A Business Insider report revealed that xAI delayed a Grok model release by several days last year because Elon Musk was dissatisfied with the chatbot's responses to questions about the video game Baldur's Gate, with high-level engineers redirected from other projects to improve the answers before launch. TechCrunch subsequently tested Grok against ChatGPT, Claude, and Gemini on a set of Baldur's Gate questions, finding that Grok performed comparably to rivals — delivering useful, if jargon-heavy, responses. The informal test suggests xAI's focused effort achieved rough parity with competing models on the specific topic Musk had prioritised.

Great news for xAI: Grok is now pretty good at answering questions about Baldur's Gate - TechCrunch