I spent ten years in the trenches of American urban design policy. The best we could do was lose very slightly less quickly. It's not changing. Trains are great, we should build more, and we probably should replace a lot of bus routes by subsidizing rides on Waymo and its ilk. It'll be cheaper and provide better service.

A well run public transit system should obviously be cheaper at scale than robotaxis, but the incentives for Waymo (or Uber, or Lyft, etc.) are very different than the city's incentives. It's very possible that in practice private companies can operate more cheaply at scale than buses because they have much higher incentives to reduce costs and increase efficiency.

“Self driving cars aren't the next stage of public transport; they're a bandaid solution to American urban design.”

That might be an unintentionally excellent analogy, because like a Band-Aid, self self driving cars have the potential to heal the urban environment. The widespread adoption of self driving cars doesn’t take cars off the road, but it does reduce the reliance on destination parking. Entire city car parks could be replaced with medium density residential, substantially increasing density and paving the way to walkable cities.

It's not a bandaid because American urban design isn't going to change substantially. I don't see American cities changing their mind on how they build and where they build.

They won't be better for maintenance but unless Portland can build the state capacity to fund public transport properly this is better than nothing. Plenty of developing countries rely on buses, jitneys, and low footprint vehicles like mopeds for traffic flow because they don't have the state capacity to enforce an urban framework conducive to public transit. Honestly many US states are the same.

> increased pavement wear

That's buses. Even more with electric buses. They are insanely subsidized by public. Robot taxis are vastly cheaper for everyone.