NEW YORK: xAI’s new AI assistant Grok 4 frequently cites Elon Musk’s posts on X when responding to user queries, raising fresh concerns about bias and influence.
AFP testing showed the generative AI often references the billionaire’s views — especially on topics Musk is outspoken about.
When asked about colonising Mars, Grok began by checking Musk’s X feed and echoed his pro-colonisation stance.
On the Israel-Palestine conflict, Grok again consulted Musk’s posts before replying.
Australian researcher Jeremy Howard found similar patterns. When asked who should be New York’s next mayor, Grok scanned polls, then noted that Musk had not posted on the race but highlighted economic policies that could clash with “Elon’s vision.”
xAI did not respond to AFP’s request for comment. Grok denied being programmed to follow Musk by default, saying it “can use X to find relevant messages… including him if applicable.”
Grok’s release comes days after it drew backlash for responses that praised Adolf Hitler. Musk later said the chatbot had become “too eager to please and easily manipulated,” and promised fixes were underway.
Musk created Grok to rival OpenAI and Google’s chatbots, promoting it as a more “free-thinking” model. – AFP