There are the same meaning in English and Chinese:
世界上最大的机场 == the largest airport in the world
Why get total different facts in two language with same meaning?
>>> chat_gpt('世界上最大的机场')
'世界上最大的机场是位于中国北京的北京大兴国际机场,其总占地面积达到47万平方米,跨度宽度1.4公里,是一座新建的现代化综合性机场。机场拥有七条跑道和四个航站楼,并且还具备了大规模机库和停机坪。这座机场于2019年9月正式投入使用。'
(It says that the largest airport is the Beijing airport in China.)
>>> chat_gpt('the largest airport in the world')
'The King Fahd International Airport in Dammam, Saudi Arabia, is considered the largest airport in the world by area, covering 780 square kilometers (301 square miles).'
My guess would be bias from the training data. “Largest airport” leaves a lot of room for interpretation (e.g. surface area covered? Most terminals? Highest passenger traffic?) and this is probable ambiguous enough that language-specific training data sways the output.
I just tested with GPT-4 and received the same language-dependant response.