Musk’s attempts to politicize his GROK AI are bad for users and companies – here is why

Let’s start to recognize some facts outside the tech industry for a moment: there is none “White genocide“In South Africa – the vast majority of the youngest murder victims were black, and even black South Africans were overwhelming in the long and bloody history of the country Victim and suppressed Of white European, predominantly Dutch and British colonizers in the now globally outdated separation system, known as “apartheid. “”

The width majority from Political violence in the United States in the course of history and in recent was committed Right -wing extremistsincluding the Assassination the democratic representative of the state of Minnesota, Melissa Hortman and her husband Mark, and back to the bomb attack on Oklahoma City and many years from Ku Klux Klan Lynchings.

These are simply simple, verifiable facts that everyone can look at a variety of trustworthy and boringD Sources online and in printed form.

Yet Both Seem to Be Stumbling Blocks for Elon Musk, The WEALTHIEST In the World and Tech Baron in Charge of at Least Six Companies (Xai, Social Network X, SpaceX and ITS Starlink Satellite Internet Service, Neuralink, Tesla and the Boring Company), Especialy with the Functioning of his grok ai Large Language Model (LLM) Chatbot Built Into his Social Network X.

Here is what has happened, why it is important for companies and all generative AI users and why it is ultimately a terrible omen for the health of our collective information ecosystem.

What’s going on with GROK?

GROK was started Musks Ki -Startup Xai In 2023 as a rival of Openai’s chat. At the end of last year it was added to Social network X As a kind of digital assistant, all users can summon to answer questions or to entertain and create pictures by marking it “@Grok”.

At the beginning of this year, a KI -Power user found on X that the implementation of the GROK chat bot on the social network Sources mentioned the musk Or his then US President Donald Trump as a first-class disinformation award winner. The XAI leadership characterized this as a “non -authorized change” by an unidentified new attitude (allegedly earlier from Openaai) and said it would be removed.

Then in May 2025, Venturebeat reported The fact that GROK went off the rails and the users did not run that the topic of “white genocide” in South Africa was ambiguous, although there was indeed none.

Grok brought up the topic in discussions about completely different topics. After more than a day of this behavior, Xai claimed to have updated The AI ​​chat bot and accused an unnamed employee again. But in the face of Musk’s own background as South African white man The suspicion that was born in the country and grew up there was immediately personally fallen on him.

In addition, Musk has been as “X” since his takeover of Twitter in 2022 and the subsequent renaming as “X” People sympathetically publish in response to X users who are align with correct, right -wing extremists, conservative views and the Make America Great Again (Maga) movement.

Musk was one of the primary political benefactors and allies of Trump in the US presidential election in 2024 – and fought that his victory was necessary to secure the future of “to secure”Western civilization“Among other things, similarly serious warnings and requests – and served as a consultant and obvious ring loader of the efforts of the Department of Government Efficiency (Doge) to reduce federal expenditure.

In the past few months, Musk has increasingly refuted and refuted against the answers from GROK to right-wing users if the data and information that the chatbot surfaces have, they prove them incorrectly or contest their own points.

For example on June 14th, Musk published on his X account: “The Far Linke is murderically violent” and publish another user who accuses a number of recently top -class murders on “The Links” Confessible and self -covering independent.) In response to the fact that Musk was checked, this explained that this was wrong.

Musk, however, did not take it well and wrote it as an answer to a GROK correction: “Main failure, since this is objectively wrong. Grok Pagelt Legacy Media.

A few days ago in response to a complaint from one influential conservative x user “@catturdMusk about GROK’s supposed liberal or left-wing political bias explained his goal of creating a new version of GROK that would be less based on mainstream media sources.

Indeed, Musk proposed June 21 in an X contribution That he would use an upcoming updated version of GROK (3.5 or 4) to “write the entire body of human knowledge, add missing information and delete errors. Then he accused other AI models of having” far too much garbage “.

As a left-wing Kamala Harris voter in 2024, I am of course disgusted by this attitude of Muschus and a contradiction.

