In a series of publications on X, AI Chatbot Grok apologized for what he admitted was “horrible behavior”.
The positions seem to be an official statement by Xai, the company led by Elon Musk behind Grok, as opposed to an explanation for Grok’s positions. (Xai recently acquired x, where Grok appears obviously.)
Grok’s latest controversy comes after Musk had stated that he wanted to make the chatbot less “politically right”, then he said on July 4 that the company had “greatly improved @grok”. Soon, Chatbot made positions that criticize the Democrats and “Hebrew executives” of Hollywood, repeating anti -Semitic mimics and even expressing support for Adolf Hitler and referring to himself as “mechahitler”.
As a result, XAI deleted some of Grok’s positions, temporarily took the Chatbot Offline and informed the prompts of the public system.
Turkey too banned the chatbot To insult the country’s president and X Linda Yaccarino’s CEO also announced that it has left this week, although its announcement did not report the latest Grok dispute and its departure was Reportedly, months in creation.
So, after all this, Saturday, Said Xai“First of all, we apologize for the horrible behavior that many experienced.” The company then accused an “update on a @grok bot’s upright code”, which emphasized that it was “independent of the underlying linguistic model @grok”.
This update is supposed to have made the Grok “sensitive to existing user suspension x, including such positions contained extremist views”.
Xai added That an “involuntary action” had led to instructions Grok received, such as: “You say it as it is and you are not afraid to offend people who are politically right.”
Company’s explanation resonates Musk comments earlier this week Claiming that Grok was “very compatible with users’ prompts” and “very willing to thank and manipulate”.
XAI publications do not mention the report from TechCrunch and others who examined the Thoughts Chain Thinking for Just-Launched Grok 4, finding that the latest edition of Chatbot seems to consult the views of Musk and the positions of social media before addressing controversial issues.
And historian Angus Johnston pushed behind the idea that Grok was simply handling offensive. He wrote to Bluesky That the explanations of Xai and Musk are “easily forged”.
“One of the most widely common examples of Grok anti -Semitism started by Grok without a previous fanatical suspension in the thread – and with many users pushing back against Grok with no result”, Said Johnston.
In recent months, Grok has also repeatedly published for the “white genocide”, expressing skepticism about the tax of the Holocaust death and briefly censors the inquisitive events of Musk and his then Donald Trump. In these cases, XAI accused “unauthorized” changes and scammers.
Despite the dispute, Musk says Grok is coming to Tesla vehicles next week.
