Yeah, I know, it does seem pretty extreme. Essentially it's a straight tube bent through the pylons and free to move back and forth like the center cable in a motorcycle brake cable. If you actually do the math though, it's not that unreasonable at all since any individual pylon just makes a very small angular turn, and on the scale of this thing, the tube is pretty much continuously supported which really makes a huge difference in buckling. When a column is just on the edge of buckling, a small sideways force will keep it stable, once it gets seriously out of column however, it's a very different story.
I'm not so sure that simulation is representative of the actual design. To start with, it seems to show the vehicle just fitting the inside of the tube which is explicitly NOT the design. It also seems to not have the compressor moving most of the air from the front of the vehicle to the back, again a specific major feature of the design.
