This article is ludicrously short on technical details -- anyone have more?
CO2 + 4H2 → CH4 + 2H2O
NASA is investigating its use for harvesting water from exhaled CO2 on the ISS, and it has been proposed as a way to make fuel on mars. [http://en.wikipedia.org/wiki/Sabatier_reaction]
Based on the way it's presented in the video, I'm guessing he may have found a way to catalyze this reaction in reverse using methane and water as fuel, and probably also reharvesting the leftover H2 in the process. (Granted, I have no clue if any of this is possible).
This process must inevitably produce CO2, but the question is, does it produce significantly less of it than the current "generate power in one place and ship it around the country" system? If so, it must be heralded for what it is - an incremental improvement in energy efficiency; and companies will treat it as such, as it will be cheaper. But it's not an be-all, end-all solution to our problems.
Abstract: A high temperature electrochemical system, such as a solid oxide fuel cell system, generates hydrogen and optionally electricity in a fuel cell mode. At least a part of the generated hydrogen is separated and stored or provided to a hydrogen using device. A solid oxide regenerative fuel cell system stores carbon dioxide in a fuel cell mode. The system generates a methane fuel in an electrolysis mode from the stored carbon dioxide and water by using a Sabatier subsystem. Alternatively, the system generates a hydrogen fuel in an electrolysis mode from water alone.
If this thing is cheap enough it could be worthwhile, but only in cold climates where you can use the generated heat.
However, I have seen people include it in the past as a way of manipulating the numbers. I doubt this is the case with this number. It should be close to 60%