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

[Bet TSLA]

I don't particularly like Lex or his style, but he's a brilliant guy, and probably the camera doesn't help him (in the past I happened to host a small TV show and you'd be surprised how much television "steals" from a person personality/enthusiasm/personality. I bet in person he's much less boring.
He's humble, he can listen and he actually can follow people in very nerdy paths of conversation. Not a superpower but a real skill. How many podcasters out there can really talk to Karpathy?
Also, he's not afraid to talk for a lot of time, even if the conversation appears to be boring. That is likely something that helps people connect with him (if you haven't, check out is interview with Rick Rubin). Intimate conversation take time and aren't really flashy. It's a different kind of Youtube, distant from the boombastic podcasters or interviewers out there, with dopamine cuts and tight editing.
It's like old school, serious stuff, and smart people like to be taken seriously.

Just a minor point to add to your analysis -- I listen to his podcasts. I think I've watched a single Youtube interview, one of the Donald Knuth ones, because I wanted to see what he looked like now. So I just hear the voices and the words.

 

[petit_bateau]

I believe that I'm still in the minority here that think that the semi's GMs are going to be terrible. Even at volume, I think they will be low. It's not obvious to me that delivering semis should be a catalyst if the expected margins are poor? At LRIP the margins will be negative (typical), but when you're selling a vehicle that is 30% more expensive than a MS Plaid, but with 10x more kWh, it will always be much lower margin IMO.

	Kwh	$/kWh	Pack cost	Price	Pack$/price
Plaid	100	125	$12,500	$136,000	9.2%
Semi	1000	125	$125,000	$180,000	69.4%

I think the semi is more about the mission...

FWIW my GM% projections for the Semi blend are/were as follows. Note my Semi blend assumes some shorter range and lower load versions in the mix. So I am not quite as pessimistic as you, but still somewhat cautious :

 

[MaChiMiB]

Surprised no one has pointed out that the stock is

GREEN

GREEN

GREEN

I thought this was an investment forum.

Also a black cat just walked under the ladder I'm currently standing on.

Some less green in Berlin though.
Progress looks good.
No drone flyovers were allowed last weekend (and are generally only allowed on weekends). 50m alt limit too, because of the nearby airport.

https://twitter.com/x/status/1587098887457849344

 

[Tony73]

I believe that I'm still in the minority here that think that the semi's GMs are going to be terrible. Even at volume, I think they will be low. It's not obvious to me that delivering semis should be a catalyst if the expected margins are poor? At LRIP the margins will be negative (typical), but when you're selling a vehicle that is 30% more expensive than a MS Plaid, but with 10x more kWh, it will always be much lower margin IMO.

	Kwh	$/kWh	Pack cost	Price	Pack$/price
Plaid	100	125	$12,500	$136,000	9.2%
Semi	1000	125	$125,000	$180,000	69.4%

I think the semi is more about the mission...

There is zero chance semi pricing will stay at $180k for the 500 mile version when pricing is updated. The only question is whether reservation holders will get the old or updated price.
The short range electric semis currently on the market cost over $250k so there is a lot of room to increase prices and still be the best value on the market by far.

 

[ClimateOptimst]

I'm very confused about the semi semi ramp. Aren't they just building a handful out of Reno?

Don't we have to build a whole production facility in Texas? Has it even been started?

This seems more like a concession to the marketing and IR departments at Pepsi and Tesla than start of production. (I'm not criticizing that, If I was Pepsi I'd get a lot of pictures of Pepsla trucks in cool paint jobs on TV as fast as I can.)

Anybody know what's really going on with production?

Electrek has a few articles saying the Nevada line will be capable of producing 5 trucks/week. There was a more recent article at Teslarati (I tried to find it again but couldn't) saying they'd talked to sources and the line would produce 3 trucks/week & was largely manual assembly. If true, it's basically a pilot line until the real production starts coming out of Austin presumably.

 

[H Mak]

And NKLA is up 12.8%.

Time to jump ship then, after today's stellar performance. It's just down the 94% now from its high

 

[betstarship]

When watching the Fridman podcast with Karpathy, the latter talked a lot about how Tesla has a lot of copy and pasting lately between teams (even gave an example of FSD Vision and Optimus)...

