Elon Musk‘s AI chatbot Grok has stirred uproar after it reportedly told users that its owner is “smarter and physically fitter” than anyone else in the world.
Several of Grok’s remarks on Musk, such as how Musk could hypothetically beat Mike Tyson in a boxing match, surfaced on X. The chatbot reportedly also ranked him above Leonardo da Vinci, and said he was funnier than Jerry Seinfeld.
Musk addresses the issue
Musk called the hyped-up answers a consequence of “manipulation by adversarial prompting”. He added that this had led to Grok saying “absurdly positive things” about him.
“For the record, I am a fat retard,” Musk said.
Earlier today, Grok was unfortunately manipulated by adversarial prompting into saying absurdly positive things about me.
— Elon Musk (@elonmusk) November 20, 2025
For the record, I am a fat retard 😀
After Musk’s statement, Grok too dialled down on its praise and began giving restrained answers about him, placing him in its top 10 humans rather than ranking him first, the Washington Post reported.
Grok’s objectivity under scanner
The remarks renewed concerns over Grok’s objectivity and growing scrutiny of Musk launching the latest version of its chatbot, Grok 4.
Users have also noted that Grok, which Musk has pitched as an ‘anti-woke truth seeking alternative’ frequently echoes his public political views, prompting concerns about bias.
In the past, Grok has referenced “white genocide,” questioned Holocaust death tolls, and reportedly made derogatory remarks about politicians, which xAI, the Musk-founded start-up that develops Grok, has attributed to “unauthorised modifications” or rogue employees, Russia Today reported.
A number of AI researchers pointed out that Grok is known for being intentionally humorous and sarcastic, unlike traditional models that adopt a more neutral point of view, according to Moneycontrol.
Grok’s remarks were likely generated to keep its brand of personality-driven engagement rather than serve as factual comparisons, Moneycontrol reported, quoting experts.
