Bingo.
Elon is capitalizing on the trend away from car ownership, not so much on the "love" of public transportation.
A super-cheap autonomous bus like Elon describes would be absolutely loved by urban millennials. It'd likely be a tenth the cost of Uber and even less than a private autonomous taxi.
How I imagine what Elon is imagining is a nearly full point-to-point transportation system. While it may not be as fast as a personal autonomous vehicle,
Possibly faster.
it'd still be much faster than current public transportation,
Much slower than existing metro rail systems, and slower than existing light rail systems.
won't create more traffic/congestion compared to an urban SUV,
A bit less than taxis. But the congestion level is almost entirely determined by the drivers of private cars. In the busiest, most popular locations, induced demand causes them to fill up any available space, if they're allowed in. Enough private car drivers have weird biases about this that they'll drive their private cars even if it's significantly *slower* than the alternatives. In a city big enough and dense enough to need *mass* transportation, you *cannot* avoid congestion as long as you allow private cars into the lanes. Heck, if the lanes open up enough, you even get joyriders clogging them up.
and likely far cheaper in exchange for the longer timeframe.
Yeah, cheaper than a private SUV!
It'd operate sort of like an elevator, except each car could operate in a pre-determined space and it'd automatically snake itself around that area as efficiently as possible as people request pick-ups and drop-offs. It's a hybrid of public and private transportation, and I Imagine it'll be nothing like any current form of public transportation.
Interestingly, elevators actually have a problem with scaling up to huge passenger volumes, you just don't see it very often because most buildings don't have nearly enough people travelling for it to be an issue. It shows up in skyscrapers where added elevators to combat the traffic start crowding out rentable space.
Look, this is pretty much automated taxis. There's massive inefficiency in taxis right now -- drivers all clustering around the most profitable locations, for example -- so automated taxis are great. But it isn't *mass* transportation; it doesn't scale up. Consider what percentage of people take taxis vs. subways in major cities which have both. It's not just due to the price... it's because during rush hour, the taxis are stuck in traffic, and the subways are actually *faster*.
Mass transportation is what you need when you just have so many people moving that you can't give everyone their own two-ton armored vehicle and still fit them into the street. (And the distances are too far to walk.) That's why it was called "mass" transportation.
Personally, I see a market for this in dense, walkable urban areas (like where I live) for something between an Uber and a city bus.
Maybe, but it'll face the same congestion as Uber or a cab. In rush hour it will be slower than a subway, metro, light rail, or Vancouver Skytrain. (Look, fully automated electric trains, already in use.) This sort of "intermediate" option is not mass transportation, it's a driverless taxi.
Which is fine, except that Musk claimed he was addressing "high passenger-density urban transport". Which shows that he *does not know what he's talking about*, because he immediately proposes smaller buses which have inherently lower density.