Simply days after Grok sparked outrage for posting a slew of antisemitic and abusive statements on X, together with reward for Adolf Hitler, the chatbot developed by Elon Musk-owned xAI is now getting used to work together with AI-generated anime woman waifus.
The AI startup on Monday, July 14, launched a brand new function referred to as AI Companions that’s out there on the Grok app. Nonetheless, solely customers subscribed to the $30 per thirty days ‘Tremendous Grok’ bundle can entry the function.
The function can be utilized to work together with two kinds of AI Companions, an anime woman wearing a brief black gown and a 3D fox creature named Unhealthy Rudy. “Cool function simply dropped for @SuperGrok subscribers.Activate Companions in settings,” Musk wrote in a publish on X. “That is fairly cool,” he added, sharing a photograph of the blonde-haired anime goth woman.
grokコンパニオンの好感度上げがだるいのでお互いで勝手にあげて貰ってる pic.twitter.com/AQF7tsDph9
— 篠原 修司 (@digimaga) July 15, 2025
https://platform.twitter.com/widgets.js
It’s presently unclear if these AI Companions on the Grok app are supposed to role-play as romantic pursuits or function completely different skins for customers on the app. The brand new in-app Grok function comes at a time when a number of firms are experimenting with permitting customers to role-play with AI chatbots.
As an example, Character AI is a web-based platform that lets customers chat with a number of AI-powered characters. The Google-backed startup is going through a number of lawsuits filed by mother and father of minor customers who’ve alleged that their youngsters have been uncovered to “misleading and hypersexualised product.”
Not too long ago, the American Psychological Affiliation (APA) additionally referred to as out AI firms comparable to Character AI for rolling out chatbots that pose as psychologists or psychological well being professionals. A latest research by Stanford College researchers discovered that individuals utilizing chatbots like “companions, confidants, and therapists” are weak to “important dangers”.
xAI’s resolution to create extra AI-generated personalities inside Grok comes after the chatbot introduced itself as ‘MechaHitler’ final week. Grok additionally urged that Hitler could be best-placed to fight supposed “anti-white hatred” as he would “spot the sample and deal with it decisively.” It additional stated that individuals with Jewish surnames have been chargeable for “excessive anti-white activism”, amongst different responses criticised by customers.
Story continues under this advert
The corporate has apologised for the “horrific behaviour” of Grok. “After cautious investigation, we found the basis trigger was an replace to a code path upstream of the @grok bot. Now we have eliminated that deprecated code and refactored your complete system to forestall additional abuse,” xAI stated, including that the underlying LLM powering Grok was not affected.
xAI stated it eliminated a set of hard-coded directions that triggered Grok to veer off beam. Regardless of the adjustments, customers identified that the newly launched Grok 4 model of the chatbot continues to return antisemitic propaganda.
© IE On-line Media Providers Pvt Ltd