I wish he'd stop saying stuff like that. Let the software speak for itself. When he makes bold predictions like that, my eyes roll given his previous predictions about FSD.
Agreed.
Part of the problem is that his tweet is read and interpretted by a wide range of people, and the nuance/details he may think are implied are probably going to be ignored or missed.
For example, in Elon's line:
"We are starting to get to the point where, once known bugs are fixed, it will take over a year of driving to get even one intervention."
As Usain mentioned, the "over a year to get even one intervention" is probably related to v. 12.6 or even beyond...so it could just be a prediction based on early looks at far-ish future versions. But, critics are going to read the sentence as an evaluation of the current, publicly-available build...which is definitely nowhere near a year between interventions.
I also suspect that the "once known bugs are fixed" part is doing a lot of heavy lifting. There are probably a lot of known bugs, and those fixes are going to take a long while. And, while those bugs exist, people are going to be intervening multiple times a day.
It would also probably be useful if Elon was more specific about his meaning of "intervention" -- I suspect he means a critical or safety-related interventions. Minor interventions are still very frequent in many areas if you don't want to annoy other drivers. I "intervene" multiple times a day when the map data gives the car a 25mph speed limit in a 55mph zone and it starts to slow down. All I have to do is raise the speed via the scroll wheel...but I think that's still an "intervention.". (Is faulty map data is considered a "known bug?"). I also intervene to have the car behave more naturally, for example so it's quicker to pull away from stop signs when other cars are around. Not normally a safety issue...but helps to make sure other drivers aren't confused or annoyed.