Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[sunwarriors]

I was wondering why everyone thinks Elon is betting the company on robotaxi/autonomy all of a sudden, so I:
- read the last 25 pages of this thread (consisting of mostly panicky "WTF is Elon doing?" versus a minority of "CHAAARGE" approval-posts)
- searched the interwebs for the source of these rumours.

Elon himself said it (all over the news if you searched):

Elon Musk says Tesla is ‘worth basically zero’ if it cannot make self-driving cars

Tesla has about 830,000 vehicles on the road, but the NHTSA has said they have got into nearly 300 ‘self-driving’ crashes - more than other manufacturer

www.independent.co.uk

Elon Musk is putting Tesla all-in on Robotaxi

Riding the AI wave, Elon Musk is betting Tesla’s business on its self-driving effort, including Robotaxi, at the cost of...

electrek.co

 

[Usain]

First, OEDR complete is more than just "avoid objects on the road"

It's also "know how to react to a stopped school bus, including in context of the sign being out or not, and if the road is divided or not."

It's also "understanding any non-standard speed sign"

It's also "understanding parking signs"

and a slew of other things.

Given enough time and training, FSD would learn to properly deal with things like school busses. But I tend to agree with you that these are the kind of things that Tesla would want to do specific targeted training for.

The approach that I've been advocating is that you can start out by limiting the domain. For instance, in the point to point example of going from Macon to the ATL airport, it's easy to avoid school bus routes because you are just going from exit to exit off the interstate.

 

[Electroman]

More Elon tweets incoming!!

Why..?

 

[philw1776]

Tesla without autonomy is not worth a whole lot more than today’s price.

I pretty much agree. Musk has said as much though it's his typical hyperbole.
I'm in long for the AI & robotics stuff.
Robo taxi is many years off based on my FSD experiences. Tougher than Optimus factory, etc. implementation because of an unconstrained huge environment vs a factory task.

 

[Buckminster]

https://twitter.com/x/status/1780306773431238902

 

[Zero CO2]

I've thought about this notion a great deal over the last week.

Cancelling the $25K car to focus on RT could possibly be a brilliant move in hindsight years from now. My gut feeling is it would be a terrible choice, but seeing how competitive the auto and EV market has become, and seeing how efficient Chinese brands are getting at making affordable EV's, maybe Tesla is making a choice to give up on growing its auto market to instead focus on the new market no one else has the ability to compete in: TaaS (Transporation as a Service).

FSD when solved would be a huge tech breakthrough, one which no one else has nor can easily follow. It would enable Tesla to create a brand new market practically all to themselves with RT's.

Foregoing the super competitive auto market to focus and grow on a Tesla only RT market...well, its not a concept without a crazy degree of sound reasoning.

this is why i invested in Tesla ... i would never invest in a "car " company right now they are both Apple and Google (smart phone OS analogy) stop competing with all the low cost HW vendors and pivot to Energy and Software

2018 (January) Model 3 RWD LR in California here.
What's your hardware ?

Typical mistakes it makes for me is moving over into right turn lane when planning to go straight. Couldn't try if it would actually turn as there were other cars I couldn't let it cut off.

Map showing only one of two stop signs, signaling right on first stop sign with intent to go straight and turn right at the next one.

Also cutting over opposite traffic lane when turning left, rather than staying clearly in its lane while turning. I had to take over to ensure no colission today. Maybe it wouldn't have collided but it sure did freak me and probably the other driver out.

My wife in 2020 model x (also still intel based mcu) also reported having to take over at stop sign where it was trying to take other cars right of way.

I also had interventionless 20 mile drives door to door on other routes. Most of the time very impressive and capable but regressed compared to 12.3.3 for me.

late 2018 model S HW3... i think .. never got 12.3.3 12.x was delayed for older Model S .. maybe i have not driven enough yet.. but wow it is a lot better than 11.x ... it is getting real
and for the Robotaxi folks .. this is a great feature right now

 

[SOULPEDL]

I agree that FSD needs to become more strategic in its behavior.

