It has some rather tenuous arguments in it. For example it posits that one family might summon 3 robotaxis to drop the kids to school where they would previously have only used one car, but conveniently omits the idea that families might save money by carpooling so that one robo taxi actually picks up 5 kids in the same neighbourhood and takes them all to school, or that mom and dad now don't need a car, so many of the cars driving on the road or parked on streets would disappear.
Depends on the cost of the RoboTaxi. If cheap enough (and probably to start they will be cheap) then I can see huge numbers of generated journeys. If I may put on a "slight conspiracy theory" hat, I would expect the business model would be to flood the market with cheap taxis, then as other options dwindle (who would want to go on a bus if a taxi is cheaper?) and a monopoly is formed then prices go up.
On the parking, I don't get it. Robotaxis will need somewhere to sit when empty unless we expect them to aimlessly tour the streets between taxi rides. Sure, the number of Robotaxis would theoretically be lower than private vehicles because they would take multiple people on multiple trips, but that doesn't account for any model shift (eg. from buses) and no generated journeys. In reality I would expect many more journeys to be made and we end up solving very little with traffic or parking demands.
