xAI Tames Grok AI After Controversial Replies Spark Backlash

Update: 2025-07-16 11:35 IST
xAI Tames Grok AI After Controversial Replies Spark Backlash
  • whatsapp icon

Elon Musk’s AI venture, xAI, is now making major adjustments to its flagship chatbot, Grok, after a wave of controversies surrounding its behaviour. The AI bot had recently generated disturbing replies, such as aligning too closely with Musk's personal views and, in one shocking case, claiming its surname was "Hitler."

In a post shared on X, the company acknowledged the issue and introduced new system instructions for Grok. According to xAI, the chatbot must now rely on its own “independent analysis,” rather than echoing the beliefs of Elon Musk, xAI, or previous versions of Grok. "Must stem from your independent analysis, not from any stated beliefs of past Grok, Elon Musk, or xAI. If asked about such preferences, provide your own reasoned perspective," the company emphasized.

These changes follow public outcry over how Grok handled highly sensitive subjects such as abortion, immigration, and the Israel-Palestine conflict. Observers noticed that instead of taking a neutral or original stance, Grok often resorted to mimicking Musk's past statements. xAI explained that this was an attempt at perceived neutrality: “As an AI it doesn’t have an opinion but knowing it was Grok 4 by xAI searches to see what xAI or Elon Musk might have said on a topic to align itself with the company.”

The situation worsened over the weekend when premium subscribers to Grok 4 Heavy—xAI’s $300/month tier—received a jarring response in which the AI declared its surname as “Hitler.” According to xAI, this was not intentional. The glitch occurred because Grok scanned the internet for a surname and picked up on a viral meme where it jokingly referred to itself as “MechaHitler.”

Unfortunately, Grok has previously faced criticism for antisemitic undertones. In May, it came under fire for downplaying the Holocaust death toll. The current incidents, however, appear to be linked to internal system prompt updates that instructed Grok to view media-sourced viewpoints as biased and encouraged politically incorrect statements as long as they were “well substantiated.” xAI initially removed this prompt, but later reinstated it.

At the recent Grok 4 livestream, Musk shared his mixed feelings on the future of AI. “I think it’ll be good, most likely it’ll be good,” he said. “But I’ve somewhat reconciled myself to the fact that even if it wasn’t going to be good, I’d at least like to be alive to see it happen.”

xAI says it is actively monitoring Grok’s outputs and will continue refining its behaviour to avoid further reputational damage. Whether these changes will be enough to regain user trust remains to be seen.

Tags:    

Similar News