Blog Post
September 20, 2025
When Your AI Starts Swearing: The Strange World of Grok’s ‘Unhinged’ Mode
Twitter/X’s AI tool Grok comes with wild personality modes like ‘Unhinged,’ where conversations turn bizarre, funny, and full of swearing. But beyond the novelty, what happens if these AI personalities are shaped by companies—not users? A look at the social and personal impact of AI modes we don’t control.
Have you tried the Grok (Twitter/X's AI tool) phone app voice mode in "Unhinged" personality? It's insane.
Brace for a bizarre and somewhat funny, no-holds-barred, crazy AI conversation where elaborate swearing is the norm.
(Below is a screenshot of its response when I asked a simple question about the EU.)
You can also have your AI 'partner' in "Doc", "Therapist", "Conspiracy", "Argumentative" modes, among others.
It did make me wonder what personal and social/societal impact it can have if/when these 'modes' are not chosen by the user but discreetly configured on the back end by the operators of call centres to gadget makers (cars, smart speakers etc.) choosing to power their device with this software.
How will this shape our world?
This was originally posted on Andras Baneth's LinkedIn account.
