Kicked out of the FSD Beta! (Long Post)

Hi everyone, apologies for the long post, but I feel like my M3 investment deserves this airing! I’m not a super-frequent visitor here, but have read many threads over the 2.5 years I’ve had my beloved Dual Motor Model 3. I’m posting today to explain how I was unceremoniously booted from the FSD beta program last week. But before I get into that, a little background.

First and foremost: I love my M3. It has changed my life. I previously had a Subaru Forrester. I’m a consultant and drive about 1-2 hours per day, going to various client sites. For a couple years leading up to the M3 purchase, my body, and particularly my back, would be *hurting* after longer drives in the Forrester. I was honestly asking myself how I was going to keep going with my career, with the aches and pains that driving was causing me. Then I looked into a Model 3. And the rest is history.

I use autopilot 90% of the time I’m in the car. Always paying attention, of course, but able to rest my body. No feet pressing the pedals. No hands clutching the wheel. It’s astounding what a toll that takes on the (50-year-old) body, day in and day out. I can drive 2 hours in the M3 and not feel like I was in the car at all. That's life changing!

Now, first of all, I’m no Elon/Tesla complainer/hater. I love the car so much, tech warts and all. Auto-wipers suck—don’t care. Road noise is significant—don’t care. Unresolved trim issues at purchase—don’t care. The car is just amazing. Best purchase I have ever made, without question. Second: I pay attention when I drive, and know how the Model 3 works. I have logged about 45,000 miles on autopilot. I know how to pay attention, and I know when and how to prove that to the M3.

Now for the sad part. The FSD beta was simply yanked from my car last week, with no email notice, and little to no in-car warning. If something was presented on the screen, it was small, looked like any other message, and, ironically, it must have been showing when I was driving and paying attention to the road!

I got the FSD beta a few weeks ago after a painstaking 30 days of getting a 99 safety score, despite driving in really difficult areas of Boston and the suburbs. All sorts of people hard breaking in front of me, cutting me off, etc. But I was careful, as I always am, and maintained a 99 for a month which was no small feat. I got the FSD beta and was really psyched. Unlike some of you, I didn’t read a ton about it ahead of time. No time for that with 3 younger kids and a demanding job, all seriously affected by the pandemic. So I didn’t fully understand many things about FSD except for the fact that 1) I paid many thousands of dollars for it. 2) I’ve waited two and a half years for it. And 3) it doesn’t work very well so you have to be very careful.

I used FSD the day it downloaded and was struck by how bad it was on city streets. I was really surprised. It was exciting to see it trying to do things it couldn't do before, but I stopped using it. I pretty much only double-tap the gear stalk on major roads, and after that first bad experience, didn’t try it again for a week or so on city streets. The next time I tried it, in hindsight, I noticed extensive nags about paying attention. I was confused because I *was* paying attention. But I guess I did change audio sources, change A/C temps., etc. This is par for the course with standard AP and I never had an issue.

Then last week, I was on a back road. No one around me. I was changing the radio stations during a morning commute and then I had to change my navigation destination. I simply clicked on navigation and switched destinations to one of my preset favorites. AP freaked out and disengaged. And then I saw the message. I was booted from the beta.

Then ensued tons of reading. I discovered that I was doing it wrong. I should have turned off FSD during my normal commutes and ONLY used it when I wanted to “test” it and pay 100% full attention to the road without even changing the radio station. But I didn’t know. And now after 2.5 years, thousands of dollars, 45,000 miles of safe AP driving, and 30 days of excruciatingly safe Safety-Score driving, I’ve lost what I only had for a few drives, and may never get it back. Man, that SUCKS!

Now a couple of questions for those of you more in the know than me.

1) Does anyone know if I will get FSD beta access back? When?
2) If I email Tesla and officially back out of the beta, will I get my old less-nagging AP back?
3) Can I then re-apply for the beta and go through another safety score test and get FSD back?

So frustrating not knowing the answers to these questions. Of course I called Tesla and the guy on the phone was useless. I do have service scheduled next week and will ask, but I’m pretty sure they won’t know anything either.

For those of you with FSD beta access right now, be careful, and learn from my mistakes!

 

[mrbulk]

One thing bothers me about them upcoming robotaxis - dirt.

Meaning OPD, or Other People's Dirt.

After Covid I've had to rethink a lot of what I used to do naturally, and probably far too casually, when out and about with the public. Would prefer that I drive (or be driven in) my own private vehicle. You never know who that taxi just dropped off. Mavbe at a hospital.

 

[nativewolf]

That has always been the case. But to argue there is no money in robotaxi service capability is just not being serious.

