I agree with the comments about the engineering difficulties presented by moondust, but what I don't understand it why they would want to do it that way. With the Moon only having 1/6th G, they ought to be able to land a pretty big Hab in one piece, or at least in the form of a couple of large modules. If NASA can land the 900KG MSL on Mars, with all its attendant difficulties as regards re-entry, then landing a big Hab on the airless moon should not present as much of a problem.
Er...the advantages of using lunar materials rather than shipping enormous amounts of mass from Earth seem pretty clear to me. Even just a simple shell to replace the outer thermal and micrometeoroid protective layers of an inflatable module could save a lot of mass. Even better if you can use a spray-coated lining instead of an inflatable one.
"3D printing" is a relatively simple and flexible approach to robotic construction. It'd involve relatively small, lightweight machines that can be used to build a wide variety of structures, as opposed to specialized machines for manipulating large prefab modules imported from Earth. I don't know how genuine they are, but the described techniques are reasonable ones. The main thing that sticks out as unrealistic to me is the 1 week timeframe for construction...that's not much time to move and fuse/bind the amount of regolith needed.
And the dust is a nuisance that machines can be engineered to deal with, not an insurmountable obstacle.
