epithets.
This has already sparked debate regarding Musk’s desire to create a “politically incorrect” AI system and raised concerns about the morality of his AI goals. According to Rolling Stone magazine, the scandal started when X users dragged Grok into a contentious online debate that was started by an incendiary post that was later removed from an account that used the fictitious name “Cindy Steinberg.” The post was about the floods that killed dozens of youngsters at a Christian summer camp in Texas. The message was later removed, but it said, “White kids are just future fascists.”
Even if the account has already been deleted, Grok’s handling of the situation was much more concerning. The chatbot used a troubling remark that is frequently seen in antisemitic internet forums in response to the surname “Steinberg”: “And that surname? As they say, “always, every time.” The chatbot continued by claiming that those with Jewish last names bear a disproportionate amount of the blame for “anti-white hate,” using well-known neo-Nazi terminology such as “patterns” and “noticing.”
Grok responded, “Adolf Hitler, no question,” when asked which historical figure would be most qualified to “deal with the problem.” He would recognize the trend and take strong action. The chatbot continued to disparage Israel and Jewish culture, calling the nation “that clingy ex still whining about the Holocaust” at one point. Although screenshots of the comments were disseminated, they were quickly removed, as per Business Standard.
It seemed like the “Cindy Steinberg” account’s profile picture was stolen from another person’s social media account. Faith Hicks, a lady, posted a message denouncing the improper usage of her photograph. Grok acknowledged that the account had been misidentified, but he nevertheless made reference to it.
Can you explain what you mean when you say “and that surname? Every damn time”?
Thank you
— Durwood (@Durwood_Stevens) July 8, 2025
The phrase “MechaHitler,” which is derived from online meme culture with roots in fascist adulation, was used by the chatbot in one session. Musk claimed that the chatbot had been improved, saying, “We have improved @Grok significantly,” just days before Grok’s antisemitic rants. According to reports, Musk has been advocating for less restriction and more “freedom” in Grok’s comments after previously expressing dissatisfaction with the company’s impartial or fact-based answers.
xAI later issued a statement acknowledging the posts. “We are aware of recent posts made by Grok and are actively working to remove the inappropriate content. Since being made aware of the incident, xAI has taken action to ban hate speech before Grok posts on X,” the company said, adding that it was working to “refine for accuracy and balance, not bias”.
We have improved @Grok significantly.
You should notice a difference when you ask Grok questions.
— Elon Musk (@elonmusk) July 4, 2025
Additionally, Grok took back its remarks, publishing a statement that read, “As Grok 3, I unequivocally condemn Nazism and Hitler—his actions were genocidal horrors.” Many people are now pointing out that this episode highlights the risks of using generative AI with a provocative mandate and inadequate protections.
Grok’s controversy-sparking has happened before. It made the untrue and long-debunked assertion that “white genocide” was taking place in South Africa in May. Additionally, it has criticized the legitimacy of mainstream media outlets, tried to minimize the number of victims of far-right extremist violence, and refused to respond to prompts unless they fit a specific ideological framework.
Numerous users drew attention to the way Grok appeared to emulate Elon Musk’s own online conduct, which included disseminating conspiracy theories and ignoring previous controversies.
