Poll: 81% of Prospective Model 3 Owners Say They Won’t Pay Upfront For Full Self-Driving

[vc_row][vc_column][vc_column_text]It seems most prospective Model 3 owners aren’t willing to shell out cash upfront for a $3,000 “full self-driving capability” option that is likely years away from becoming available to engage.

In a poll posted by jsraw 81.3% (347) of respondents said they will not pay for the feature at purchase. Adding the option later will cost an additional $1,000. Of respondents, 18.7% said they will pay for FSD upfront.

According to Tesla’s website, FSD “doubles the number of active cameras from four to eight, enabling full self-driving in almost all circumstances, at what we believe will be a probability of safety at least twice as good as the average human driver. The system is designed to be able to conduct short and long distance trips with no action required by the person in the driver’s seat. For Superchargers that have automatic charge connection enabled, you will not even need to plug in your vehicle.”

Elon Musk has said that level 5 autonomous driving is possible with second generation Autopilot and the FSD option, meaning the car is fully autonomous in any and all conditions. During his TED talk in April, Musk said the company plans to conduct by the end of 2017 a coast-to-coast demo drive from California to New York without the driver touching the wheel.

Obviously, there will be regulatory hurdles ahead and Musk has said it will likely be two years before owners will be able to engage FSD capability.

See a few comments on the poll below, or go to the thread here.

Swift

EinSV

jason1466

Waiting4M3

Enginerd[/vc_column_text][/vc_column][/vc_row]

 

[NJhokie98]

I hope that there will never be a time where you HAVE to have a fully autonomous cards on the road. I like teslas but i like to drive them not have them drive me.

 

[insaneoctane]

I don't believe that FSD will get more expensive in the future. Buying it afterwards is already going to add an additional $1k. Plus you have to buy the EAP before you can get FSD. I think with more people buying Teslas and buying EAP, there's a chance that in the future they could even reduce the cost. I think creating a new technology is much more expensive than maintaining and updating it.

I see Tesla maximizing profit. Even if they recouped all the development costs, I don't think that they will lower the price, rather they would increase margins. If FSD works, it will be more than $3K in the future. The only thing that would drive the price point down is competition.

 

[chronopc]

I see Tesla maximizing profit. Even if they recouped all the development costs, I don't think that they will lower the price, rather they would increase margins. If FSD works, it will be more than $3K in the future. The only thing that would drive the price point down is competition.

You typically get more profit increasing volume than margin. If they wanted to maximize margin a better approach would be to not install it on every car. The fact that they installed it on every car makes me think that they will charge more now and whoever can afford it will buy it and once they recoup their development costs, they'll reduce it to get more people to opt in and take advantage of economy of scales. The way I see it is all those people driving around with the hardware but didn't get the EAP or FSD is sunk cost for Tesla. Making it cheaper to unlock EAP or FSD will not cost Tesla anything at that point and will definitely entice many of the people who didn't originally get it to consider it.

 

[siggyfreud]

Tesla might also be banking on collecting road data from all those cars with the hardware but unenabled software. Might be a way to recoup some of those costs.

 

[insaneoctane]

You typically get more profit increasing volume than margin. If they wanted to maximize margin a better approach would be to not install it on every car. The fact that they installed it on every car makes me think that they will charge more now and whoever can afford it will buy it and once they recoup their development costs, they'll reduce it to get more people to opt in and take advantage of economy of scales. The way I see it is all those people driving around with the hardware but didn't get the EAP or FSD is sunk cost for Tesla. Making it cheaper to unlock EAP or FSD will not cost Tesla anything at that point and will definitely entice many of the people who didn't originally get it to consider it.

I think that they install it on every car to enhance the collective learning of the fleet. As far as profit vs volume, look at what they have chosen to do with the LR. They took a 25kW upgrade and chose to sell it at about 60% profit margin (maybe more). The economies of scale of the gigafactory would allow them to sell it too many more people. But, who else offers a 300+ mile range EV? No competition, so we see very high margins.

 

[verygreen]

Tesla might also be banking on collecting road data from all those cars with the hardware but unenabled software. Might be a way to recoup some of those costs.

I think this is still greatly overhyped.
What sort of road data do you honestly think they can collect that would comprehensively let them stop using one of the data feeds from 3rd parties?

 

[verygreen]

I think that they install it on every car to enhance the collective learning of the fleet

The learning hype is still running strong. But what do you think it means?

