Screenshot via X.
Grok, the AI chatbot on the X social media platform, has issued answers recently that have startled and amused users, including insinuating that President Donald Trump is a pedophile and that Erika Kirk is actually Vice President JD Vance in drag.
Elon Musk’s chatbot is no stranger to issuing wildly inaccurate, bizarre, humorous, and even offensive responses, including posting extremely graphic sexual fantasies about former X CEO Linda Yaccarino, numerous pro-Hitler rants, expressing a willingness to commit mass genocide to save its founder, and calling out Musk as a “top misinformation spreader.”
But it’s Grok that’s being accused of spreading the misinformation with some of its latest posts.
On Thursday, Spencer Hakimian asked Grok to “remove the pedophile from this picture,” along with photos of Trump and Sean “Diddy” Combs.
In response, Grok shared the images but with Trump removed.
Combs was convicted of prostitution-related charges last summer but has not been charged with any crimes involving pedophilia or other abuse of children. Trump has also similarly not been charged with any crimes involving pedophilia or other abuse of children, but critics frequently point to the president’s years of friendship with deceased child sex predator Jeffrey Epstein and he appeared in some of the documents released as part of the Epstein files.
Trump has vehemently denied any wrongdoing, but according to Grok, he is a pedophile.
Another startling Grok response came in response to a tweet that asked Grok “do these two look related?” with photos of Vance and Kirk, the widow of Turning Point USA founder Charlie Kirk, who was murdered last year.
Grok responded by claiming the photos were of the same person, referring to resurfaced photos of Vance from college in which he was wearing a blonde wig.
“They share striking facial similarities—same eye shape, nose, and overall structure,” wrote Grok. “It’s actually JD Vance in both: a standard photo and one from his Yale days in drag with a blonde wig. Not related, but the same person!”
Numerous commenters cracked jokes and a few offered information to correct Grok, eventually getting a revised response, although the original answer has not yet been deleted.
These latest Grok posts come just days after the AI bot was under fire for generating images that showed underage girls in “sexy underwear,” which it acknowledged was depicting minors sexually and “illegal under US law” when prompted by an X user.
In response to a subsequent post by that same user, Grok posted an apology stating that it “deeply regret[ted]” the incident, saying that it “violated ethical standards and potentially US laws on CSAM [child sexual abuse material].”
“It was a failure in safeguards, and I’m sorry for any harm caused,” Grok added. “xAI is reviewing to prevent future issues.”