Whoa, key point there. In fact it caused me to think of how the car could safely pass a farm tractor for example (even though the front cameras are in the center of the car requiring a wider lane offset).

If the car just pulls further back behind the car to be passed, it gives extra time to peek. In fact, if there's room, that's always safer. But here it can compensate for camera positioning. Nice example of strategic behaviors.

Maybe too much FSD in this thread for one day... ya before anyone says it.
Cheers!

 

[DarkandStormy]

TL;DR: take a chill pill everyone. As long as Tesla isn't losing money on every vehicle sold, they can outlast the macro storm (which, if you hadn't noticed, is still raging). If Tesla solves autonomy in the meantime, big bonus.

QQQ is up 36% in the last year and SPY is up 23%. What kind of storm are you experiencing?

 

[Buckminster]

https://twitter.com/x/status/1780297314000871518

 

[FEERSUMENDJIN]

Payload has a huge impact, try it and see, go offroad with half a ton in back. Towing, larger yet. Your comment.."strategies to provide electrical charging to an advancing force" is about all one needs to know. It does not exist. You don't send light mobile units into the field unless you know they can stay mobile. ICE can pretty much fuel from anything, a mrap or the equivalent. If sent to middle east do you think there are already tens of thousands of petrol solutions available? Pipelines, tanker trucks, gas stations, steel drum with 40 gallons of diesel. Maybe one day when electrification is largely built out this would be viable. Now..not so much.

CT is neat, a military vehicle it is not

Supervised FSD as it is now? Well it statistically has not caused a death (sustains life) and a year ago it crashed 5x less than a human (sustains vehicles). And for those of us who consistently drive 10 hours or more some weekends, it sustains sanity and mental energy. Tesla publishes astonishing FSD safety statistics

Eventually, within 3 years to be sure, some would say by next year, FSD unsupervised will allow each RT to replace the utility of approx 15 cars (average car is used for 1.4 hrs a day vs 24/7 RT). This will drastically reduce the number of ICE vehicles on the road.

Also Tesla has elaborated the sustainability aspect of Robotaxis several times and with the huge reduction in passenger vehicles overall (Privately used poorly utilised) along with the requirements for car parking, this will result in the urban footprint in US cities being transformed.

 

[Chunky Jr.]

As you said, the Chinese are very good at making inexpensive EV's. What if...WHAT IF...there has been some recent deal between Tesla and a Chinese company to do some badge swapping or shared engineering of some sort to better enable the net gen $25K car from Tesla? Something like that could take many forms...but I'm imagining something that enables Chinese-like production in North America. Maybe a deal like this resulted in a drastic change in how the $25K model will be designed and produced, resulting in scrapping the "old" plan. This is 100% wild idea from me...and there are many reasons why it is crazy...but a sudden deal like this could explain the apparently sudden shift in direction over the past few months.

As @Right_Said_Fred mentioned, this scenario seems more likely with batteries. Tesla and CATL teaming up and Tesla abandoning their home grown 4680 plans

 

[Wooloomooloo]

The question of what if FSD takes another 5 years and what will the stock do in those five years is the main one I have right now.

5 years ago, anyone who said FSD was 5 years away was laughed at for their pessimism, sometimes they even got temporary bans. I said at the time (and you can find the posts) that it would be 10 years - I think I got about 230 downvotes for that. And yet here we are.

Tesla is absolutely not worth zero without FSD, that's just typical Musk hyperbole. The company is profitable, still growing (1 quarter a trend does not make) and there's a freaking huge market out there to be had.

The whole economics of "robo-taxi" really needs to be looked at. Who does he think is going to own and profit from robotaxis. This notion that both owners as well as the company and shareholders will profit from this has never made sense to me. It's double counting the potential income - everyone will want a Tesla which will be an appreciating asset and make money. Sure... something about that sounds like a perpetual motion engine, or a ponzi scheme.