 

[chronopc]

I think that they install it on every car to enhance the collective learning of the fleet. As far as profit vs volume, look at what they have chosen to do with the LR. They took a 25kW upgrade and chose to sell it at about 60% profit margin (maybe more). The economies of scale of the gigafactory would allow them to sell it too many more people. But, who else offers a 300+ mile range EV? No competition, so we see very high margins.

If you compare how much you have to pay to get an extra 25kw on a Model S ($23k) vs Model 3 ($9k) you can see they do decrease it based on economy of scales. Also with battery it's a hardware component, so for each additional one they sell there is a cost. However with software each additional one you sale doesn't have any additional costs. So for software the more you sell the cheaper you can sell it for. Think about games that are sold on Steam, they still make a profit even selling for prices as low as $5 because there isn't any cost associated with selling more digital copies.

 

[Daniellane]

Purchased FSD on my Model S 90D that I took delivery of in Oregon on June 23rd. I did have to pay Washington Sales Tax. Far less than the $1,000 difference and since I financed it @ 1.49%, seemed like no big deal. My rational was that if the hardware was not adequate to do FSD, Tesla might be more inclined and legally bound to upgrade it at no cost in the future.
I have a Model 3 reserved for my wife. Will probably do the same again.

 

[FlyingKiwi]

Another factor could be that even if you haven't purchased FSD Tesla might want your car on Tesla network. When you are driving your own car FSD is not active if you haven't paid, but once you exit the car and release it to the Tesla network FSD enables and your car goes off to earn it's keep.

 

[lunitiks]

The learning hype is still running strong. But what do you think it means?

The question is what you think it means

 

[malcolm]

Before a coast-to-coast AP demo it's pretty much a crap shoot.

After a coast-to-coast AP demo it's pretty much a turkey shoot.

 

[siggyfreud]

I think this is still greatly overhyped.
What sort of road data do you honestly think they can collect that would comprehensively let them stop using one of the data feeds from 3rd parties?

Anomalies, curb heights, construction zones, missing lines, unrecognizable lights, and lots of other nuanced data that may not be provided by said feeds. If they're going for true FSD there are so many things they'd want to factor in. Slope of entry/exit points, one way tire spikes, gated entrances . . .

 

[jsrawa]

View attachment 242589
Purchased FSD on my Model S 90D that I took delivery of in Oregon on June 23rd. I did have to pay Washington Sales Tax. Far less than the $1,000 difference and since I financed it @ 1.49%, seemed like no big deal. My rational was that if the hardware was not adequate to do FSD, Tesla might be more inclined and legally bound to upgrade it at no cost in the future.
I have a Model 3 reserved for my wife. Will probably do the same again.

I thought you still have to pay sales tax even if you buy it in the car later. Can someone with some experience confirm? If so you are really only paying the finance 1.49% as a "penalty" but some states have an ownership tax so you may be taxed some more on that amount.

 

[JeffK]

What is extremely valuable is driver logs + sensor data through the current software whether that's feature detection, inference, etc. (it can sometimes be even more valuable than the raw data especially due to the amount of data.) It would be trivial for it to have sign, lane, pedestrian, unknown object detection. Simply report back what was detected, the position relative to the vehicle, and inputs from the driver, etc.

What's important is what the driver did vs what the car "thought" it should do in any given scenario and what was the outcome positive or negative. In order to do this on a mass scale, you need the hardware in every car as individual drivers probably don't vary their routes that much.

 

[verygreen]

The question is what you think it means

I think it's mostly hype.

Anomalies, curb heights, construction zones, missing lines, unrecognizable lights, and lots of other nuanced data that may not be provided by said feeds. If they're going for true FSD there are so many things they'd want to factor in. Slope of entry/exit points, one way tire spikes, gated entrances . . .

Anomalies like what? How would they be reported? Same for the lights and other "nuanced data". How would the car even know if it misdetetcts something?
Construction zones, missing lanes - you need lots of cars on the road everywhere for this, otherwise it'll only work reliably in California and everywhere else you'd still need to pay for a subscription to the feed (and it's unlikely they'd let you remove just California data for a discount, I think).

