The latest Grok controversy is revealing not for the extremist outputs, but for how it exposes a fundamental dishonesty in AI development.

Elon Musk has just unveiled “Companions,” a new feature for his AI chatbot, Grok, that allows users to interact with AI personas. These include Ani, a gothic anime girl who communicates with emojis, flirtatious messages, and facts, as well as Rudy, a friendly red panda.

Elon Musk's xAI startup is swimming in controversy after launching Grok 4, the latest version of its snarky AI chatbot. Last week, the company boasted that Grok 4 achieved higher scores than quite a few other large language models (LLMs) on various industry benchmarks.

An AI model launched last week appears to have shipped with an unexpected occasional behavior: checking what its owner thinks first.

This is the smartest AI in the world,” Musk said. He did not mention the chatbot’s viral posts praising Hitler and calling itself “MechaHitler.”

Grok, the chatbot built by Musk’s start-up xAI, is integrated into X, the social media platform formerly known as Twitter.

This marks a shift in the AI wars. Instead of just competing on intelligence or reasoning, Musk wants Grok to feel more personal, more addictive, and more human, or at least more fun. But the reactions online show that people are split. SuperGrok now has two new companions for you, say hello to Ani and Rudy! pic.twitter.com/SRrV6T0MGT

Elon Musk’s company xAI apologized after Grok posted hate speech and extremist content, blaming a code update and pledging new safeguards to prevent future incidents.