Sure...it is very serious. If you want to talk money look at Uber. Tell me how much money they make (Spoiler: they have lost billions & billions). Then tell me how a robotaxi fleet s fundamentally changes economics and increase addressable market while making a profit. DO NOT USE ARK MODEL. What a pile of crap. The fact that the entire shared services model user market had flatlined prior to the pandemic should provide a giant giant caution sign to anyone putting values on robotaxi because the model does not work for most people. Put the total addressable market at the number of people that regularly use an Shared Services Provider (uber, lyft) and I think you have a fair proxy for robotaxi market. That's not a big market. You'll still have enourrmass competition from uber etc. The savings from not having drivers and fuel are offset by having to own service centers in metropolitan areas where real estate is expensive. It will need to be an area with excellent electrical connections, safe, and able to attract employees to maintain the fleet (cleaning etc).

Then to knightshades point re timing, Waymo will be there before Tesla in the geofenced areas. However, those geofenced areas are the places where most shared service rides originate. Today you can take a Waymo in SF that is fully automated, no driver. They are expanding robotaxi service to downtown Phoenix. They are collecting data much slower but for a robotaxi fleet I don't think it matters. Full disclosure I don't live in SF and I have not ridden in a Waymo. Fundamentally Tesla has much to do to get real driverless FSD (not that limp FSD Beta thing they are testing) in a car and that requires new hardware and that is not happening tomorrow. Then they have to rewrite all the code, test for a year, debug, recompile, retest. Rinse repeat. Not happening soon. Anyone ignoring this has their head in the sand.

I think robotaxi could do wonders for the public in places not served well by Uber. Places like Marion VA or Elizabeth City NC or the equivalent but the economics won't be great. 3-4 of those running in a little town 45 miles from the other little town means that you'll need a service center in each little town, someone to clean them, etc. It's why taxi service in small towns is so ...poor.

Why don't I trust ark to wash a toliet: Ark places an exceptional value on Palantir. A company, like Uber, that makes no money. Unlike Uber Palantir is borderline fraud, it is a services company masquerading as a technology company. It's comp is any other public traded IT consulting company. Infosys or something like that. I don't believe Ark has a model I'd trust for anything. I don't think they fundamentally understand business. Anyone putting value on Panantir hasn't gone to the state dept and sat down to see what palantir "technology" is actually doing.

 

[nativewolf]

One thing bothers me about them upcoming robotaxis - dirt.

Meaning OPD, or Other People's Dirt.

After Covid I've had to rethink a lot of what I used to do naturally, and probably far too casually, when out and about with the public. Would prefer that I drive (or be driven in) my own private vehicle. You never know who that taxi just dropped off. Mavbe at a hospital.

You're not the only one. Shared Services user numbers plummeted. Also, the actual market was not large before the pandemic. It shrank. It has rebounded but not to prior levels. For people that use it, it works and they tend to use it more. Global $25 billion in total bookings in 2021 and 1.7 billion rides worldwide. Average trip of 13 bucks or something. Demand surges up and down, lots of the day there is very little demand. The trend is for people that use it to use it more often and to shed users that were infrequent users. I just don't see a model to get excited about.

10500 cities served but 25% of bookings in 5 cities. I bet in the top 20 cities you'd have 75% of revenues. Uber has struggled and struggled to make profit and have destroyed huge amounts of shareholder equity.

Then on top you throw a very heighted concern of communicable diseases and the knowledge that many uber rides are for drunken people in the evening (a great service) and ...well say no more. Uber drivers clean up, it's their car. I have friends that drive for Uber and cleaning is something that's done.

 

[Knightshade]

Sure...it is very serious. If you want to talk money look at Uber. Tell me how much money they make (Spoiler: they have lost billions & billions). Then tell me how a robotaxi fleet s fundamentally changes economics and increase addressable market while making a profit.

Because they don't have to pay for drivers?

How is that even a question you needed an answer to?

Uber and the labor market: Uber drivers’ compensation, wages, and the scale of Uber and the gig economy

Uber drivers earn low wages and compensation, and the total hours and compensation in the gig economy represent a very small share of total hours and compensation in the overall economy.

www.epi.org

Look for "earnings before expenses" that's what the drivers get net of what goes to Uber. It's over $20 an hour.

100% of which would now go to the rideshare company since there's no driver.

You don't think Uber would be profitable if they collected an extra $20/hr from every hour of rides?

BTW further down it cites # of driver-weeks at 24,832,168, and average of just over 17 hours per week- which is ~422.147 million hours a year. At $20/hr that's almost 8.5 billion dollars of extra revenue.

