A disastrous idea
Even if he could actually build it at all (there's a lot of doubt about that), the slightest problem would destroy most of the tube, the vehicle and the people on board. A sudden tube breach will have some seriously nasty effects because of the release of a lot of energy as the vacuum is lost.
Could he build it so that it would work for a while, and do so at a cost that makes the whole thing seem feasible (if you pretend none of the failure modes will ever happen)? Probably not. Pump-down is going to be very slow and they don't yet have a feasible design for letting people board the capsule without first re-pressurizing the whole system then pumping it down again (which is not just slow but requires a lot of energy).
Then there are the failure modes. Thermal expansion is going to cause it to buckle, with disastrous results. Differential thermal expansion (top is warmer than the bottom) is going to cause it to buckle, with disastrous results. Using invar rather than steel might avoid the thermal expansion problems but invar is quite a bit more expensive. Failure of maglev on a capsule going at the speed of a bullet, even if it drops onto wheels, is going to cause the capsule to graze the tube, with disastrous results.
If the capsule develops a leak, that won't destroy the whole system (At last, a failure mode that doesn't destroy the whole system) but the occupants will not disembark in good condition, or under their own power. But they'll probably be freeze-dried, so they'll be lighter to lift out and cost less to cremate (no water, so they'll burn more easily).
Oh, and... A car hitting one of the supports (most of his design versions are on pillars, not underground) will have disastrous results. A bullet through the pipe will have disastrous results - no need for a bomb, just somebody with a handgun.
For more detailed analyses, see any of these Thunderf00t videos. Note: later videos re-use some footage from the earlier ones so that each can stand alone, so if you watch the whole lot you'll see the same points several times. From oldest to newest:
How the Hyperloop can kill you!
The Hyperloop: BUSTED!
Elon Musks Hyperloop: BUSTED!
Hyperloop crashes and BURNS!!!
Entire Hyperloop could be destroyed in SECONDS!
Hyperloop, HyperSPEED, HYPERMADNESS!
To be fair, there have been several attempts to debunk the Thunderf00t analyses. Most prominently by Shane Killian. However, I long ago found that when Killian talks about anything outside his areas of expertise (whatever they might be, if in fact there are any), he's wrong. You'll have to judge for yourselves.
Then again, everybody thought Musk was crazy for trying to land the first stage of the Space-X vehicles so he could re-use them. So maybe he's right on this one. However, with Hyperloop, Musk has made the concept public domain for anyone to build. He's not risking any of his own money or any significant amount of his time on it.
I think this concept was based on a "back of a fag packet" calculation, then announced, then Musk did more detailed calculations and realized how unworkable it was, so made it PD. Your mileage may vary. Particularly if you ride on an imploding Hyperloop.