...I wonder if Musk is just bored and the rest of Tesla is now just a matter of copy and pasting new factories, vehicles, robotics, and customization + scale for the foreseeable future. Thus, he's going to be able to be "distracted" with Twitter.

 

[UkNorthampton]

When watching the Fridman podcast with Karpathy, the latter talked a lot about how Tesla has a lot of copy and pasting lately between teams (even gave an example of FSD Vision and Optimus)...

...I wonder if Musk is just bored and the rest of Tesla is now just a matter of copy and pasting new factories, vehicles, robotics, and customization + scale for the foreseeable future. Thus, he's going to be able to be "distracted" with Twitter.

I think Elon will focus on the most important and hardest aspects affecting continuation of sentient life. Tesla's in his to-do list, but not the only thing and I think there's a great set of people at Tesla to handle most of it.

 

[ZachF]

Surprised no one has pointed out that the stock is

GREEN

GREEN

GREEN

I thought this was an investment forum.

Also a black cat just walked under the ladder I'm currently standing on.

Bickering feeds the chaos gods who grant us our gains!

 

[dhrivnak]

Soo... your point is that if you don't count all the relevant factors, then humans are as good as robots?

So, no, they aren't. Electric motors are 90% efficient, humans are only 20% efflicient WHILE they're working. (that's -4x)

Humans only work about 6.5 hours per day, while robots work about 20 hrs (-3x). And humans take vacations (-10%). And sick time (pick a number, but add training time for the replacement worker).

This is a losing arguement. On current energy costs alone, it's already 20x cheaper to run a robot. And I don't have to pay retirement benefits to the robot.

Just no.

And what happens to taxes when most people no longer need to work but still need retirement, health care and living. Those are costs that do not truly disappear.

 

[Xepa777]

When watching the Fridman podcast with Karpathy, the latter talked a lot about how Tesla has a lot of copy and pasting lately between teams (even gave an example of FSD Vision and Optimus)...

...I wonder if Musk is just bored and the rest of Tesla is now just a matter of copy and pasting new factories, vehicles, robotics, and customization + scale for the foreseeable future. Thus, he's going to be able to be "distracted" with Twitter.

This is exactly what I've been saying and a big reason why Tesla is so de-risked.

• Semi-production to reach global scale will take 5+ years worth of ramp, with finding ways to automate the ramp.
• Energy storage (both megapack and power wall) has a basic framework with a huge backlog, just about battery management (i.e. eventually transitioning power walls to LFPs)
• Solar panels/roofs have challenges, but most are non-engineering challenges and more about the permits/governments/ad-hoc parts of installation and service, not an area Elon's passionate about I presume.
• Cars are largely copy/paste. Maybe if Cybertruck does prove to be wildly profitable, making more cars similarly in structure to that (but there's already a blueprint there).
• FSD and Dojo are unsolved, but there's a pathway there and it's not Elon's forte as a physics/hard engineer sorta guy, so he's just setting the milestones and motivating the team.
• Optimus does have a lot of physical real world problems left to solve, so I suppose that's an area Elon can have more direct involvement with.

Any dual SpaceX follower sees the slant Elon has provided on Starship development. And yah looks like Twitter now will take a lot of time.

But that's good. It means Tesla is self-sufficient and de-risked. A well-oiled machine. And it's just a waiting game until it continues to pay off for us investors, whether that's in 2023 or 2028.

 

[ZeApelido]

First Argo AI going kaput, now George Hotz announcing he is leaving comma.ai.

The Hero’s Journey

I’m taking some time away from comma. In some ways, I feel like it’s all planned out, it only remains to see if it works or not. I rewatched my first Lex Fridman and realized I hadn’t said much new on the self driving front for the past couple years. My two part blog post from 2019 remains a...

geohot.github.io

 

[Todd Burch]

Remarkably, um, stable share price the last few days. Coincidence?

Answer: Nope.

 

[insaneoctane]

That's part of it but it's actually more of a direct marketing effort than that.

With EV's the next new thing, it's difficult for many to get excited about buying a new ICE car. These ads get consumers excited about something new, EV's, which gets them thinking about how much down payment they have and how much they can afford monthly. It gets consumers thinking about a new car, even if it is an EV. If they can get them into one of their showrooms ready for a new car, it's a real letdown when they find out how far off a new car is. So, it's a simple "switcharoo" to get them into a brand-new ICE car to satisfy their desire for a new car. They will cite figures on how high resale value have been and tell them the first batch of EV's will be quite expensive and the prices will likely fall as EV's become more ubiquitous (OK, the car salesman is not going to use the "ubiquitous" word, but you get the idea).