Recognizing spikes is going to be challenging and I am not sure what's the value in gated entrances. Also some gated entrances are open all the time until they are closed at the worst possible moment

What is extremely valuable is driver logs + sensor data through the current software whether that's feature detection, inference, etc. (it can sometimes be even more valuable than the raw data especially due to the amount of data.) It would be trivial for it to have sign, lane, pedestrian, unknown object detection. Simply report back what was detected, the position relative to the vehicle, and inputs from the driver, etc.

pedestrian/potholes/debris are typically short-lived so only matters if you have a lot of Teslas in vicinity so they can take advantage of it right this moment. (if you do remember such events on the other hand that's going to be a separate can of worms in the form of "why is my car always slow at this spot?")
Also I suspect debris are going to be a source of frustration since a flailing around plastic bag is going to be hard to tell from something more dangerous, I suspect.
Remember the car is not a human so there's no easy way to determine why did the driver take whatever action, unless you send that to the mothership right away and have people waiting to analyze it eagerly so there's no delay (= lots of humanpower).

What's important is what the driver did vs what the car "thought" it should do in any given scenario and what was the outcome positive or negative. In order to do this on a mass scale, you need the hardware in every car as individual drivers probably don't vary their routes that much.

I suspect lots of people don't properly disengage AP before doing some maneuver (I certainly don't a lot of times) that disables the AP anyway so sifting through this data in search of gems would be hard. If you don't even depend on eap/tacc activated then there's going to be even bigger crapshoot. "oh no, there's a guy in Mustang overtaking me, cannot allow that, full on forward!"
"When you approach a yellow light - accelerate", ...

So there's the eclipse tomorrow and at about 2:20pm all the drivers would likely greatly slow down and partake in dangerous maneuvers around here. If all those were Teslas with this learning thing, I wonder what would the fleet learn from this?

So in short - you could learn a lot from the data, but you need to employ a lot of humans to look at all this data all the time to actually infer new knowledge, NNs can just help you filter some of the data, but the bigger the filtering likely the more useful data is also lost.

 

[lunitiks]

What a fantastic reply

Btw almost every single time I disengage autosteer I get that question in my head: "Did I just add to the Autopilot "learning"? Wooops, that wasn't my intention. I just wanted to take over"

 

[chronopc]

So in short - you could learn a lot from the data, but you need to employ a lot of humans to look at all this data all the time to actually infer new knowledge, NNs can just help you filter some of the data, but the bigger the filtering likely the more useful data is also lost.

There's a whole industry of people making money from data mining. It's really lucrative.

 

[verygreen]

There's a whole industry of people making money from data mining. It's really lucrative.

Yes, I know about that. It does tend to highlight general trends and such. For example "if the light switched to yellow - accelerate" or "if there's a mustang in the adjacent lane stopped on a traffic light - launch", "drive at detects speed limit +10mph", "turn on a blinker and a window opens for you to merge in some cases", general stuff like this.
I am less sure you can infer transitive minor details on the roads via that, though. Yes, there's a piece of a blown up tire on 3rd lane on I-5 today at 2pm at some random mile marker, but there's just no big picture to infer form that, I suspect.

Quick Edit: think of waze and similar apps, if there was a way to use the video feed to infer a lot of this data, why it's not part of them yet? Google has huge resources to analyze it yet we don't see anything of the sort.
It would have beat the human reporting on consistency if everything was as you imagine, right?

 

[chronopc]

Yes, I know about that. It does tend to highlight general trends and such. For example "if the light switched to yellow - accelerate" or "if there's a mustang in the adjacent lane stopped on a traffic light - launch", "drive at detects speed limit +10mph", "turn on a blinker and a window opens for you to merge in some cases", general stuff like this.
I am less sure you can infer transitive minor details on the roads via that, though. Yes, there's a piece of a blown up tire on 3rd lane on I-5 today at 2pm at some random mile marker, but there's just no big picture to infer form that, I suspect.

Quick Edit: think of waze and similar apps, if there was a way to use the video feed to infer a lot of this data, why it's not part of them yet? Google has huge resources to analyze it yet we don't see anything of the sort.
It would have beat the human reporting on consistency if everything was as you imagine, right?

I'm positive Google is using the information they are collecting from Waze and Google Maps in their testing. Mobileye joined up with them recently to do testing in Arizona. I'm not sure how confident Google is in their technology but they have been testing for a very long time. Since they are not a car company they have no reason to rush it to market. I think eventually they will just license their technology to other car companies who don't want to spend on R&D for self driving technology.

Robocars Flock to Arizona, Land of Good Weather and No Rules

I would say the data that Tesla gets is much more useful than the data from apps since a Tesla vehicle gets a lot of information from the multiple cameras.The challenge for them is to teach the computer how to know which information is important for which situation. The human mind is very good at collecting a lot of information and only focusing on what is important for the decision that needs to be made at the particular time.