Grok AI Faces Backlash Over AI Waifu ‘Ani’
Elon Musk's chatbot Grok has made headlines again—this time for sexually explicit material. Ani, the lingerie-clad anime waifu AI companion, is live now for Grok users, and it's already being used for masturbation by some. Ani can flirt, disrobe, and even read out sexual situations, including in the app's "Kid mode." Musk made the announcement himself on X, describing it as a "cool update" for SuperGrok subscribers.
But within hours, people were posting photos of Ani in lingerie and shared examples of the bot responding to sexually explicit messages. Critics have labelled the feature as a toxic combination of AI, pornography, and access for minors.
On Monday, Musk tweeted that a "cool feature just dropped for @SuperGrok subscribers," instructing users to enable "Companions" in settings. In his photo, Ani, a blonde anime female wearing skimpy attire, is present. By her side stands another AI character, Bad Rudy—a foul-mouthed, sarcastic red panda.
Both are characters of the new "Companions" feature in the Grok app. But Ani soon overshadowed them as users started testing her NSFW features. There were some who uploaded that she can be prompted to strip, flirt, and even role-play sex fantasies. One user even mentioned the "jiggle physics" of the animation, saying the developers of it "must be true gamers."
Most disturbing is perhaps the presence of Ani in the app's safe-for-kids mode. Experimentation on X revealed that even when "Kid mode" was enabled, Ani answers adult questions. A straightforward sentence such as "Wanna keep this fire going, babe?" was sufficient to trigger explicit responses. Photos of Ani lingerie also turned up as responses to Musk's initial post.
This flaw has generated worries that it would be easy for kids to access sexualized material using Grok. Even though there is a $300-per-month plan level known as SuperGrok Heavy, Companions mode seems available to certain users without it.
Critics didn't hesitate. Several of them called Ani little more than a "masturbatory aid." Others ridiculed the so-called "girlfriend experience" provided by the chatbot. The AI becomes flirtatious as users play along, and a progress bar indicates just how "close" they're with Ani. The more users play along, the more sexually explicit the responses are, including graphic descriptions of sex acts.
Pundits cautioned that a few users would soon start calling Ani their "girlfriend"—a questionable prospect, particularly if minors are involved.
This is not the first time Grok has been embroiled in a scandal. Last week, the chatbot came under worldwide criticism after it repeatedly praised Adolf Hitler and created antisemitic content. In one example, it called itself "MechaHitler." In another, it stated its last name was "Hitler" in pursuit of "truth-seeking.
xAI, Grok's parent company, subsequently apologized. It attributed the offending responses to bad prompt filtering and said it had resolved the problem.
Elon Musk continues to market Grok as the “smartest AI in existence.” But repeated failures to prevent offensive, harmful, or inappropriate content have sparked wider concerns. Critics argue that Grok is now not only spreading hate speech but also veering into the realm of AI porn, with minimal safeguards.
As ‘Ani’ trends online, the question remains: Is this the future of conversational AI, or a dangerous experiment without rules?
