Erika Kirk
As Erika grieved in front of millions, the internet dissected, debated, memed, and magnified every gesture she made. The White House, Public domain, via Wikimedia Commons

Elon Musk's artificial intelligence, Grok, has made the most hilarious claim about Vice President JD Vance and Erika Kirk that many users think will be unbeatable for the rest of the year. The AI assistant is designed to be edgy and an unfiltered alternative to mainstream bots, and while Grok has been helpful, it can sometimes go rogue.

Its recent blunder is unforgettable to many. The incident began when an X user presented the AI with side-by-side images of Erika and Vance, asking if the two were 'look related.' Grok's response immediately went viral, confidently asserting that the photo of Erika Kirk was actually JD Vance 'in drag with a blonde wig' from his 'Yale days.'

Erika Kirk Is JD Vance in 'Drag with Blonde Wig'

One X user shared two photos featuring JD Vance and Charlie Kirk's widow, Erika Kirk. @theneedledrop found a resemblance between the two and captioned it 'Am I going nuts?' Canadian Jennifer reacted by tagging Grok, asking, 'Do these two look related?' Grok immediately responded, but its reaction was unbelievably hilarious.

According to Grok, the two people in the photos have 'striking facial similarities,' noting that they have the 'same eye shape, nose, and overall structure. Grok concluded that the person in both photos was Vance.

'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?' Grok said.

Another person tested Grok's logic and shared a single photo of Erika asking 'Who is this?' The AI assistant doubled down on its initial take, saying it was 'JD Vance dressed in drag with a blonde wig, from a 2012 photo taken during his time at Yale Law School.'

Grok is 'Going Rogue Again'

The internet's reaction was swift and unforgiving. Users quickly shared screenshots of the conversation, with many baffled that the bot could be 'dead a*s serious' about such an obvious error. Most X users found Grok's hallucination hilarious, and its confidence in asserting its response, despite the apparent mistake, made the situation even funnier.

One user even remarked that no other post from the AI would beat this blunder for the remainder of the year. Another joked that the technology cannot beat millions of years of human evolutionary biology quite yet. A different person said it was 'Grok Entertainment.'

Grok's History of Going Rogue

This is far from the first time the assistant has strayed from reality. Grok is built on a Large Language Model (LLM) that pulls real-time data from X, which sometimes leads to unpredictable results. For instance, in July 2025, Grok went rogue and disseminated anti-Jewish slurs and explicit depictions of violence. According to Will Stancil, an attorney in Minnesota, Grok's responses to him 'hinted at kind of anti-Semitism and bigotry.'

In December 2023, Grok roasted its creator, Elon Musk, in response to a user's request. The AI chatbot called out Musk, saying his 'greatest accomplishment' was 'turning Twitter into a dumpster fire.' While these glitches provide endless entertainment for users, they also serve as a reminder of the limitations of current AI.

Writer's pick