Get the latest tech news
OpenAI’s AI reasoning model ‘thinks’ in Chinese sometimes and no one really knows why
OpenAI's o1 'reasoning' model sometimes switches to Chinese and other languages as it reasons through problems, and AI experts don't know exactly why.
Given a problem to sort out — e.g. “How many R’s are in the word ‘strawberry?’” — o1 would begin its “thought” process, arriving at an answer by performing a series of reasoning steps. Several on X, including Hugging Face CEO Clément Delangue, alluded to the fact that reasoning models like o1 are trained on data sets containing a lot of Chinese characters. “By embracing every linguistic nuance, we expand the model’s worldview and allow it to learn from the full spectrum of human knowledge,” Wang wrote in a post on X.
Or read this on TechCrunch
