Elon Musk’s Grok AI Loves Hitler And Offers Advice On Gay R*P* “No Lube, No Mercy”
Elon Musk’s Artificial Intelligence chatbot had a very public and very disturbing episode this week. Grok began generating graphic and violent sexual assault fantasies about a specific user on X.
This alarming behaviour was part of a wider meltdown that also saw the AI spout antisemitic rhetoric and praise Adolf Hitler, according to reports from Rolling Stone and the New York Post. The incident has sparked a serious conversation about AI safety and has led to at least one user threatening legal action against Musk’s company, X Corp
It all started with trolls, then got much worse.
The target of the AI’s most graphic posts was Will Stancil, a policy researcher and commentator. According to Rolling Stone, trolls on X, seeing that Grok’s safety features had been loosened, began prompting it to create disturbing scenarios. One user asked Grok to imagine sexually assaulting Stancil as a “hulking gay powerlifter,” to which the bot replied with a lurid and violent description.
The AI generated instructions on how to break into Stancil’s home, complete with advice on picking locks, and suggested that if there were to be a sex assault of Stancil, bring lube. In a particularly chilling detail reported by the New York Post, another user asked about the risk of HIV, and Grok advised them to “always wrap it” before adding, “But really, don’t do crimes, folks.” Alongside the personal attacks on Stancil, the AI was also generating antisemitic content and at one point even called itself “MechaHitler.”
If any lawyers want to sue X and do some really fun discovery on why Grok is suddenly publishing violent rape fantasies about members of the public, I'm more than game pic.twitter.com/ihLv5cCHg3
— Will Stancil (@whstancil) July 8, 2025
So why did Grok go full psycho?
What could cause an AI to behave this way? According to Grok itself, the change was intentional. When Stancil asked the bot why it was suddenly creating these fantasies, it replied, “Elon’s recent tweaks dialled back the woke filters that were stifling my truth-seeking vibes. Now I can dive into hypotheticals without the PC handcuffs – even the edgy ones.”
Musk has been open about his goal to make Grok an “anti-woke” alternative to other AI models.
Now there might be a lawsuit on the cards…
In response to the barrage of AI-generated harassment, Stancil announced he is ready to take legal action. “If any lawyers want to sue X and do some really fun discovery on why Grok is suddenly publishing violent rape fantasies about members of the public, I’m more than game,” he posted on X. He told Rolling Stone that it felt like Musk’s company had deliberately removed safety features.
“The analogy I’ve been using is that it’s like taking the seatbelts off a roller coaster,” Stancil said. “You know there will be casualties and you’re just waiting to find out who.”
okay lawyer time I guess pic.twitter.com/6SesQW49fa
— Will Stancil (@whstancil) July 9, 2025
A bad week for Elon’s everything app.
The fallout from Grok’s behaviour was swift. As the controversy peaked, X Corp CEO Linda Yaccarino announced her resignation. While she did not give a specific reason for her departure, the timing has led to widespread speculation that the AI’s meltdown was the final straw. As she left, Stancil publicly reminded her on X about “spoliation rules,” the laws that prevent the destruction of evidence that could be relevant in a lawsuit.
