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

[jabloomf1230]

[

fair enough. We'll see in 2 years where they are and compare to Tesla. I suspect we'll see Waymo beginning to deploy the fleet and Tesla to start going through permitting in CA. We have no idea where Waymos cost will be at scale nor what the market will hold if Tesla and Waymo go to a price war in 3-4 years.

It's a battle of two philosophies. Tesla has tested FSD everywhere in NA with their own production vehicles. Waymo has retrofitted someone else's production vehicles and tested them within small geofenced areas. Both approaches may end up being profitable. But for now they are both technological curiosities. They are competitors and each company's profitability will be impacted by the other's.

 

[aubreymcfato]

"New"? I guess...

https://twitter.com/x/status/1799405994843222249

They voted no also last time, fwiw.

 

[GOVA]

I hear the complaints about what Elon has said in the past about FSD progress and I don't have the same sentiment. Why? I don't look to his X posts for progress, I look to my own experience on FSD and see that we are progressing quickly.

I have this edge case in my town. A roundabout where 2 lanes have a "right of way" and 3 lanes have a yield. For local townspeople, we maneuver this with ease. But anyone new to this roundabout yields everywhere because it is confusing. FSD V11 and prior would get confused here and either disengage or yield when not necessary. FSD v12.3.6 has handled this roundabout 6 times now consecutively with no intervention or disengagement.

Robotaxi does not need to arrive tomorrow. It needs to arrive first with the competition a few years behind . . .and I think this is what we will see.

Green has right of way. Yellow needs to yield.
View attachment 1054735

Roundabouts, IMO, are the greatest traffic inventions ever. Normally, those are quite simple which make them very effective. The example you posted does modify the basic rules of a runabout. This is bad and should be avoided at all cost.

Why would you invite a wheel and use a square to move forward?

The fact you mentioned that FSD has handled this well, only indicates its high degree of development.

 

[mrmage]

Googles last product reveal was gemini AI. how did that go? And sure, google can do software...maybe. Maybe not. Thats why they had to buy deepmind to get anywhere in AI.
But hardware? Show me the large scale global physical hardware rollout by google? What google hardware do you own? Mass production of physical products is WAY harder than rolling out software, and definitely requires different skills.

I had some pixel phones, google assistant, Nest products, cameras, etc. 40 million phones sounds like mass production. And a sensor array is basically electronics + cpu + i/o, and certainly not rocket science.

And again... you cannot compare a few hundred (or even thousand) waymo vehicles with LIDAR to Tesla. Tesla order cameras for their cars by the million. How many LIDAR units do you think waymo ordered in the last year? You think waymo can get LIDAR cheaper than a camera? (They need cameras as well)

I never compared camera costs to LIDAR, so not sure why you brought it up. I said the cost of LIDAR is not a significant factor because its amortized over the life of the car. They're only $1k now ($4k for 4), and only getting cheaper in the future.

In August 2023 waymo told regulators they had 250 cars, 100 of which were in use at any one time.
Yeah... really not holding my breath for global robotaxi domination by that company...

In June 2024, Tesla had 0 robotaxies...

Anyway, if you're bagging on LIDAR because Elon said it was too expensive, that only applies to Tesla. Tesla is selling cars to consumers, so adding $4000 of LIDAR sensors would destroy sales and margins.

Waymo has a different business model and can absorb the cost. They only make dedicated robotaxies, the sensors will always be used for generating revenue, so they can amortize the cost over the revenue generating life of the vehicle.

 

[Usain]

You're looking at this wrong from a "how does it scale" perspective. If you want to know the true ratio of how it will operate at scale, it's more logical to check the number of interventions, assign a time cost per intervention (some will be a click on a computer, others would be physical recoveries), factor in a safety margin and that would be your time cost for the fleet. Ballpark, given the number of interventions and miles driven, it's much much much lower than an Uber driver cost.

By your logic, Tesla is in a much worse position because it can't move a meter without driver supervision, which is a ratio that's orders of magnitude worse than Waymo's today.

The basic problem is that we don't actually know the number of interventions or the cost for each. What I see in Waymo videos is that it is very common to get the "We are working to get you moving again" message. I believe that each one of those is a human intervention. And it's an unhappy customer who is waiting in a car that refuses to go forward for no apparent reason. Relief drivers are needed less frequently, but I see a lot of those as well.

So while I don't know for sure, my impression is that there is a lot of behind the scenes human monitoring and intervention with each and every Waymo trip.

