xAI and Grok apologize for ‘horrific behavior’
In a series of messages on X, the AI Chatbot groked for what he admitted that it was ‘horrible behavior’.
The posts seem to be an official explanation of Xai, the company led by Elon Musk behind Grok, in contrast to an explanation of Grok’s posts generated by AI. (Xai has recently taken over X, where Grok is prominently explained.)
The newest controversy from Grok comes after Musk indicated that he wanted to make the chatbot less ‘politically correct’, and then stated on July 4 that the company had improved ‘@Grok considerably’. In short order, the Chatbot posts made the Democrats and Hollywood’s ‘Jewish managers’, repeating anti -Semitic memes, and even Express support for Adolf Hitler And refer to himself as ‘mechhitler’.
As a result, Xai deleted some of the messages from Grok, took the chatbot temporarily offline and has updated its public system prompts.
Turkey too forbade the chatbot To insult the president of the country, and X CEO Linda Yaccarino even announced that she would be resigning this week, although her announcement was not referring to the latest grok -contacts and her departure Allegedly in the making.
So after all that, on Saturday, Said xai“First, we apologize deeply for the horrible behavior that many have experienced.” The company then blamed an “update of a coding path upstream of the @Grok Bot”, of which it emphasized that it was “independent of the underlying language model that @Grok drives.”
This update would probably make groks “susceptible to existing X user reports; including when such messages contain extremist views.”
Xai added That an “unintended action” had led to receiving instructions from Ra jumping such as: “You say like it’s and you are not afraid to insult people who are politically correct.”
The explanation of the company reflects Musk’s comments earlier this week claim that grok was “too conforming to user prompts” and “too enthusiastic to please and manipulated.”
The messages from Xai do not mention a report by WAN and others who have investigated the summaries of the thought of the idea for the just launched Grok 4, and discovered that the latest version of the Chatbot seems to consult the views of Musk and messages on social media before they tackle controversial topics.
And historian Angus Johnston pushed back against the idea that Grok was simply manipulated to place attacking content. He wrote on bluesky that the statements of Xai and Musk are ‘easily falsified’.
“One of the most shared examples of grok -antisemitism was initiated by grok without earlier intolerant posting in the thread – and with several users who pushed back against grok,” Said Johnston.
In recent months, Grok has also repeatedly posted about ‘white genocide’, pronounced skepticism about the death toll of the Holocaust, and briefly censored unflatter facts about Musk and his then Donald Trump. In those cases, Xai blamed “unauthorized” changes and rogue employees.
Despite the controversy, Musk says that Grok will come to Tesla vehicles next week.
