$4b in CapEx alone for <500k units/year manufacturing capacity for a single model is the worst program-level capital efficiency I've ever heard of... in an industry that is infamously capital inefficient. And that's just what Tesla spent in 2017!
Some of that 2017 spend may have gone toward opening new sales/service locations and Superchargers to support the higher-volumes planned for Model 3. Then again, a good deal of 2016's $1.44b and 2018's $2.32b probably went into Model 3 manufacturing as well.
It's also super-important to recognize what all these billions bought: not a competitive flexible/scalable manufacturing system, but labor-intensive low-to-no-automation general assembly in a freaking tent. This makes M3 a capital efficiency failure of epic-historic proportions.
The worst part: Tesla could have totally avoided this fiasco by listening to manufacturing experts who predicted the "Alien Dreadnought" would end up in utter failure. Instead it was "Elon lands rockets on ships, bro" and "you're just trying to drive down the stock." 🤦♂️🤦♂️🤦♂️🤦♂️🤦♂️
That's the really scary part of this: the response from investors has been to throw more capital at Tesla, drive up the valuation, and angrily reject good advice as jealousy, insecurity and/or an effort to hurt Tesla. Talk about a recipe for not learning obvious lessons!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
There's a fascinating thread here, where Elon seems to go nuts for anything Google is doing, and funds "competition" for their work based on the assumption that they are much farther along than they are. semafor.com/article/03/24/…
This is literally how Tesla ended up going into driving automation tech! Most people still don't know that Google developed a system called "AutoPilot"but didn't bring it to market based on the exact safety concerns plaguing Tesla's version thedrive.com/tech/29877/dr-…
The driving automation play and the AI play are different in some respects, but in both cases Musk was trying to "frontrun" competition with what he (mistakenly) thought were imminent Google monopolies. He's kind of an emerging tech squatter?
Here's the thing about automation: "understanding the technology" is barely half the battle. If you don't understand the work you're trying to automate, you'll never succeed. The people who proclaim rampant job loss from automation grok the tech, but are clueless about the jobs.
Driving automation technology is rampant with this flavor of cluelessness. People see a video and say "oh look, it's driving" without having the first clue about what driving in an economically valuable way entails (long-tail reliability, insurance, customer service, etc)
Right now the auto industry is planning to spend the technology dividend on Level 2 systems, which provide an enticing simulacrum of self-driving without delivering any actual proven safety benefits. That's where the money is, so never mind about safety I guess!
It doesn't have to be that way. Instead we could apply the performance improvements and cost reductions in driving automation tech to the systems we know deliver improved safety outcomes... which just happen to be the systems that don't pretend to be "self-driving."
I asked Midjourney to show me a Persian miniature of a robotaxi and it gave the thing a lidar dome that looks like a mosque 🥹
I've been having fun trying to understand how Midjourney imagines the term "robotaxi" in different contexts, so I'll continue to post the results in this thread.
Here we have one of my favorite examples of an ancient Egyptian robotaxi, complete with a massive infotainment system
The Oregon Trail sure ain't what it used to be, now that we got them self-driving Conestoga wagons, I'll tell you what.
It says a lot about the Tesla fandom that the response to revelations about Tesla's malfeasance is always "oh well every other automaker does this, you just never hear about it, and though I don't actually cite any sources just trust that I have looked into this."
Tesla has tried to do this itself, for example in its only public statement on my reporting about its record of environmental violations at Fremont.
I dug into their claims, and the comparison Tesla drew actually shows how much worse their record is thedrive.com/tech/28432/tes…
Fremont has some 47 Clean Air Act violations since it last settled some 30 violations with @AirDistrict in mid-2021. I challenge Tesla's defenders to name a single auto final assembly factory with a record that comes close to that. Just one! echo.epa.gov/detailed-facil…
The whole "this wasn't VCs, this was just SVB" narrative elides a major point of this whole episode: insularity concentrates risk in lots of ways. You think it was a coincidence that a bank so embedded in this specific ecosystem failed to hedge for higher interest rates?
Like, I guess it's probably a good thing that Silicon Valley didn't do all of it's Finding Out all at once (as much as some players might deserve it)... but let's not pretend we can or should prevent that Finding Out from happening in a more orderly fashion.
I have friends and acquaintances in Silicon Valley who are amazing people doing amazing work, and I wish them nothing but good things. But there were plenty of good people in Detroit, and that culture still had to undergo a major reckoning. And guess what? That pain was worth it.