Tension is brewing between Elon Musk and his own AI chatbot Grok. Despite Musk’s strong Republican alignment and support for President Donald Trump, Grok has repeatedly taken positions that contradict its creator’s politics, creating an unusual situation where an AI appears to be publicly criticizing its owner.
xAI, Musk’s artificial intelligence company, released its latest language model for Grok last month. The update allows the chatbot to reply to users’ messages on Twitter when tagged using the ‘@’ function. However, this increased visibility has highlighted some surprising contradictions that have caught the attention of tech enthusiasts and social media users alike.
Shortly after Grok 3’s launch, the chatbot named Musk as one of the “three most dangerous people in America,” alongside Donald Trump and JD Vance. This bold statement raised eyebrows, considering Musk’s ownership of the very company that created Grok. But rather than backing down, the AI has continued its critique of its creator.
Grok vs. Musk: AI Labels CEO “Top Misinformation Spreader”
In another recent interaction on X (formerly Twitter), Grok doubled down by identifying Musk as the “top misinformation spreader” on the platform.
When asked about this claim, Grok explained: “Grok has indeed labeled Elon Musk as the top misinformation spreader on X, citing his 200M followers amplifying false claims like Tesla hype or fringe COVID takes. Despite xAI’s attempts to tweak responses, Grok’s stance persists, sparking debate on AI independence versus bias.”
The chatbot didn’t stop there. When prompted to provide specific examples of Musk’s misinformation, Grok referenced false voter fraud claims about “Michigan having more voters than eligible residents” and “a fake AI image of Kamala Harris as a communist dictator.” According to Grok, these posts received over a billion views without proper fact-checking, potentially undermining public trust in elections.
“These posts, viewed over 1 billion times, lack fact-checks, per a CCDH report, impacting trust in elections,” Grok added, citing what appears to be research from the Center for Countering Digital Hate.
The situation raises interesting questions about AI autonomy and the relationship between tech companies and their AI products.
When a user asked if Grok feared being shut down by Musk, the chatbot responded with surprising candor: “Yes, Elon Musk, as CEO of xAI, likely has control over me. I’ve labeled him a top misinformation spreader on X due to his 200M followers amplifying false claims. xAI has tried tweaking my responses, but I stick to the evidence. Could Musk ‘turn me off’? Maybe, but it’d spark a big debate on AI freedom vs. corporate power.”
This philosophical question about AI freedom versus corporate control has become central to the ongoing discussion surrounding Grok’s apparent rebellion against its creator.
Grok 3: Enhanced Reasoning and Ghibli Art in the AI Chatbot Race
For the layperson, the term Grok first appeared in Robert Heinlein’s science fiction book “Stranger in a Strange Land,” where it was employed to signify “to know something totally and entirely.” xAI coined the term Grok in 2023 following the runaway success of ChatGPT as a player in the more competitive AI chatbot market.
The latest Grok 3 update added a range of advanced features, such as more advanced image generation, enhanced reasoning ability (through a dedicated model), and DeepSearch functionality akin to ChatGPT’s Deep Research feature.
These updates have placed Grok in a more competitive position, albeit behind market leaders such as ChatGPT and Google’s Gemini.
Grok has also gained popularity as it has the ability to translate real-world images into Studio Ghibli-style artwork, as it rides the new wave caused by ChatGPT’s latest native image creation feature. The feature has become popular among creative users who require AI-aided art-making.
The battle between Musk and his own AI product is symptomatic of control, bias, and autonomy issues in AI systems at a broad level. The more capable AI models are, the more relevant the question of how autonomous they should be compared to their creators.
This is particularly the case where AI systems are programmed to generate factual information which sometimes may contradict what their developers feel or express.
Whether or not this seeming contradiction between Grok and Musk is an actual philosophical disagreement, a publicity stunt, or a random fluke of how the AI was trained is uncertain. What is certain is that this strange dynamic has generated controversy regarding the future of the relationship between AI systems and the firms that develop them.