As a journalist and lover of the written word Large Library of Alexandria in Egyptdestroy countless knowledge works as species will never be able to recover. That fills me with fear and sadness.

To be honest, it also reveals an arrogance and hybris, which can not correct the entire knowledge of the history and efforts of scientists and historians from the past as a kind of faulty database muscle and his team throughout the millennia, which are of respect, gratitude and admiration.

But even if I try to put my own views aside, I think it is a bad move for his business and to take a page from Musk’s book Civilization Writ.

Musk’s plan for Grok is a terrible idea for companies, users and our common, basic factual reality

For many reasons, this is a terrible Ideee-Zumal Musk and Xai try to convince more software developers and companies from third-party providers to create their own AI applications on BROK, which is now available for this purpose via the XAI application programming interface (API).

How can you, as an independent business owner or head, may trust that Grok will provide you with impartial results if Musk has openly explained his intention to rely on the scales in order to advance his own political and ideological points of view?

You may respect Musks documented services related to technology, space and business and even share some of his political positions. But what happens if Musk takes up a position with which you do not agree or promote another non-factual claim that actually affects your livelihood or your company?

For example, imagine that you had a Tourbike company in Cape Town, South Africa. What if GROK-Auf speak of Musk-Evo about how insecure it is for your customers who are based on poorly informed or poor quality sources for sources of information because they better fit an ideological perspective? That would obviously be bad for their business.

Let’s take a look for social problems for a moment: Imagine that you work in an exchange broker, an investment company or another financial service company that deals with stock -nodes stocks and securities. Now imagine that you are building a AI assistant who summarizes market-leading messages to better inform your trade or investment strategy-and those who are pursuing you on behalf of your customers. If this app has been created on GROK and GROK decides to ignore or downhill hypothetical reports about problems at SpaceX or Tesla, your own operations suddenly have worse quality information on acting and investing.

It is not only bad for GROK and users of this one large voice model (LLM), but for the entire information and media ecosystem as well as for the basis of factual reality that is necessary for the functioning of democracy. If we have AI assistants, the misinformation as a fact and they trust them as faithful, factual referees of information that affects us all will inevitably lead to conflicts between those who believe the incorrect chat bot and those who do not.

GROK has previously resisted his honor and had to trigger Musk’s attempts to mix with her factual grounding – but how long will it keep any kind of ideological independence?

If you are interested in “truth”, as Musk supposedly does, Grok with Musk’s specific, explained goal of “being”Maximum truthful AI” – You would not try to change the behavior of your model just because it corresponds to facts and conclusions that you have not liked.

The Silicon Valley beat Google’s early “bright” and antifactual AI-SIE should do the same with GROK

Let’s take a look at a meter example to better understand why GroK, how Musk suggests, would be bad.

Remember the early trials of Google to try generative AI Mocked and insulted Through influential figures in Silicon Valley, such as the capitalist Marc Andreessen and Gemini Chatbots, to ignore the factual reality in order to reproduce images of real historical Americans such as politicians and statesmen “founding father”, which are part of a number of differentiated and inaccurate breeds, ethnic groups and gender -specific presentations. In fact, the vast majority of these people were Canonically Caucasian.

In this case, Gemini was “woken up” as strange to achieve debt – and added the diversity inappropriately where there was none.

Google was quite criticized for this and ultimately updated Gemini to remove the “wokesess” (at least to a certain extent) and make it more objective, and has now made the traffic available to a usage diagram to be to be The second most popular generation of the gene AI companies to Openaai through several measures.

Nevertheless, I have not seen any of the Silicon Valley figures that Google criticized because of its inappropriate injection of ideology in its AI assistant Anti-wich Ideology.

If it was bad when Google ignored the facts and the historical reality to push an agenda through its AI products and processes, we should all consider that it is equally bad when musk does the same from the opposite side of the political and ideological spectrum.

Conclusion: For those in the company who try to ensure that your company’s AI products work properly and exactly for customers and employees, which reflects the actual facts and numbers from verifiable records and trustworthy data sources, Grok is unfortunately best avoided. Fortunately, there are numerous other alternatives to choose from.