You said by my logic Tesla is in a worse postion. But Tesla is not in a worse position because Tesla hasn't started yet. Waymo is losing millions on its robotaxi network and Tesla is losing zero.

Once Tesla starts its service, it won't take long for the performance of Tesla's system to surpass Waymo. Tesla will win is because end-to-end is a better approach and Waymo will not be able to catch up. Waymo will be behind in both hardware cost and core driving technology.

It won't be a fair fight.

 

[UkNorthampton]

These are my sentiments today although i did vote in favor of the original plan and ratification.
Events of the last few days have established the consequences of Mr. Musk's reactive postures regarding Tesla issues. Pre-pandemic he seemed consistently acting in the best interests of shareholders including his own long term prospects. During and Post- Covid-19 it seems he might have, at best, "long Covid".

The prospects of Tesla remain stellar if stable relentless innovation continues. Whether that will be the prognosis (assuming all this is fundamentally diseased) is not at all clear.

200+ years ago, Wellington won a pivotal battle/war with a ragtag bunch of units from different countries and speaking multiple languages, some of whom until recently were loyal to Napoleon.

People matter, leadership matters, planning for the future matters.

Lots of examples from history. Marlborough, Rommel.

Key people, ability to attract the best talent, capital efficiency and culture matter to the success of Tesla.

Short term, new leadership might switch to one for one ICE-replacing cars at higher profits ditching longer term ambitious development.

That might lead to higher share price, more institutional buying and less controversy.

Longer term, Tesla won't hit its potential and we will risk bad actors controlling the future including robots and AI.

 

[cliff harris]

I had some pixel phones, google assistant, Nest products, cameras, etc. 40 million phones sounds like mass production. And a sensor array is basically electronics + cpu + i/o, and certainly not rocket science.

All of which are phones, not vehicles. Ask apple how hard it is to make a car.
And apple don't even make their own phones, foxconn make em. You make it sound trivial to make a car with an FSD capable sensor suite. How long have waymo been at it, and they STILL dont even come close to breaking even?
Its truly comical for anybody to keep flogging the dead horse that is the LIDAR+maps approach. Its doomed. The only question is how many more billion get incinerated before google do what they always do...and give up.

 

[Usain]

You don't know their costs.

It's true that their numbers are not public. I'm just waiting for someone at Waymo to say, "We see a clear path to profitability." I haven't heard that yet. From everything we do know, Waymo is still losing a lot of money.

Simple extrapolation says 3 years to 50k, so summer 2027. When do you predict Tesla will have 50k driverless robotaxis in service?

I don't think Waymo will be able to get to 50k in 3 years because the losses will be too high to keep expanding. I believe Tesla is far more likely to be at 50k driverless robotaxis in 3 years because Tesla's cost per vehicle is far lower. And more importantly, their cost to run the service will be lower with a superior end-to-end FSD system.

How will Tesla will avoid these costs?

At first they won't. But it won't take long before the need for human intervention becomes so low these costs all but go away.

They can start a service with safety drivers today. I'm sure they will on 8/8, along with multiple grandiose claims.
Make a real prediction. How many paid driverless rides will they provide in December 2025? Waymo should do ~3 million that month.

I don't think 2025 matters much. It's some time around 2026 or 2027 when I expect Tesla will start to dominate. That's my prediction.

Why? H/W cost? What's the incremental h/w cost per mile? A nickel? A dime? What will it be in two years? Decent solid-state lidar is already sub-$500 with a specification war driving rapid improvement.

Tesla will build the whole robotaxi vehicle at a much lower cost than Waymo will ever be able to acquire outfitted vehicles from a third party. I don't think that is in dispute.

That's one of the things we expect to see on 8/8. We will see a super low-cost, purpose-built robotaxi. Tesla will build them very fast and very cheap using the unboxed process. Nobody will be able to match it.

 

[willow_hiller]

Rob Maurer shares both his contemporaneous and present thoughts on Elon's compensation package, and why he's voting for ratification:

https://twitter.com/x/status/1799536504412205175

 

[Usain]

Can't confirm this is actually 12.4, but was posted on X to be so. This would drive me crazy...

https://twitter.com/x/status/1799332647241662853

Omar is not seeing that behavior. But if it is common I'm sure it will be fixed, probably before 12.4.x goes to wider release.

 

[mrmage]

All of which are phones, not vehicles. Ask apple how hard it is to make a car.
And apple don't even make their own phones, foxconn make em. You make it sound trivial to make a car with an FSD capable sensor suite. How long have waymo been at it, and they STILL dont even come close to breaking even?
Its truly comical for anybody to keep flogging the dead horse that is the LIDAR+maps approach. Its doomed. The only question is how many more billion get incinerated before google do what they always do...and give up.

