xAI’s Grok AI chatbot under scrutiny over “racist and offensive” posts

Elon Musk’s microblogging platform X (formerly Twitter) is investigating “racist and offensive” posts, which are reported to have been generated by the xAI's Grok AI chatbot. Multiple X users flagged the issue with screenshots of Grok producing offensive remarks after users' prompts.

According to Sky News (via Reuters), the safety teams have been examining user claims and whether the chatbot produced “hate-filled, racist posts” in response to user prompts. Users pointed out that the Grok AI chatbot is producing offensive posts surrounding religion and groups of football fans, making racially insensitive remarks about traffic incidents.

The issue arose nearly two months after Grok AI was under fire, generating sexualised images portraying women being digitally undressed. The UK government also threatened the platform with a ban in the country, and the latest report raises fresh concerns about AI safety standards and content moderation on generative AI systems.

According to a GBN report, the chatbot made offensive jokes about major tragedies linked to Liverpool and Manchester United, due to which both teams have demanded that the platform remove the content immediately. 

As of now, xAI has not provided any statement on the user claims, and Reuters reports suggest that the case is under investigation within the company. The episode now adds to a growing list of controversies surrounding Grok AI, but it also raises a concern over how these models are being trained. 

Well, it's not the first time AI has made racist claims. In mid-February 2024, Google’s image generation tool was also scrutinised for generating historically inaccurate diversity, like Black Vikings, Asian Nazi soldiers, or diverse U.S. Founding Fathers. Sundar Pichai also addressed the issue as "unacceptable,” and soon the feature was entirely removed on February 22 while committing to fixes.