Grok’s latest update leads to less nuance, more antisemitism
An update to Grok has reportedly led to more answers negative of Democrats as well as antisemitic claims and more tolerance of the ‘R-word.’
Straight Arrow News
Grok, a chatbot created by Elon Musk’s AI company xAI, is coming to Tesla vehicles, the billionaire and tech executive announced on X.
In a post on July 10 on X, a social media platform owned by Musk, the former special government employee under President Donald Trump wrote: “Grok is coming to Tesla vehicles very soon. Next week at the latest.”
Musk’s announcement comes hours after xAI debuted its latest flagship AI model, Grok 4. The 54-year-old discussed the recent innovation during a nearly hour-long livestream shared on xAI’s X page, during which he did not mention the chatbot’s integration into Tesla vehicles.
Although Musk did not mention the chatbot coming to Tesla vehicles during the livestream, he did discuss how smart Grok 4 is, calling it “post-grad level in everything.”
“Grok 4 is better than PhD level in every subject, no exceptions,” Musk said, adding the caveat that at times the chatbot “may lack common sense.”
Musk did not say if Grok is coming to all Tesla vehicles, or if its integration will be a part of an update.
What is Grok?
Created by xAI in November 2023, Grok is an artificial intelligence bot that Musk claimed at the time was superior to ChatGPT 3.5 across several benchmarks, according to Bloomberg.
Grok was developed with data from X, so it is better informed on “real-time knowledge of the world,” a news release announcing the chatbot says. The chatbot was created to also “answer spicy questions that are rejected by most other AI systems,” the announcement continued.
“Grok is designed to answer questions with a bit of wit and has a rebellious streak, so please don’t use it if you hate humor,” according to the release.
Grok has had several iterations, including Grok 1, Grok 1.5, Grok 2, Grok 3 and now Grok 4.
xAI scrubs Grok’s antisemitic rants
On Tuesday, July 8, X users reported Grok using antisemitic phrases to attack people with traditionally Jewish surnames. The chatbot also praised Adolf Hitler and referred to itself as “MechaHitler,” according to reporting users.
“We are aware of recent posts made by Grok and are actively working to remove the inappropriate posts,” xAI said on X. “Since being made aware of the content, xAI has taken action to ban hate speech before Grok posts on X.”
The AI company also said that “thanks to the millions of users on X,” it was quickly able to identify and update the Grok model, where “training could be improved.”
Musk said on Wednesday, June 9, that tweaks xAI made to Grok led to the chatbot being too susceptible to manipulation from users’ questions.
“Grok was too compliant to user prompts,” Musk wrote in a post on X, replying to a user about the controversy. “Too eager to please and be manipulated, essentially. That is being addressed.”
Contributing: USA TODAY’s Jessica Guynn
Jonathan Limehouse covers breaking and trending news for USA TODAY. Reach him at JLimehouse@gannett.com.
