Checked out Tesla's results call and was surprised how irrational Musk was on a bunch of key issues. Likely he got arrogant and does not go deep enough when considering a topic but they can't afford that.
Faster charging, Musk was dismissive about its importance. Lots of things to consider , optics vs ICE, competition, how people in different geographies live and drive, what they want, where range becomes sufficient and charging times more important. length of a trip in minutes not miles. For EVs to have over 90% share, in 10 years, they need to charge in 5 mins, no doubt. Excluding car as a service where battery chemistry and charging have to be optimized for cost.
Here is a key to Elon-speak. Elon Musk is quick to promise things he's worked out the Physics on and knows they are Physically possible. He may be optimistic about how long it's going to take for delivery, but he knows it is physically possible. If Elon dismisses something as "unimportant" he's also run the numbers and found that whatever the person is asking about is either physically impossible or so cost prohibitive it isn't worth talking about.
I'm sure he's run the numbers and with the battery tech that he is pretty sure will be out there in the next 5-10 years, charging in 5 minutes is not going to happen.
For EVs to beat out ICE in the market, first the public needs to be educated and they are still quite ignorant of EV capabilities. Then the cost of batteries need to come down to a point where an electric drive train with decent range can compete pretty much toe to toe with comparable ICE. We're not quite there yet, but the Model 3 is close. The cost of ownership is less, but people need to see a lower sticker price.
Third EVs don't need to charge as fast as an ICE, they just need to get up into the same ranges as ICE. On long distance road trips, few people drive more than about 8 hours in a day. Some people will want to make minimal stops, but they would be willing to make one stop for a meal while the car charges. Then if it was common for lodging places to have chargers, people would just plug in the car at their stopping place for the night.
For day to day charging EVs are more convenient than ICE if you have some kind of charging while you park at home or work. If you don't it might be a bit of a hassle, but apartment/condo chargers will start to become more common and cities will install AC charging in neighborhoods where people need to park on the street. If cities charged for it, the system could pay for itself.
Car as a service, they keep talking about sharing not their own fleet. Don't wanna say much on this one actually but there are many reasons why they need their own fleet. They need different hardware for ownership and car as a service, sharing is less capital efficient and further down the road people won't buy cars if car as a service is available, there won't be cars to share.
There is a lot of talk about this and while I think car sharing services will grow and be especially popular in city cores, I suspect a lot of people will still prefer to own their own car.
As far as Tesla getting into this game, there is the risk that car makers that get into the ride share business could run afoul of antitrust laws. United Airlines was once a division of Boeing. They were split off by an antitrust suit.
Autonomy, again they talk about averages but that's like killing a person to save 10 and we don't do that. If i help an old lady cross the street, I don't get to punch a cop in the face. Even if I help two old ladies cross the street, I still don't get a free punch. Car crashes are judged on a case by case basis because every person has human rights and the right to life is pretty important.
Autonomous cars have to be better than the best human driver in every possible scenario. That's what's sufficient, anything less is not good enough. The average driver is a terrible metric , the median driver is likely already 10 times better.
Might seem hard to do it it but the best human driver has an atrocious perception-reaction time, limited experience, limited sensors, limited precision and ability to quickly choose the ideal solution. Its one asset over current computers is its perception, it might be slow but it can interpret the sensor data pretty well.
The autonomous car has much better sensors, 10-100 times faster perception-reaction time, high precision, will always choose the ideal path and all it needs to absolutely crush the best human, is for it's perception software to not suck.
If in a give situation, a human would have done better, the car is to blame. That is just and fair.
Humans have some advantages over an autonomous system. Humans are far, far better at filtering unneeded data and humans can act on instinct. The airliner that landed on the Hudson River 10 or so years back, the pilot made a last second decision that saved everyone's life that was nobody had ever thought of doing. He nosed into the water instead of coming in for a belly landing (how everyone said it should be done). The problem with a belly landing with the engine layout of modern airliners is there is a high risk the engines when they catch the water will twist the wing off and take a chunk of the fuselage with it, allowing the plane to sink immediately. By nosing in, the nose of the plane took most of the impact and the wings were moving slow enough when they hit the water, the engines didn't rip the wing off.
The pilot did something no computer would have done. People can do that, a computer won't. When you throw an AI into a situation that is too bizarre and not tested in the lab, anything can happen, including very bad results.
That's the nightmare autonomous driving developers have to deal with. The number of edge conditions with a plane are smaller than the number of edge conditions with autonomous driving, though the stakes are smaller. Planes come apart much more spectacularly than cars usually do. Car accidents are much more common than aircraft accidents, but they are more survivable if the car doesn't disintegrate or crush too much.
Ultimately you have to play by the odds with this stuff. Car accidents will happen whether computers or humans are driving. Our culture is more forgiving of human error than computer error, so the standard we hold autonomous cars to is a much higher bar.
With the analysts, Musk messed up too. Tesla has a communication problem , journalists and some analysts are not informed at all on EVs and Tesla. For years Musk has been getting upset instead of acting to address the problem. The call is an opportunity for Tesla to address certain things, to inform, educate. And ofc, Tesla messed up with M3 production and now they are getting spanked a bit but that's not unfair.They have a communication problem and they've made it worse.
So yeah, Musk is supposed to be rational and its alarming to see that Tesla is on the cusp of making huge mistakes, silly mistakes. Maybe he needs to hire an adviser that doubts everything instead of always agreeing with him. Or maybe he needs to allocate more time for Tesla and not rush to conclusions.
Other than that it was all good, they seem to be sorting out M3 production but they'll be really busy for the next few quarters.
Tesla is not great at communication. The public wants to hear from Elon, but really he is not that great at conveying what's going on. He over thinks what he's trying to say.