Hey fam: Supersized newsletter today. This Triad goes to four.
It’s about Tesla and Elon Musk and autonomous vehicles. It gets fairly technical at times. But it’s really about how we’ve turned over large portions of America to insanely rich people who have no idea what they’re doing.
It’s about oligarchy and technology. I hope you’ll take the ride with me—and share it if you are so inclined.
P.S.: If this edition ends abruptly in your inbox, click the title to get to the full version on the site. Like I said—it’s a long one.
For years, Elon Musk has been promising that Teslas will operate completely autonomously in “Full Self Driving” (FSD) mode. And when I say years, I mean years:
December 2015: “We’re going to end up with complete autonomy, and I think we will have complete autonomy in approximately two years.”
January 2016: “In ~2 years, summon should work anywhere connected by land & not blocked by borders, eg you’re in LA and the car is in NY.”
June 2016: “I really would consider autonomous driving to be basically a solved problem. . . . I think we’re basically less than two years away from complete autonomy, complete—safer than a human. However regulators will take at least another year.”
October 2016: By the end of 2017 Tesla will demonstrate a fully autonomous drive from “a home in L.A., to Times Square . . . without the need for a single touch, including the charging.”
March 2018: “I think probably by end of next year [end of 2019] self-driving will encompass essentially all modes of driving”
February 2019: “I think we will be feature complete—full self-driving—this year. Meaning the car will be able to find you in a parking lot, pick you up, take you all the way to your destination without an intervention, this year."
I’m going to stop the litany here, but it continued. For a decade Tesla’s “Full Self Driving” has been a running joke. Every year, Musk says that FSD is months, or weeks away. Every year it slips over the horizon. And every year the media and the stock market pretend that Musk hasn’t failed.
But now Tesla’s FSD is really, truly, absolutely about to happen.
Maybe.
Tesla “Robotaxis” are coming to Austin in June and Elon Musk says they are the future of the company. Facing declining consumer sales, the company’s valuation is based on the idea that Tesla is about to transform the auto industry by turning car ownership into something akin to software-as-a-service.
As usual, the details are murky. Fortune reports that as of this week Tesla still has not briefed many of the relevant authorities in Austin about what the company’s Robotaxis are, how they will operate, or how first responders and law enforcement are supposed to deal with them. And the cars that appear on the street aren’t going to be the “Robotaxis” Musk keeps promising. They will be modified Model Y’s.
Also, there will only be 10 to 20 of them.
You may be forgiven for wondering if this Robotaxi launch will be a real proof-of-concept for autonomous driving, or the Tesla version of Amazon’s “Just Walk Out” technology.
Of course we already have full self-driving technology. It’s called Waymo, it’s run by Google, and it operates a driverless taxi service in a growing number of cities across America. Just a few days ago, Tesla’s head of self-driving admitted that Tesla FSD is “a couple years” behind Waymo, which in Elon-speak probably means a decade. Or more.
So in a sense, Musk was correct that autonomous driving is a “solved problem.” Just not by him.
We have been chasing the driverless car for a hundred years. During that time people have tried all sorts of paradigms. At the 1939 World’s Fair, General Motors proposed a system of remote-controlled cars. (Put a pin in that.) At another phase, people thought that beacons on roadways could emit signals to guide cars or that “world maps” could give the car’s brain a detailed enough view for them to navigate on their own. Starting in the early 2000s the industry settled on the idea of making cars that navigated by “sensing” their surroundings.
This movement was kickstarted by a DARPA challenge (government intervention for the win) and only after university researchers showed that the sensing paradigm held promise did tech companies pile into the space.
Google started working on self-driving cars in 2009 and it was a slog. Initially they focused on the use of cameras to understand what was happening around the vehicle’s environment. But no matter how smart AI got, the cameras weren’t enough.
Eventually, Google took a hybrid approach. Its vehicles still had cameras trained on mountains of data. But they also relied on incredibly detailed maps of the drive area and—most importantly—advanced active sensors such as radar and lidar. It was only by combining all three of these techniques that Google was able to build a Level 4 autonomous vehicle.
And they work.
Under its Waymo brand, Google has driverless taxi fleets operating in eight cities (so far). They use a custom Jaguar, fitted with cameras, sensors, and a fairly high-powered onboard computer. While Elon has been making FSD promises, Waymo vehicles have driven 40 million real-world miles.
What happened?
Elon Musk was stupid. That’s what happened.
LiDAR (stands for “light detection and ranging”) is a technology that uses lasers to measure distance. It’s been around since the 1960s, but in the 2000s, lidar started making its way into automotive platforms to power what we now call advanced safety features.
Lane-departure detection, front-collision avoidance, adaptive cruise control, emergency break assist—all of these features are powered by lidar. Porting lidar into the automobile has been the most important car-safety advance since the airbag.
Elon Musk hated it.
Musk began a crusade against lidar in 2019, calling it “a fool’s errand.” He called lidar “a crutch.” Over the years he doubled down on this assessment calling lidar “freaking stupid” and “expensive and unnecessary.”
