This technology has been implemented by using Optical Character Recognition (OCR) technology, which is able to extract text from images. This OCR technology able to read typewritten or printed text, handwritten or whatever document that can be scanned into digital form.
In fact, this OCR technology has been existed for a while. But Google didn’t implement it because of the accuracy problem - as Google target to improve the accuracy as much as possible before implement it to achieve the highest quality.
You can view the example of these documents here: Steady success in a volatile world