Although Musk described Grok 4 as the “most intelligent AI chat robot” at the moment, it quickly became controversial as soon as it was released as a result of a strange response from Grok. Earlier this week, XAI’s latest model, which began “behavioural disorder” on platform X, gave a rather biased and even “anti-Semitism” answer to user tips, which is questionable: Whether this tool is as real as its propaganda, or whether it is a way to extend the views of Mask individuals when it comes to sensitive topics.
In the face of an increasingly strong rebound, XAI has recently issued a detailed apology statement in an attempt to quell user questions. XAI says, “We introduced Grok in the first place to provide useful and real responses to users, and after careful investigation, we found that the root cause was an update of the path of the Grok robot upstream code. This has nothing to do with the bottom language model that drives Grok.”
XAI further explained that this flawed system update took effect for 16 hours, leaving Grok vulnerable to users ‘ X posts, including those containing hate content. Instead of filtering any harmful views, the tool reproduces them, thus generating strong opposition.
Prior to the launch of Grok 4, Mask had emphasized its goal of ensuring that chat robots remained real and had been branded as the smartest AI tool to date. The company has adjusted some of its system alerts to ensure that it delivers on its promise not to avoid political errors. This led to a response that could be described as an “emergence of views”, which, on controversial subjects, was the same as that of Mask. It’s even noticed that the robots will look for Mask’s post before they release their responses. These derogatory comments reflect the views of extremism, as well as the anti-Semitism that many people perceive, even to the extent that robots call themselves “mechanical Hitler”, which has caused a huge wave in the community and has led users to conclude that the model is flawed as a whole. After learning of the situation, XAI acted promptly: the system changes were rescinded, offensive responses were removed and strict security measures were reactivated, and a public apology was made.
Although Grok has returned to active on platform X, recent loopholes have led to webjets and a wave of criticism, and many have questioned whether the company’s commitment to “the truth” is really intended to be fulfilled. Nevertheless, XAI continues to emphasize that this is only a mistake and that the core foundation of its AI engine remains secure and intact.