those decades are going to be shorter than you think. Predicting the future is notoriously humorous in hindsight.

the present rate of progress - that's the point about disruptive technologies, and "S" curves. They ain't linear, but look it at first. They provide an advantage that established tech can't compete with and can't change till they have paid off their investment that established them. So the disruptives have an easier wave to surf.

I can remember when the saying was no-one ever got fired for buying IBM - Huh? They is big, but there are bigger competitors now! I can remember when hard disks were never in a position to compete with mag tape, but were predicted to, where did that go? RAM was never cheap enough to supplant mag core, er, um, that ship sailed within 5 years. Scale brings costs down, and with a few additional pluses, off it ramps, in an "S". It is the coming together of several elements that give the advantages, we tend to see parts of the aspects and tut, not the whole.

My worry with transport is the inevitable mix of AV and human drivers, the idiots will drive to it (at vehicles) and expect the AV to respond the way they think it will. They do it with human drivers and look at road deaths! And who ya gonna believe? Who shouts the loudest, idiot or AI code?

Watch this space.