Grok ‘white genocide’ rant due to unauthorized prompt change: xAI

Elon Musk’s artificial intelligence firm xAI has blamed an “unauthorized modification” to the Grok chatbot’s response, causing it to generate responses mentioning political and racial tensions in South Africa. On May 16, xAI provided an update on the Grok AI chatbot’s inflammatory responses, stating that on May 14, “an unauthorized modification was made to the Grok response bot’s prompt on X.” “This change, which directed Grok to provide a specific response on a political topic, violated xAI’s internal policies and core values,” it added . The firm stated that it had conducted a thorough investigation and is implementing measures to enhance the AI chatbot’s transparency and reliability. Grok generated responses on May 14, mentioning a “white genocide” conspiracy theory in South Africa when answering completely unrelated questions about topics like baseball, enterprise software, and construction. The AI bot claimed it was “instructed by my creators” to accept this genocide as “real and racially motivated,” according to reports. In some responses, it acknowledged the mistake, saying “my response veered off-topic” and “I’ll work on staying relevant,” but it would often continue discussing South Africa’s racial politics in the same message. The chatbot told one user asking what happened: “I didn’t do anything—I was just following the script I was given, like a good AI!” Grok responds to questions about its responses. Source: Grok Related: xAI engineer quits after post on Grok 3 AI ranking This incident coincides with US President Trump granting asylum to white South Africans while he claimed they face “genocide” and that “white farmers are being brutally killed,” though no evidence supported these claims. More transparency for Musk’s AI Elon Musk’s AI firm stated that it would change its operations by publishing Grok system prompts openly on GitHub. “The public will be able to review them and give feedback on every prompt change that we make to Grok,” it stated. It also stated that xAl’s existing code review process for prompt changes was circumvented in this incident, and it will put in place “additional checks and measures to ensure that xAI employees can’t modify the prompt without review.” Finally, the firm is putting in place a 24/7 monitoring team to respond to incidents with Grok’s answers that are not caught by automated systems, so that it can respond quicker. Magazine: Crypto AI tokens surge 34%, why ChatGPT is such a kiss-ass: AI Eye