Let me remove LIDAR because you seem to have a bone to pick. Let's say Waymo has a safe, driverless, robotaxi that has an additional component called Something Useless (SU) that costs an extra $1k / year (pick a number). SU does nothing at all. Tesla has a robotaxi service that does not have SU, so it operates with lower overhead.

I contend that Waymo has a perfectly competitive service despite paying for SU, for any reasonable cost for SU. The margin will be a little lower, but it's easily bearable.

 

[Usain]

Let me remove LIDAR because you seem to have a bone to pick. Let's say Waymo has a safe, driverless, robotaxi that has an additional component called Something Useless (SU) that costs an extra $1k / year (pick a number). SU does nothing at all. Tesla has a robotaxi service that does not have SU, so it operates with lower overhead.

I contend that Waymo has a perfectly competitive service despite paying for SU, for any reasonable cost for SU. The margin will be a little lower, but it's easily bearable.

The cost difference between Waymo's robotaxi vehicles and Tesla's robotaxi vehicles will easily be more on the order of $20K to $30K. And I'm probably being generous.

 

[GatorMeat]

https://twitter.com/x/status/1799132862928789740

listen to the rest. He continues on claiming EVs have limited range, aren't for roadtrips, and when he tells this to Elon, he agrees with him and then he goes on babling more about how bad EV range is.

 

[GatorMeat]

I admit Im in awe of the steel balls this man have in completely contradicting himself multiple times in the span of months. Elon and EVs were his enemy not long ago.

He does a disservice by not showing the whole clip. DJT goes on telling stories about poor range, EVs are bad for road trips, DJT tells this to Elon (DJT being the all-knowing expert in everything) and Elon agrees with him.

 

[GatorMeat]

https://twitter.com/x/status/1799132862928789740

DJT contintues: "EVs are not for road trips, they have bad range, I tell this to Elon and he agrees with me. EVs are for short range....

 

[GatorMeat]

He wants 25% of voting power using different share class. But unfortunately this can’t happen post IPO. His concern with AI has been there for years. So developing it with a great chance of black rock and vanguard choosing to militarize it or control the direction of it, is a valid concern of his. This really shouldn’t be surprising.

Now the fact that he bought twitter and now wants this , is definitely not a good look. But he also felt a major issue was brewing with the media and direction of this country. Wether one chooses to believe that or not, is up to them. But he believes it.

From Reuters:

“In a separate post on X, he said he would be fine with a dual-class share structure to achieve his goal of getting 25% voting control, but was told it was impossible after Tesla's initial public offering.

"It's weird that a crazy multi-class share structure like Meta has, which gives the next 20+ generations of Zuckerbergs control, is fine pre-IPO, but even a reasonable dual-class is not allowed post-IPO," he said, referring to the Facebook parent's founder Mark Zuckerberg.“

Zuck has proven to be a good CEO and made a course change when the market told him to change his focus. Elon tells us all to GFY

 

[AndrewZ]

DJT contintues: "EVs are not for road trips, they have bad range, I tell this to Elon and he agrees with me. EVs are for short range....

Irks me bad when people share little snippets of videos to push a narrative while omitting important bits preceding and following the snippet, but that’s social media for you

 

[Electroman]

If he wants his package, he should probably quit posting stuff like this. He looks like an ass and a crazy MAGA person.

https://twitter.com/x/status/1799417761539805431

What is his pay package to do with his opinion and his decision to express that opinion in social media?

If you contract me to mow your lawn and when it comes to payment you find out that I have been a vocal Putin admirer, you think you should withhold my payment for the work I already did?

 

[GatorMeat]

It's beyond ridiculous to claim Tesla investors should replace a CEO who just accomplished this:

https://twitter.com/x/status/1798792222743122164

Get this: Tesla is nothing without Elon Musk, without him Tesla is just a bigger Rivian, and one would be better off to just buy RIVN since it's a lot cheaper.

And who do you plan to replace him with? Trevor Milton? Peter Rawlinson?

Maybe Tesla's management structure should match SpaceX, where there is a professional at the top running the show instead of having all the execs report to Elon who fills a CEO role that he is not good at, like at Tesla.

 

[Electroman]

Which is related to this issue, because Delaware essentially said that the board was not doing its job,

Okay I get that. It is one the establishment's method of conducting its lawfare to neutralize Musk. If you don't stop him from saying bad things about us, we know how to make his life miserable.