The manufacturers know their dealers don't have any EV's on the lot to sell the EV curious and will "do the right thing" and get them into an ICE car. One more ICE car before EV's are "ready".

Get the mark ready to buy and slip them into a shiny new ICE car while making them feel smart for not jumping on the EV early-adopter bandwagon too soon. The charging networks are still being built out, the EV's aren't even ready yet, and will be expensive and somewhat unproven. Tesla are perpetually over-priced, so do the 'sensible' thing and slip into this shiny new ICE car with proven reliability and cost efficiency. They are just using the thought of EV's to get people mentally ready to buy a new car. This will work, right up until it won't anymore.

Believe it or not, but I kinda think this is business as usual for F and GM. They're marketing departments are pretending Tesla doesn't exist. They know many of their truck buyers are not interested in the fake-climate change driven EVs yet. For many of their buyers it's keeping the brand relevant so that their buyers don't leave the brand. Supply chain and chip woes still seem to have their ICEV stock low, so Osborning seems less likely from that perspective too. Once they start piling up inventory like before Covid, this strategy might get tweaked.

 

[Cosmacelf]

But it's potentially devastating to the other US companies who are trying to wing it on their own and trying desperately to leverage their sunk costs in ICE infrastructure and last generation manufacturing technology. China hasn't been successful in the US, and the IRA legislation might keep them at bay for a few years more, but they are way ahead of US manufacturers.

It's also likely to have some nasty effects on European manufacturers as well. There is a lot less protectionism in Europe and Chinese imports are gaining more traction there. Diess was perhaps their best hope for being able to play catch up and they got rid of him.

If so, it’s their own damn fault. Tesla started selling cars in 2008. They let Tesla have a 16! year head start before getting serious (assuming they are serious now, which not all are).

 

[Todd Burch]

First Argo AI going kaput, now George Hotz announcing he is leaving comma.ai.

The Hero’s Journey

I’m taking some time away from comma. In some ways, I feel like it’s all planned out, it only remains to see if it works or not. I rewatched my first Lex Fridman and realized I hadn’t said much new on the self driving front for the past couple years. My two part blog post from 2019 remains a...

geohot.github.io

Two possibilities I can imagine based on what we're seeing the last few days WRT autonomous driving companies:

1. Due to economic conditions, self-driving hopefuls who aren't as far ahead as others thought aren't seeing the money come in to support continued R&D on such a hard problem and are therefore abandoning it, or

2. Competitors are seeing how far ahead Tesla is on FSD and throwing in the towel.

 

[insaneoctane]

I wrote up something from looking at a FB/META grumbling on twitter. This was meant to some of my friends who are not stock savvy. So, figured to get your views before sending to them.
Am I on the right track?

View attachment 869545

Fun data. Seems like market cap may be a factor to consider too...

 

[SOULPEDL]

Two possibilities I can imagine based on what we're seeing the last few days WRT autonomous driving companies:

1. Due to economic conditions, self-driving hopefuls who aren't as far ahead as others thought aren't seeing the money come in to support continued R&D on such a hard problem and are therefore abandoning it, or and

2. Competitors are seeing how far ahead Tesla is on FSD and throwing in the towel.

FIFY

 

[Cosmacelf]

From the competition is coming department. Hive mind at reddit says they are running commercials just to stop people from buying EVs that are available today.

They are doing this to get people into dealerships. A good salesman will sell two cars when someone walks in. ”Let me put your name on this 2 year long waiting list, and then buy this ICE in the interim…”.

 

[Ogre]

If so, it’s their own damn fault. Tesla started selling cars in 2008. They let Tesla have a 16! year head start before getting serious (assuming they are now serious now, which not all are).

Yeah. I was just pointing out who would be most impacted by the Chinese… not trying to defend them.

The most likely scenario is these companies are toast of what the Chinese do. They have tens of billions owed on production equipment and resources which won’t be remotely useful in 5 years.

This just accellerates that a bit.