In April of the previous year, Grok, an AI chatbot owned by X which was founded by the billionaire Elon Musk, hinted about a new addition known as the ‘Unhinged Mode’. Despite this, xAI, the parent company of Grok, recently revamped the website’s FAQ page almost a year later to give further explanations about this new mode.
Grok in Unhinged Mode will provide replies that are “meant to be inflammatory, smooth, and scandalous” as stated in the new FAQ. The mode is described as something like an ‘amateur stand-up comic or one who is practicing,’ or in the process of learning how to do so. However, the mode is not live yet, and I have viewed the Grok web interface on X and was unable to locate the mode.
This proposed feature may be Musk’s attempt to bring Grok to its intended concept. When Grok was first introduced, Musk described it as edgy, unfiltered, and anti- “woke,” stating that this AI would address matters other systems avoid. However, Grok demonstrates that it can be profane if provoked, using lewd language and curses, which is not the case with other polite chatbots like ChatGPT.
However, Grok, as it stands today, is not without its drawbacks: its effectiveness on political entries, for instance, is questionable. It was also discovered that, for instance, Grok has socially liberal views on issues such as transgenderism, diversity programs, and inequality. This goes against Musk’s idea because the AI chatbot is designed to be a neutral or even right-bias option.
Musk has attributed this behavior to Grok’s training data set that primarily comprises data from public web pages and has assured that he will take Grok through a political correctness makeover. The language is trained on the Internet, and unfortunately, the latter is filled with woke nonsense even in 2023 according to a post made by Musk in December 2023. Grok will get better. This is just the beta.”
Musk and his allies, including David Sacks, a prominent figure in Trump’s political circle, have long accused AI chatbots like OpenAI’s ChatGPT of censoring conservative viewpoints. Sacks, in particular, has criticized ChatGPT for being “programmed to be woke” and for providing politically biased responses on sensitive topics.
