Grok Calls Out Elon Musk For 'Exaggerated' Comments About xAI's Focus On Truth

Department of Government Efficiency (DOGE) chief Elon Musk has just been called out by the artificial intelligence assistant created by his AI company, xAI, for supposedly "exaggerating" his claims about xAI being focused on work that produces the truth, regardless of political correctness.

Musk re-shared a post by xAI engineer Igor Babuschkin, which called on backend engineers to work on Grok, xAI's chatbot that has been gaining much attention from consumers worldwide in recent weeks.

"xAI is the only major AI company with an absolute focus on truth, whether politically correct or not," the tech billionaire wrote Monday.

Musk is Embellishing His Claims Around xAI: Grok

Many X users flooded Musk's posts with positive comments, saying Grok is indeed one of the best, if not the best, AI assistant available today.

"Grok is legit. No other AI I've used hits facts like Grok," said one user, while research scientist Houman David Hemmati said the AI assistant "has done things ChatGPT refuses to even consider."

However, one user asked Grok about the veracity of the X owner's statements about xAI.

"Elon's comments on xAI are partly true but exaggerated...however, claiming it's the only major AI firm prioritizing truth is debatable," Grok responded, noting how Anthropic and OpenAI "also value accuracy."

Another user asked a similar question on the "accuracy" of Musk's claims about his AI startup. Grok noted that Musk's use of the word "only" is debated.

In another response to a user's question on whether Musk's statements were "right," Grok said it can be true that XAI "likely uniquely prioritizes truth in AI," but it did reiterate that Musk's claim "oversimplifies the issue" on the key factors that make a good and safe AI product.

X Users Make Suggestions to Improve Grok

While Grok itself and other X users doubt the correctness of Musk's claims around xAI, some were focused on providing the xAI team with advice on how they can improve their AI chatbot.

"Grok needs a sense of subjective self if you want it to be able to consider if what it is saying is true. Right now it gets confused a lot and is easy to trick," said one user.

Another user pointed out that "safe" AI systems should exhibit "raw honesty" and also raised the matter on whether AI should always "prioritize truth over sensitivity."

The latest version of Grok was released last month, with experts noting how the chatbot has better reasoning skills than other AI assistants.