If you back out "losses" from other companies Uber holds stock in, their OPERATIONAL loss Q1 2022 was only 400 million...or 1.6B annualized.
So they'd show a profit almost 7 billion dollars if they got that extra money.... (well a bit less due to the mainteance and fuel costs would fall on them-- but again EVS have vastly lower costs in both)

On top of that (if it increased TAM enough)- RTs could charge HALF what Uber does for rides and still be significantly profitable

 

[besnova]

I got 3 strikes but I thought it was moved to 5?

 

[SVMike]

Because they don't have to pay for drivers?

How is that even a question you needed an answer to?

Oops...You are missing the part where the money that goes to the drivers is also paying for the driver's vehicle, insurance and fuel

 

[Daniel in SD]

Oops...You are missing the part where the money that goes to the drivers is also paying for the driver's vehicle, insurance and fuel

I believe the data he's using is net income.
Where I live a 30 mile 1 hour ride in an UberX costs $58.72 (7.50+.3*60+.87*30+1.15+5.97). $20 net income for the driver seems plausible to me.

EDIT: seems like the $20 is not net income though it seems like they must be counting time waiting for a fare. Bottom line is that customers are paying about $2 a mile. I think robotaxis could be quite profitable at that rate.

 

[nativewolf]

It's a lot profit business because you have to pay a human driver and pay NYC gasoline prices.

Switch that to $0 for labor and EV rates and suddenly the profit ain't so low.

The other bigger piece there is exactly all those smaller cities you mention-- because suddenly all the places it doesn't make sense TODAY to have significant taxi service due to labor and fuel costs, it now DOES make sense... meaning your TAM is much much larger than todays Waymo or taxi services.

It's a very big market.



My dubiousness isn't on there being a large, profitable, market, the math is pretty clear there--- it's on getting to generalized L4 anytime soon.

I'm also dubious of the MEGA RT bulls who think they'll end individual car ownership but that's another topic... (I do think they'd reduce the # of households that have 2-3 or more vehicles though)

First let me say that I really enjoy all your posts on FSD. Many thanks.

Finding service center space in NYC is simply a matter of economics and some physical constraints but it is going to cost some $. You will need a space with lots of juice- an issue in NYC, you will replace drivers with cleaners - unionized cleaners in NYC. I don't see the lots and lots of profit but maybe we'll get to see sometime.

Small towns: Sure, could be lots of demand for taxi in a small town but you'll have to have a service capability. They demand dies away after dinner. Small town life but it means utilization is low for many hours of the day. Maybe they'd run Teslas out to Lumberton from Wilmington or Raleigh or Fayetnam and then back? Do you have an employee and service center in Lumberton? I don't know. I do know that taxi service has never made it big in small towns. I guess I am hopeful but skeptic.

 

[Knightshade]

First let me say that I really enjoy all your posts on FSD. Many thanks.

Finding service center space in NYC is simply a matter of economics and some physical constraints but it is going to cost some $. You will need a space with lots of juice- an issue in NYC, you will replace drivers with cleaners - unionized cleaners in NYC. I don't see the lots and lots of profit but maybe we'll get to see sometime.

What do you think the robots Tesla is making are for?

They'll plug and unplug robotaxi cars at superchargers, and clean them too.

Small towns: Sure, could be lots of demand for taxi in a small town but you'll have to have a service capability. They demand dies away after dinner. Small town life but it means utilization is low for many hours of the day.

That's why it works

With a taxi you have to have a human sitting around all the time waiting for fares.... so intermittent fares sucks for this model. No human will do this job for ONLY fare money because they won't make enough to live. Thus no service at all, or very limited offerings.

I can't even get an uber to my house half the time because I live "in the country" despite being not terribly far from multiple major cities because it makes no sense for a human uber driver to hang around here waiting for call once in a while.


With an RT it just sits there asleep until the app wakes it. So it makes perfect sense to actually service areas like mine now. Even better, the car can sit there 24/7 versus the single uber driver that needs to go home and sleep and live their life most of the time.

Maybe they'd run Teslas out to Lumberton from Wilmington or Raleigh or Fayetnam and then back? Do you have an employee and service center in Lumberton? I don't know. I do know that taxi service has never made it big in small towns. I guess I am hopeful but skeptic.

As I've said I don't buy the RTs will replace first cars in most households idea-- so most trips won't be nearly that long.