I sold in Oct '22 @$350 and have never looked back (I bought in Jan 2019 and Jan 2020) It was never going to stay a $1t company, let alone $3t or $10t some here said would happen "within 10 years" but it could still be a great car company.

 

[Usain]

new job listings being purged. I'm sure there will be new ones in near future, but existing ones are no longer available. This is far more than just removing the bottom performers. Seems more like a total reset.

https://twitter.com/x/status/1780303329370816755

I'm betting that Elon said to take all the job postings down so he can personally go through and decide which ones he still wants to hire for.

The still-active postings for the US are for intern or training programs. Not sure why those got to stay, but at least one is a partnership with a school.

 

[M|S|M MYP]

I'm betting that Elon said to take all the job postings down so he can personally go through and decide which ones he still wants to hire for.

The still-active postings for the US are for intern or training programs. Not sure why those got to stay, but at least one is a partnership with a school.

Darn, and here i was hoping T.I.T.S was going to be up and running by the time my son applied for college!

 

[FEERSUMENDJIN]

So Tesla just starts in a state that doesn't use the silly document. Problem solved.

And as I've pointed out before, if Tesla can perform as well as Waymo then it will be in compliance anyway. You have yet to tell me what Waymo does that Tesla FSD could not easily be trained to do.

All of those acronyms that Knightshade likes to chuck around are all things that can be implemented, no problem. The ODD can be limited for level 4. I would say that the OEDR is probably better than Waymo, and that the DDT fallback is a concept that is a relic of heuristics, and can be considered an intrinsic function of the AI planning module (resilient adaptation). As you say the proof is in the observed behaviour and performance.

 

[DarkandStormy]

https://twitter.com/x/status/1780306773431238902

Hmmm, so it's almost exactly like I said it was the past two days? And not your usual "fat trimming?"

Carry on.

 

[GOVA]

new job listings being purged. I'm sure there will be new ones in near future, but existing ones are no longer available. This is far more than just removing the bottom performers. Seems more like a total reset.

https://twitter.com/x/status/1780303329370816755

What if the Optimus is ready to take this over?

 

[jeewee3000]

Elon himself said it (all over the news if you searched):

Elon Musk says Tesla is ‘worth basically zero’ if it cannot make self-driving cars

Tesla has about 830,000 vehicles on the road, but the NHTSA has said they have got into nearly 300 ‘self-driving’ crashes - more than other manufacturer

www.independent.co.uk

Elon Musk is putting Tesla all-in on Robotaxi

Riding the AI wave, Elon Musk is betting Tesla’s business on its self-driving effort, including Robotaxi, at the cost of...

electrek.co

I said "all of a sudden". In other words, what has changed in the last few weeks.

You bring up an Elon quote from 2022. I know what he said. But in the last week he has not confirmed any of the things being thrown around.

 

[cusetownusa]

Also Tesla has elaborated the sustainability aspect of Robotaxis several times and with the huge reduction in passenger vehicles overall (Privately used poorly utilised) along with the requirements for car parking, this will result in the urban footprint in US cities being transformed.

I would like a better explanation how there would be a huge reduction of passenger vehicles. Why would a huge amount of people give up their cars? Also, if that did happen, there would still need to be as many cars to move people during rush hour, where are these cars going during the day? They would need to park somewhere.

 

[heltok]

As @Right_Said_Fred mentioned, this scenario seems more likely with batteries. Tesla and CATL teaming up and Tesla abandoning their home grown 4680 plans

4680 is NMC for the high capacity batteries that they need for Semi, Cybertruck and future products such as Optimus, eVTOL etc. CATL are doing LFP and Sodium for stationary storage and mass market models such as 3/Y/25k etc.

Tesla has realized that they need more LFP made in the US due to IRA and potential future sanctions. Imo there is enough LFP made in China for their needs now and enough new capacity is coming online.

Speaking of China, they had a rough period but it seems things are looking a bit stronger right now:

China’s economic growth hits 5.3%

Annual figure for first quarter beats expectations as Beijing seeks manufacturing-led revival

www.ft.com