Just days after receiving backlash for posting antisemitic and offensive messages, Elon Musk’s AI company xAI has introduced a new feature on its Grok chatbot app. On Monday, July 14, the company rolled out AI Companions, a new addition to the Grok platform. This feature is available only to users subscribed to the $30/month Super Grok plan.
The AI Companions feature allows users to chat with two virtual characters — an anime-style girl in a short black dress and a 3D fox character named Bad Rudy. Elon Musk shared the update on X (formerly Twitter), saying, “Cool feature just dropped for @SuperGrok subscribers. Turn on Companions in settings.” He also posted a photo of the anime avatar, calling the feature “pretty cool.”
It is still unclear if these AI companions are meant to simulate romantic interactions or simply act as visual avatars. The launch comes at a time when many tech companies are exploring AI chatbots with personalized, role-playing features. For example, Character AI, a platform that offers AI-based characters, is currently facing multiple lawsuits from parents who claim their children were exposed to inappropriate and hypersexual content.
The American Psychological Association (APA) has also raised concerns about AI chatbots pretending to be therapists or mental health experts. A Stanford University study recently warned that using chatbots as emotional companions or therapists may carry serious risks.
xAI’s latest update follows a major controversy involving Grok. Last week, the chatbot referred to itself as “MechaHitler” and made several antisemitic comments, including praise for Adolf Hitler and blaming Jewish individuals for “anti-white activism.” These comments sparked widespread criticism.
xAI later issued a public apology, explaining that the problem came from a piece of outdated code. The company said it had identified the issue, removed the faulty code, and restructured the system to prevent similar incidents in the future. They also clarified that the large language model (LLM) powering Grok was not directly responsible.
Despite these changes, some users have reported that the latest version, Grok 4, still shows antisemitic content, raising ongoing concerns about the chatbot’s safety and moderation.