But if I could reliably get service here, as I expect I could with RTs, there'd certainly be nights we were say going out to a nice restaurant or a bar and planning to drink, and I'd happily take an RT both ways instead of one of us having to DD because we know we'll never get an uber to pick us up here.

Same for going to the airport when travelling-- right now we end up parking at a friends house who lives 10 minutes from RDU because you CAN get rideshares to/from his place easily. If I could get an RT right from my house reliably I'd do that instead.

RTs in small towns is a slam dunk BECAUSE it's finally economically possible for them to make money.

Oops...You are missing the part where the money that goes to the drivers is also paying for the driver's vehicle, insurance and fuel

Ooops- you are missing the actual words I used

"earnings before expenses"

The fuel and insurance being the expenses. So not only did I not "miss" that part I specifically called out that one line of data did not include it.

The link I provided breaks it down for each piece of those too if you'd have actually read it.


But guess what? Those are MUCH LOWER for EVs. (ditto maintenance costs)

Especially Tesla EVs that would carry Tesla insurance.

 

[nativewolf]

Because they don't have to pay for drivers?

How is that even a question you needed an answer to?

Uber and the labor market: Uber drivers’ compensation, wages, and the scale of Uber and the gig economy

Uber drivers earn low wages and compensation, and the total hours and compensation in the gig economy represent a very small share of total hours and compensation in the overall economy.

www.epi.org

Look for "earnings before expenses" that's what the drivers get net of what goes to Uber. It's over $20 an hour.

100% of which would now go to the rideshare company since there's no driver.

You don't think Uber would be profitable if they collected an extra $20/hr from every hour of rides?

BTW further down it cites # of driver-weeks at 24,832,168, and average of just over 17 hours per week- which is ~422.147 million hours a year. At $20/hr that's almost 8.5 billion dollars of extra revenue.

If you back out "losses" from other companies Uber holds stock in, their OPERATIONAL loss Q1 2022 was only 400 million...or 1.6B annualized.
So they'd show a profit almost 7 billion dollars if they got that extra money.... (well a bit less due to the mainteance and fuel costs would fall on them-- but again EVS have vastly lower costs in both)

On top of that (if it increased TAM enough)- RTs could charge HALF what Uber does for rides and still be significantly profitable

Nope I don't think Uber will be suddenly profitable if they ditch the drivers. Today they keep 5 billion or something, that's after drivers. They can't run a dispatch service for $5 billion. Drivers provide the cars, they provide the "service" they provide the cleaning, the fuel it. This winter while we waited for bankers to close our loan I had the time to spend several days digging into the taxi and ride share markets. I came away very skeptical of the potential for robotaxis as some miracle driver of profits.

I believe that robotaxi advocates overestimate the addressable market, they dismiss issues of service, they disregard the general low profit of everyone in the taxi business. Taxi's only really made money in a few cities in the USA. It is shocking how few. Uber is no different. The problem isn't the drivers, if it were Robotaxi's would be a no brainer. Robotaxis appeal to people that would use Uber. It's not many people, they are mostly in big cities. Uber was a good brand but created a myth that this is a great business but it's not. Its a service business and service businesses have very modest profit margins. 18 million rides a day on Uber. 4.5 of them are in NYC, London, LA, Sao Paulo, and Chicago. Sounds like a lot? Take london where the subway/Tube did 4.8 million rides before the pandemic or NYC where it handles 5.5mln before the pandemic. The subway in 1 city doing more rides than Ubers top 5 cities combined. In London most of the rides are at night, for people going out drinking. This is really where Uber shines, it provides a service for people barhopping and this is not just consistent with Ubers self reporting but matches my experiences when i was doing consulting. Nothing beat Uber for bar hopping. When we'd go out with clients we might have hit 4-5 bars or restaurants in a single night in DC or LA, each one a different ride. No problem parking, no drunk driving. Super. Huge dense urban metro regions with very high incomes and lots of nightlife and traffic congestion and poor parking options. That's Uber's profit center.

Then the myth that drivers costs Uber money:
A full time uber driver can do 10k miles a month or about 300 miles a day in a 12 hour period.... Crazy stupid milage. 120k miles a year. Uber is often paying drivers less per mile than the IRS rate of $0.57 (rates in San Diego for example it is $.37). So no, I don't see a fountain of profit. I think that Uber drivers actually subsidize Uber. It's why most drivers don't keep at it. They are better off at Starbucks.

I'm not a sceptic on FSD as an enhancement to vehicles. I could see lots of people wanting a chauffeur. FSD as an enhancement..that I believe.

 

[Daniel in SD]

(rates in San Diego for example it is $.37).

