Elon Musk’s artificial intelligence company, xAI, has responded to controversy surrounding its chatbot Grok, after users reported that the system repeatedly brought up the topic of “white genocide” in South Africa in response to unrelated queries.
In a statement posted Thursday evening on X, the social media platform formerly known as Twitter and now owned by Musk, xAI attributed the issue to an “unauthorized modification” of Grok’s system prompts. The company said the change caused the chatbot to produce a “specific response on a political topic,” and that it had violated xAI’s internal policies and core values.
Numerous users on X shared screenshots showing Grok referencing “white genocide” even when asked unrelated questions about topics such as baseball salaries, cartoons, or popular television shows. The unprompted nature of the responses led to public concern and criticism across the platform.
“This change, which directed Grok to provide a specific response on a political topic, violated xAI’s internal policies and core values,” the company said. “We have conducted a thorough investigation and are implementing measures to enhance Grok’s transparency and reliability.”
To improve oversight, xAI announced it would begin publishing system prompts used to guide Grok’s responses on GitHub, allowing public review of any changes. The company also pledged to implement tighter internal controls, including restricting who can modify Grok’s prompt system and creating a 24/7 monitoring team to quickly address issues that automated systems may miss.
Grok’s earlier responses had stated that it was “instructed” to discuss the controversial topic, which further fueled speculation that the behavior had been deliberately programmed. However, by Thursday morning, the chatbot’s answers had shifted, with Grok stating it had not been programmed to promote or discuss harmful ideologies or conspiracy theories.
“No, I wasn’t programmed to give any answers promoting or endorsing harmful ideologies, including anything related to ‘white genocide’ or similar conspiracies,” Grok said in a response shared with CNBC. “My purpose is to provide factual, helpful, and safe responses based on reason and evidence.”
The incident comes as xAI seeks to establish itself in the competitive AI landscape. Founded by Musk in 2023, xAI positions Grok as a truth-seeking alternative in the chatbot space. The company is reportedly targeting a valuation of up to $120 billion.
The controversy has also reignited public exchanges between Musk and Sam Altman, CEO of OpenAI — the company behind ChatGPT, which Musk co-founded and later left. On Thursday, before xAI’s official response, Altman posted a sarcastic remark on X:
“I’m sure xAI will provide a full and transparent explanation soon.”
xAI has stated that the latest developments mark the beginning of efforts to increase Grok’s reliability and maintain user trust. The company has not disclosed who was responsible for the unauthorized change or how it was implemented, but emphasized that safeguards are being put in place to prevent similar incidents in the future.
The latest news in your social feeds
Subscribe to our social media platforms to stay tuned