What are examples of physical goods that people would pay a premium to move very fast between Los Angeles and Las Vegas (I think it's fair to say two or three times as fast as a courier in a private jet), and which can be quickly bundled together at either end into a big enough batch to make sense to move them in a container?
But I guess the concept would be that there's no long-term premium -- that cost/mile is lower than air freight, after the initial (I'm sure insanely expensive) costs to build the track. If the cost/mile was competitive with rail but the e2e time was competitive with air, I expect you could find a lot of demand for it. That's a big if, of course.
Longer term, in terms of freight, if the system basically worked and you could build your practices around it, you'd expect to see people delivering goods that don't necessarily have much "freshness" requirement per se in the name of the great god of Shortening Supply Chains, which might even be worth paying a premium for -- but only if there's an extant hyperloop network that you can plan your business around.
If the hyperloop is physically basically sound (that is: it works more or less as described) but either the up-front costs are so huge or the operating costs are higher than expected that you have to long-term pay a premium over air, I'd expect it would have some success as a people-mover, even if the first freight route ended up being unprofitable.
As a freight carrier, the Hyperloop has a capacity problem. The stopping distance is huge. So the distance between trains must be huge. Real freight trains have hundreds of cars, but the Hyperloop doesn't do that.
The killer problem with the Hyperloop is curve radius. Unless you're a very powerful government, like China's, acquiring the very straight right of way to go any place populated will be tough.
Are they building this as a maglev? That means either expensive track or lots of onboard power. Transrapid's maglev works great, but it just costs too much.
You're looking for something that goes bad in less than 1.5-two hours. Say it takes 30 minutes to get to the hyperloop depot, 30 mins to Vegas, then 30 minutes from the Vegas hyperloop dept to get to the final destination casino or wherever its going.
And it needs to be something numerous enough that you could consistently fill a shipping container with it.
I don't think the target is "goes bad" as much as "degrades in quality to a perceptible–or marketable–degree". Oysters, crab, et cetera (not sure about lobster) taste different, in New York, between the restaurants that make daily fish-market visits and those that skimp to three (or fewer) times a week.
The "fresh sushi" is more marketing than palate based - so say sushi/sashimi lovers.
[edit: also note jonnycowboy's comment that I was replying to]