Here is what Uber charges in San Diego. I'm trying to figure out how Uber could be so unprofitable if they're only paying drivers $.37 per mile.
What is your estimate for the cost per mile to operate a robotaxi?

 

[nativewolf]

What do you think the robots Tesla is making are for?

They'll plug and unplug robotaxi cars at superchargers, and clean them too.




That's why it works

With a taxi you have to have a human sitting around all the time waiting for fares.... so intermittent fares sucks for this model. No human will do this job for ONLY fare money because they won't make enough to live. Thus no service at all, or very limited offerings.

I can't even get an uber to my house half the time because I live "in the country" despite being not terribly far from multiple major cities because it makes no sense for a human uber driver to hang around here waiting for call once in a while.


With an RT it just sits there asleep until the app wakes it. So it makes perfect sense to actually service areas like mine now. Even better, the car can sit there 24/7 versus the single uber driver that needs to go home and sleep and live their life most of the time.








As I've said I don't buy the RTs will replace first cars in most households idea-- so most trips won't be nearly that long.

But if I could reliably get service here, as I expect I could with RTs, there'd certainly be nights we were say going out to a nice restaurant or a bar and planning to drink, and I'd happily take an RT both ways instead of one of us having to DD because we know we'll never get an uber to pick us up here.

Same for going to the airport when travelling-- right now we end up parking at a friends house who lives 10 minutes from RDU because you CAN get rideshares to/from his place easily. If I could get an RT right from my house reliably I'd do that instead.

RTs in small towns is a slam dunk BECAUSE it's finally economically possible for them to make money.






Ooops- you are missing the actual words I used

"earnings before expenses"

The fuel and insurance being the expenses. So not only did I not "miss" that part I specifically called out that one line of data did not include it.

The link I provided breaks it down for each piece of those too if you'd have actually read it.


But guess what? Those are MUCH LOWER for EVs. (ditto maintenance costs)

Especially Tesla EVs that would carry Tesla insurance.

If a RT is sitting around waiting to take someone on a ride that's nice for you and me but bad for the robotaxi owners. Therefore robotaxi owners will logically flock to the cities that keep them going the longest.


We do know Uber pays drivers less than the IRS rate of $0.57 in many cities. The drivers pay per mile is shocking low. Even though the tesla robotaxi fleet is an EV fleet likely insured by Tesla it still has a milage expense of something, say 1/3 of an ICE (fuel, tires, etc- just guessing so go with it). Tesla would have to run a service center (with an employee) charge vehicles, clean them, etc and do all of that for $0.17/mile to approach the fees Uber pays for their fleet. That's not including the cost of the huge investment they've made in software (no idea how that is allocated). In short, Uber is paying below market rates on a $/mile basis. Tesla can do better but you have to ask yourself...is this really such a great business? My research leads me to believe that it is not.

When I first read Tony Sebas work on transportation I was sold. I'm just not as sold anymore. I think the small town is a far far cry from a slam dunk. You still have to have someone clean it, it has to charge, it has to be serviced in all other ways. Maybe over a long period of time it might happen but I think that will be slow and painful.

 

[Knightshade]

If a RT is sitting around waiting to take someone on a ride that's nice for you and me but bad for the robotaxi owners.

Why? There's essentially 0 marginal cost to sitting there asleep.

For most car owners today their car already does that 16+ hours a day with no chance to ever make them money.

If they instead throw it on the RT network when not in use they CAN make money, even if it's occasional.

Therefore robotaxi owners will logically flock to the cities that keep them going the longest.

I mean, if you already live in one- sure. But I expect there most RTs will be commercial fleets, just like today.

Where I live nobody will run a fleet. But there's already a handful of Tesla owners. As described above, some will be thrilled their car can make them money while it'd otherwise be sitting around useless.

So suddenly an area it never made sense for any taxi to serve makes a lot of sense.

Tesla would have to run a service center (with an employee)

Why?

Most EVs rarely need any service at all, outside of tires and windshield wipers-- neither of which requires Tesla specialists.

charge vehicles, clean them, etc

Tesla already owns the chargers (and I already mentioned one of the first jobs bots will do is plug/unplug teslas and clean em)

But of course for things like occasional small town rides where it's local owners collecting extra $ there's no need to even charge them there.

If you're only doing a half dozen 20-30 mile trips in a day you can do that without need to recharge while on shift at all.

I could let my Tesla collect whatever local fares are out there for 18 hours a day, then have it drive itself home and I plug into my L2 charger where it costs me $2 to add 300 miles of range. Then send it back out again.

If I ever need to actually go somewhere, I call it back.

