For being a glorified dad joke generator, Grok is terrible at detecting an actual joke.
Yeeting Bricks
X owner Elon Musk’s AI chatbot and lowbrow dad joke generator Grok is accusing Golden State Warriors player Klay Thompson of throwing bricks at a house in Sacramento.
It’s important to note that Thompson did no such thing, with the chatbot repeating as fact an ongoing joke that clearly flew over its head.
“In a bizarre turn of events, NBA star Klay Thompson has been accused of vandalizing multiple houses with bricks in Sacramento,” Grok wrote in an AI-powered trending tab that has been live for days, as Ars Technica reports. “Authorities are investigating the claims after several individuals reported their houses being damaged, with windows shattered by bricks.”
At the time of writing, the post is still live. Instead of admitting to the obvious mistake, X-formerly-Twitter is holding users responsible for fact-checking its flawed AI chatbot.
‘This story is a summary of posts on X and may evolve over time,” a notice below the offending post reads. “Grok can make mistakes, verify its outputs.”
Glass Houses
Needless to say, none of what Grok claimed is even remotely true. As SF Gate points out, the chatbot likely misinterpreted joke posts that are featured right below.
Fan accounts from both the Sacramento Kings and LA Lakers had sarcastically accused Thompson of “throwing bricks,” lambasting Thompson’s abysmal performance, followed by a devastating loss against the Sacramento Kings earlier this week.
“First off… I am ok,” one Lakers fan account wrote in a tongue-in-cheek tweet. “My house was vandalized by bricks. After my hands stopped shaking, I managed to call the Sheriff… They were quick to respond. My window was gone and the police asked if I knew who did it. I said yes, it was Klay Thompson.”
While it may seem like an amusing slipup, having an AI chatbot completely garble the truth — with X-formerly-Twitter selling it off as an aggregation of real news — could be a sign of dangerous things to come, especially in a volatile election year.
The incident highlights just how flawed the tech powering AI chatbots still is, despite billions of dollars being spent on research and development.
In another alarming example, just earlier this week Meta’s AI proudly announced to a group of tens of thousands of parents that it had a gifted but academically challenged child who went to school in New York City. Facebook was forced to delete the post.
More on Grok: Elon Musk’s Grok AI Accuses Him of Going to Court for Pedophilia
Share This Article