The price tag for the tunnel is pretty much what you’d expect at the low end of civil engineering costs, so there’s no room in the pricing if things start to go south. Musk’s entire premise is that there are magical inefficiencies in the commercial tunnel boring machines he’s purchased, and he assumes that spacecraft and automotive engineering expertise is sufficiently transferable to uncover those opportunities and redesign a considerably more efficient and cheaper machine. Oh, and using commercial automobiles will be faster, cheaper, and more pax-intensive than light rail solutions. I’m … skeptical. Let’s go with “skeptical” on this one.
Based on my ballpark guesses at the cost of the tunnel, I would be surprised if The Boring Company actually made any money on this. They're liable for all cost overruns, and they have to bear the costs of developing the custom Tesla car manufacturing and safety qualification themselves.
The most galling part to me is that I can't see how they make the mandated 4400 people/hr limit. That requires 26 second headways between vehicles (assuming that 4400 people/hr is a bidirectional capacity, not unidirectional), and I don't see them getting station dwell times low enough to permit those headways--airport people movers usually run around 30-45 second dwell times.
You're thinking in train terms. There's nowhere near 26 second headway between vehicles in busy tunnels near me - cars enter and exit the road from multiple sides before entering and the headway in the tunnel is much shorter. 30-45 second headway makes sense when your station stops blocks the entrance to the tunnel, which is the norm for rail systems, but not for car tunnels.
Wouldn't the obvious solution be to multiplex cars at each station? They could have twenty Teslas at each location for passenger pickup/drop-off, only entering the launch queue when all occupants are seated and restrained.