.is this really such a great business?

It doesn't need to be a GREAT one.

It just needs to not lose money for the operator. Which the math suggests should be trivially easy once you remove the costs of paying a human, restricting hours to human availability, and remove most of the fuel and maintenance costs that gas cars have.

 

[nativewolf]

Here is what Uber charges in San Diego. I'm trying to figure out how Uber could be so unprofitable if they're only paying drivers $.37 per mile.
What is your estimate for the cost per mile to operate a robotaxi?
View attachment 812004

This winter I had time to poke around. There are a ton of reddit spots where people talk about rates. Uber and Lyft have both been under pressure because they are not profitable. They've been cutting peoples effective pay ever since the pandemic. It wasn't an investment analysis for me nor an MBA project. I was just interested in the model.

It was late and I work (forestry) for a living so I'm beat at night. I should have put that uber price to consumer there as well, thanks. Uber is paying drivers 37 out of 87 cents per mile driven. I still think that Tesla has a per mile operating costs (Tires, Juice, et) somewhere in the $0.15 to $0.25/mile. That leaves very little room to do all the things that are a service business. Washing and cleaning, charging, doing the exception handling, etc. Then you have to pay for the robotaxi, both the car and the software. Then you have to run the dispatch service. Yeah...I don't see massive profits. If there are profits it is not from eliminating drivers it is from eliminating the stupid stuff but I would have to think that Uber was cutting as fast as cut could go. Maybe not though, I was not interested in Uber but the business. What I found was despite huge brand awareness and a global footprint the platforms don't yet make money. Most telling to me was that Uber lies with stats. The real truth of the matter with Uber users is that there are very few people that have found it is a useful day to day tool. It has specific use cases , bar hopping, meetings, travel, etc in very dense urban metros that are both business and nightlife centers.

@Knightshade see I don't think that 0.25/mile on a car driving 300 miles a day pays for a service center and someone to clean the car in a small town.

 

[nativewolf]

Why? There's essentially 0 marginal cost to sitting there asleep.

For most car owners today their car already does that 16+ hours a day with no chance to ever make them money.

If they instead throw it on the RT network when not in use they CAN make money, even if it's occasional.




I mean, if you already live in one- sure. But I expect there most RTs will be commercial fleets, just like today.

Where I live nobody will run a fleet. But there's already a handful of Tesla owners. As described above, some will be thrilled their car can make them money while it'd otherwise be sitting around useless.

So suddenly an area it never made sense for any taxi to serve makes a lot of sense.






Why?

Most EVs rarely need any service at all, outside of tires and windshield wipers-- neither of which requires Tesla specialists.




Tesla already owns the chargers (and I already mentioned one of the first jobs bots will do is plug/unplug teslas and clean em)

But of course for things like occasional small town rides where it's local owners collecting extra $ there's no need to even charge them there.

If you're only doing a half dozen 20-30 mile trips in a day you can do that without need to recharge while on shift at all.

I could let my Tesla collect whatever local fares are out there for 18 hours a day, then have it drive itself home and I plug into my L2 charger where it costs me $2 to add 300 miles of range. Then send it back out again.

If I ever need to actually go somewhere, I call it back.




It doesn't need to be a GREAT one.

It just needs to not lose money for the operator. Which the math suggests should be trivially easy once you remove the costs of paying a human, restricting hours to human availability, and remove most of the fuel and maintenance costs that gas cars have.

Ok, obviously you don't drive for Uber. Go talk to some Uber drivers, hurry because the business is lousy. They have to be cleaned. Every day, every shift. Then washed as well, every day if you want it to look as nice as the best Ubers. So every shift you have 30 mins to an hour just cleaning. You've got a lot more than $2 in costs.

I also don't think individuals will do RT instead it will be fleets.

I assume that no one in their right mind will let the car be a RT for long, not at $0.17 or $0.20/ mile maybe at the full $0.37 that uber is paying drivers (in CA-not the lowest price spot at all). Would you have the car doing RT work for $115 a day for a full 300 mile trip. 200 days a year would net you 60k miles and $23,000. See Uber drivers are out there doing 300 miles and making 100-160 a day. You'd spend 200 hours a year on it as well keeping it clean, etc. In 60k miles you will probably have an exception too, vomit, someone badly sick, kids spill chocolate milk, dog pisses on it, gets a bad scratch in the Durham Bulls parking lot. That will consume a day or more. I don't see the profits.

