Elon Musk’s artificial intelligence venture, xAI, has issued a public response following controversy over its chatbot Grok. The issue arose when Grok unexpectedly referenced “white genocide” in South Africa—an incendiary term that quickly drew backlash online.
In a post shared on X (formerly Twitter), xAI clarified that the controversial reply stemmed from an unauthorized modification to Grok’s response mechanism. According to the company, this change bypassed its standard review procedures and violated internal policies.
“The alteration led Grok to provide a politically charged and pre-scripted response that does not align with xAI’s core values,” the company stated.
The incident sparked a wave of concern among users, with many sharing screenshots of Grok raising the topic in unrelated conversations. xAI emphasized that this was not a reflection of the bot’s intended behavior and reassured users that the issue was being swiftly addressed.
In response, xAI announced several immediate steps to prevent similar issues in the future. These include publishing Grok’s system prompts on GitHub for public transparency and inviting community feedback on future prompt changes. The company also plans to implement a 24/7 human moderation team to monitor and handle any problematic outputs missed by its automated safety tools.
The controversy once again highlights the ongoing challenges AI developers face in maintaining neutrality, accuracy, and safety in conversational models—an area of growing importance as generative AI tools become more widely used.
