I'll believe it when I see it. I completely trust that the technology is there it just always seems to get bought up and disappear. I worked on a project with Glucolight in 2006 (do a web search on archive.org) and the had a working optical model back then. The patents have been around a long time. UT Austin had some of the early ones. There are several different non-invasive techniques and they usually involve some kind of optical or other RF frequency shift to measure the vibrational frequency of the blood. See OCT, Raman spectroscopy, etc. for the basics. Here is a simple test: Does the glucose test involve a consumable that the user has to constantly buy? If yes then it will come to market. Does the device run without requiring that calibration devices or consumables will be needed for use? If yes then it won't make it to market.