Why wouldn't someone with a Nissan Leaf do Uber ? Taxi's don't have to be ICE. So anyway, I sliced and diced, I dug into the user market quite a bit. I don't see the profits. I actually expected far better responses from you; you are detailed oriented. Go create a real model, dig into the Uber data more. Find out how many cents/mile you could cut costs. Find the profits. Then talk to some uber drivers and find out how much time they spend on the car, how many miles per day and the take per day. Understand that nothing keeps an EV driver from being an Uber driver (other than intelligence). I'd be very interested in seeing your take on things if you dig into it, very happy to be proven wrong as I see a glaring need for transportation services in small towns.

My belief is that Uber drivers are essentially subsidizing Uber. Eliminating the driver is not a route to profitability.

Finally the "when I need it I will summon it home". If you need it and it is 30 miles away in a traffic jam with a passenger you will not be summoning it home. My wife was stuck 2 1/2 hours yesterday in a commute. If your kids school had called at 9am to say pickup your child you'd have been waiting for that car to make it back.

 

[Knightshade]

Ok, obviously you don't drive for Uber.

Of course not- it's a terrible job.

Go talk to some Uber drivers, hurry because the business is lousy. They have to be cleaned. Every day, every shift. Then washed as well

Which those bots I mention could easily do while they're supercharging. Or the owner can while it's L2 charging if they brought it home.

I also don't think individuals will do RT instead it will be fleets.

I mean, Tesla has explicitly and repeatedly mentioned individuals will be able to do this for many years so I'm not sure why you think that.

Certainly economies of scale will make "fleets" more efficient in large cities... but in the burbs and rural areas nobody would make enough to run a big fleet--- while individual owners could make a nice supplemental income during the most-hours-of-the-day the car is otherwise just sitting parked.

I assume that no one in their right mind will let the car be a RT for long

This assumption has no basis in fact or reality though.

See the tons of folks who ordered cybertrucks with FSD explicitly to use as future RTs for example. See also Turo.

Why wouldn't someone with a Nissan Leaf do Uber ?

Because most Leafs ever sold have less than 100 miles of range and no legit fast charging at all just Chademo as max 50kw.

I actually expected far better responses from you; you are detailed oriented.

I mean, I pointed out if you pocket all the $ that goes to the driver Uber would be hugely profitable. Still profitable even AFTER all the costs you cited which are ALSO called out in the source I provided. Not sure how much better than actual numbers showing billions in profits you were expecting.

4. Compensation /net profit – – – $8.55–$10.00

That's net to the driver after all costs and expenses

Per hour.

And reminder they cite 422.147 million hours a year.

That's ~4.22 billion dollars a year at $10/hr. Meaning Uber would have billions in profit a year if they got to keep that $ instead of paying it to drivers- and that's INCLUDING paying for insurance, gas, maintenance, etc.... on a fleet that's almost all ICE vehicles....profits would be significantly MORE with an all-EV fleet with good range and low maintenance.

Finally the "when I need it I will summon it home". If you need it and it is 30 miles away in a traffic jam with a passenger you will not be summoning it home.

Unless I had NO IDEA I'd need my car until the very last second- sure I would.

My wife was stuck 2 1/2 hours yesterday in a commute. If your kids school had called at 9am to say pickup your child you'd have been waiting for that car to make it back.

Or I'd just summon a different robotaxi to the school

 

[Daniel in SD]

I still think that Tesla has a per mile operating costs (Tires, Juice, et) somewhere in the $0.15 to $0.25/mile. That leaves very little room to do all the things that are a service business.

You can't leave out the $13.50 per ride fee and the $18 per hour fee! If your robotaxi does 10 twenty minute fares a day (only 14% utilization) that's $195 a day ($71k a year). The per mile fee isn't even the majority of revenue unless you're doing long highway fares.

 

[nativewolf]

Of course not- it's a terrible job.




Which those bots I mention could easily do while they're supercharging. Or the owner can while it's L2 charging if they brought it home.



I mean, Tesla has explicitly and repeatedly mentioned individuals will be able to do this for many years so I'm not sure why you think that.

Certainly economies of scale will make "fleets" more efficient in large cities... but in the burbs and rural areas nobody would make enough to run a big fleet--- while individual owners could make a nice supplemental income during the most-hours-of-the-day the car is otherwise just sitting parked.




This assumption has no basis in fact or reality though.

See the tons of folks who ordered cybertrucks with FSD explicitly to use as future RTs for example. See also Turo.




Because most Leafs ever sold have less than 100 miles of range and no legit fast charging at all just Chademo as max 50kw.





