Its all ive been thinking about.

That may be, but there's another problem. Google's translator learns languages by comparing existing translations of texts that it finds on the web. The problem is that content farms use Google's translator to make machine-quality translations of content. Google's translator then reads its own poor translations of things and thinks that it's learning from real language samples. There's no easy way for the translate robot to tell what is machine or human translated.