What galled him about lidar seems to be this last part: the cost. “You have expensive hardware that’s worthless on the car,” the visionary tech genius explained.
Musk never allowed lidar to be integrated with Tesla vehicles, but early Teslas did have radar. Until 2021, that is, when Musk became obsessed with cutting production costs and decreed that radar should be eliminated from new Tesla vehicles. And not just that: Tesla disabled the radar units on its existing cars, too.
So it wasn’t just cost. It was ideology.
Some Tesla engineers were aghast, said former employees with knowledge of his reaction, speaking on the condition of anonymity for fear of retribution. They contacted a trusted former executive for advice on how to talk Musk out of it, in previously unreported pushback. Without radar, Teslas would be susceptible to basic perception errors if the cameras were obscured by raindrops or even bright sunlight, problems that could lead to crashes.
Musk was unconvinced and overruled his engineers. In May 2021 Tesla announced it was eliminating radar on new cars. Soon after, the company began disabling radar in cars already on the road. The result, according to interviews with nearly a dozen former employees and test drivers, safety officials and other experts, was an uptick in crashes, near misses and other embarrassing mistakes by Tesla vehicles suddenly deprived of a critical sensor.
What a singular genius.
Musk was committed to a “vision only” approach for FSD that relied solely on cameras to passively collect data and onboard compute power to crunch it. Here is his sophisticated explanation for why he believed in “vision only”:
“The road system is designed for cameras (eyes) & neural nets (brains).”
Like I said: Genius.
There is an actually sophisticated version of the “vision only” argument—but it’s rooted in business imperatives, not engineering.
To believe that “vision only” FSD can work is to believe that AI is the most important process and that data is the most important commodity. Which happens to line up nicely with Musk’s ambitions. Tesla, by dint of having so many cameras on the road, has more data than anyone else. So if Musk believed that data was his moat and that the future of his empire was xAI, then he would try to make FSD run on a camera-only platform, regardless of what the results were showing.
The problem is that even in this “sophisticated” theory, Elon Musk is less like Tony Stark and more like Michael Scott.
“People don’t shoot lasers out of their eyes to drive,” Musk said recently as an explanation for why Teslas didn’t need lidar.
You tell me: Is that The Avengers or The Office?
And yet, Tesla has quietly been spending money on lidar. Why? Unclear.
I wonder if the Robotaxis Tesla puts in Austin in the coming weeks will be “vision-only” FSD or if they will be using lidar “crutches.”
Speaking of crutches, I want to highlight the disconnect between Musk’s FSD promises and the Robotaxi reality.
Musk promised a robot car that could drive from New York to Los Angeles, by itself, in 2018.
In June 2025, Musk is going to put between 10 and 20 robot taxis on the streets of Austin.
These vehicles will be strictly geofenced.
And they won’t even be truly self-driving! Because buried in all of the Robotaxi talk is the fact that Tesla has been hiring “tele ops”—meaning supervisors who watch the Robotaxis from afar and can take control of the vehicles remotely as needed.
In other words: Back to the Depression-era World’s Fair version of “autonomous vehicles” that we talked about up top.
If you’re keeping score, Elon Musk has:
Promised FSD every year for a decade.
Made terrible engineering decisions that hobbled his company’s progress.
Created a “vision-only” “self-driving” product that is vastly inferior to the state-of-the-art hybrid approach.
And is about to deliver a bait-and-switch “FSD” vehicle that is a glorified remote-control car.
This is the guy we put in charge of dismantling the federal government. Are you surprised that people are dying as a result?
Last week Bloomberg published a big profile of Microsoft CEO Satya Nadella, in which he claimed to use AI for most of his job. Here’s Bloomberg:
He likes podcasts, but instead of listening to them, he loads transcripts into the Copilot app on his iPhone so he can chat with the voice assistant about the content of an episode in the car on his commute to Redmond. At the office, he relies on Copilot to deliver summaries of messages he receives in Outlook and Teams and toggles among at least 10 custom agents from Copilot Studio. He views them as his AI chiefs of staff, delegating meeting prep, research and other tasks to the bots. “I’m an email typist,” Nadella jokes of his job, noting that Copilot is thankfully very good at triaging his messages.
And here’s Ed Zitron putting the finest possible point on this madness:
Is it because we've handed our economy to men that get paid $79 million a year to do a job they can't seem to describe, and even that, they would sooner offload to a bunch of unreliable AI models than actually do?
We live in the era of the symbolic executive, when "being good at stuff" matters far less than the appearance of doing stuff, where "what's useful" is dictated not by outputs or metrics that one can measure but rather the vibes passed between managers and executives that have worked their entire careers to escape the world of work. Our economy is run by people that don't participate in it and our tech companies are directed by people that don't experience the problems they allege to solve for their customers, as the modern executive is no longer a person with demands or responsibilities beyond their allegiance to shareholder value.
Read the whole thing. It’s all of a piece with Musk and FSD.