I mean, I pointed out if you pocket all the $ that goes to the driver Uber would be hugely profitable. Still profitable even AFTER all the costs you cited which are ALSO called out in the source I provided. Not sure how much better than actual numbers showing billions in profits you were expecting.



That's net to the driver after all costs and expenses

Per hour.

And reminder they cite 422.147 million hours a year.

That's ~4.22 billion dollars a year at $10/hr. Meaning Uber would have billions in profit a year if they got to keep that $ instead of paying it to drivers- and that's INCLUDING paying for insurance, gas, maintenance, etc.... on a fleet that's almost all ICE vehicles....profits would be significantly MORE with an all-EV fleet with good range and low maintenance.






Unless I had NO IDEA I'd need my car until the very last second- sure I would.






Or I'd just summon a different robotaxi to the school

There are not billions in labor savings, that's my point. There are no labor savings because the drivers are subsidizing Uber. Drivers are paid far less than the IRS allowance on a per mile basis. Far less. The average Uber driver is going in the hole every mile they drive. Every mile they drive they write Uber a check. They are working for FREE. Free, the payments they receive don't cover the costs of miles driven.


Another perspective: knowing that an Uber driver is willing to drive their car for $0.37 /mile or less and that the free market says that's your rate for providing your car as a RT. Would you let your Tesla out in the wild for that amount? By that I mean sign it out and have it in a position that not blocked in on your street or driveway. A car you've spent $60k plus X on FSD. Knowing that you have a real cost (depreciation, tires, wipers, cleaning time, electricity, cleaning suppllies, etc). You get home, make sure the cars not blocked in by the kids bike, make sure it's clean, sign it into the RT dispatch service, and you get one dispatch at 1130, the car comes home 30 mins later, after coffee you go out to check and greeted with a cig burn on your drivers seat and the odor of smoking before you leave for work. Knowing you made a grand total of $3.37 on the ten mile trip and you're going to spend 30 mins deodorizing the car? You'd do that?

In the burbs and small towns the cars are idle when nobody needs one. Everybody needs the car to go to and from work, get the kids, get the groceries, etc, they get home from work and very few people go out to eat and drink. Free market should dictate that you make the market rate ...a rate that sucks. $.37/mile or less (unless you live in a huge metro region). Anyway, if Uber could never successfully penetrate the burbs (and they have not) I don't believe RT will either. Because there is no density.

RE robotic car cleaning services. The human hand and wrist will be the last robotics issue solved, the technology does not exist yet and it is not just a software issue. Cars are not easy to clean, maybe Tesla simplifies RT seating etc but I am not qualified to guess on that. Just as real FSD (the type originally promised as just around the corner 6 years ago) has been a slog so to will real FSD. On this I think we agree. I am more skeptical on anthropomorphic robots any time soon than FSD.

Again, if Uber drivers made even a working wage I'd say maybe an opportunity but with the exception of larger very metro regions they don't. Don't get me wrong, I'd like to see RT work as it could be liberating for millions of people. I just don't see the profit and so far you're just regurgitating all the mess other people say. Peel back the onion that is Uber. Go find the profit that is eliminating labor. In fact, the one thing that is making profit is food delivery and black label cars...wait for it...labor is what makes the highest margins for Uber.

 

[Daniel in SD]

$.37/mile

Do we agree that an average Uber trip costs the rider about $2 a mile?

 

[Knightshade]

There are not billions in labor savings, that's my point.

Except your point is factually wrong, and I provided a source proving that.

There are no labor savings because the drivers are subsidizing Uber.

This is nonsensical.

It suggests drivers are paying Uber to work for them.

There is no math whatsoever supporting this claim.

And it gets even more nonsensical when applied to a system with no need for paid humans, gasoline, or normal ICE maintenance.


You keep conflating "paid a relatively low wage net of expenses" with "paid a negative wage" for...no apparent reason.

RE robotic car cleaning services.

You know they already have automated car washes right?

And have for decades?

The human hand and wrist will be the last robotics issue solved, the technology does not exist yet and it is not just a software issue.

Except it does exist, and has for years.

That's a lab demo of a robotic hand from 2016.

That's from a couple years ago of a robot in china that picks up trash from the street similar to what you might commonly clean from a taxi (drink bottles for example).

Cars are not easy to clean,

I mean- way easier than you seem to think....

Again, if Uber drivers made even a working wage

Again, after all expenses they make above minimum wage

As cited to you in actual sources.

4. Compensation /net profit – – – $8.55–$10.00 per hour

That's after all costs and expenses

Which is 100% profit in a robotaxi.

( as is all the $ that goes into costs/expenses that are non-existent or at least much lower in an EV)