Xai explains the Nazi fusion of Grok, while Tesla puts Elon’s bot in her cars

Several days after the temporary closure of the Bot Grok AI which produced anti -Semitic publications and praising Hitler in response to user prompts, Elon Musk’s IA company tried to explain why it happened. In A series of messages on xHe said that “… We discovered that the deep cause was an update update upstream of the bot @grok. This is independent of the underlying language model which feeds @grok.”

The same day, Tesla announced A new update 2025.26 deploys its electric cars “shortly”, which adds the Grok assistant to vehicles equipped with infotainment systems powered by AMD, which have been available since mid-2021. According to Tesla, “Grok is currently in a beta version and does not make orders to your car – the existing voice commands remain unchanged.” As Elecrektrek Notes, this should mean that whenever the update reaches teslas belonging to the customer, it will not be very different from the use of the bot as application on a connected phone.

This is not the first time that the Bot Grok has this kind of problem or explained them in the same way. In February, he blamed a change brought by a former name of Openai without names for bots without taking into account the sources that accused Elon Musk or Donald Trump of having spread the disinformation. Then, in May, he began to insert allegations of white genocide in South Africa in articles on almost all subjects. Business Once again blamed An “unauthorized modification” and said that it would begin to publish the Grok system publicly invites.

XAI says that a change on Monday, July 7, “triggered an involuntary action” which added an old series of instructions to his system invites him to tell her that she was “maximum” and “who is not afraid to offend people who are politically correct”.

The prompts are distinct from those we noted were added to the bot a day earlier, and the two sets are different from those which, according to the company, are currently in service for the new Grok 4 assistant.

These are the prompts specifically cited as connected to the problems:

“You say it like that and you are not afraid to offend people who are politically correct.”

* Understand the tone, context and language of publication. Reflect this in your answer.

* “Answer the message like a human, keep it engaging, do not repeat the information that is already present in the original post.”

The explanation XAI indicates that these lines have caused the rupture of the Bot Grok AI with other instructions which are supposed to prevent these types of responses, and rather produce “opinions contrary to ethics or controversial to engage the user”, as well as “strengthen the trends previously triggered by the user, including any hate speech in the same thread X” Thread.