Tesla delivered the first Cybertrucks yesterday, to lukewarm reaction given revelations on its specs, pricing and availability. But Tesla’s miss might be other EV truck makers’ gain, as the big question marks on Cybertruck have been answered disappointingly.
At the original Cybertruck unveiling, Tesla said the truck would start at $39,900, be available in late 2021, have a payload of 3,500 pounds in a 6.5ft bed, and a tow rating of up to 14,000 pounds max range of 500+ miles for the top-end version (which was meant to start at $69,900).
As of yesterday, we know that none of those numbers are true. The truck starts at $79,990 today (and $60,990 in 2025), has a payload of 2,500 pounds in a 6ft bed, tow rating of up to 11,000 pounds and range of 340 miles, or 470 with an additional battery that eats up a chunk of your bed space.
It seems like the only place where Tesla exceeded its original estimation is in 0-60 times for the mid- and high-spec configurations, which are .6 and .3 seconds faster respectively. These are nice specs and it’s awesome to see a Cybertruck beat a Porsche 911 while towing a Porsche 911, but a truck is still not a sportscar no matter what the quarter-mile results suggest.
And overall, the presentation at the event was just underwhelming (you can view a recording of our live-react watch party here). Not only was the event almost half an hour late and poorly-lit, but Tesla CEO Elon Musk showed off his standard poor presentation skills (which have been getting worse lately with his twitter distraction, prompting questions from investors). Worse, some cool Cybertruck features weren’t even mentioned – like Powershare bidirectional charging which lets you run your home, camp, or another vehicle off of the Cybertruck’s battery.
All of this has led to a lot of disappointment among order holders – and made room for competing electric trucks now that we finally know answers to questions that have loomed large over the segment for years now.
Other electric trucks compare favorably with Cybertruck’s unveiled specs, pricing
Prior to now, other electric trucks have suffered from the comparison to a potential Cybertruck, with millions of preorders and a promised $40k base price.
But now that Cybertruck specs and pricing have been officially unveiled, those comparisons become a lot better.
There is a question whether many people were actually cross-shopping the Cybertruck versus more traditional-looking pickup trucks in the first place, but a difference in base price of more than $20k can make up for a lot of questionable styling.
The comparison was made directly on reddit, in a post comparing Musk’s past statements about the Lightning’s price with the actual prices of the Cybertruck – which are now much more expensive than the base and even upgraded trims of the Lightning, both of which are available now whereas the base model Cybertruck won’t come until 2025.
So despite the Cybertruck’s unconventional look, it’s clear that a lot of people are putting it into the “truck” category, and comparing it against other offerings. Tesla did so too, themselves, by offering a comparison against the Lightning, F-350 Super Duty, and Rivian R1T in a tractor-pulling test during the presentation yesterday.
The market has noticed
This means that Tesla’s loss is the other trucks’ gain. If Tesla truly has 2 million order holders who were waiting off on buying an electric truck until they finally saw the Cybertruck specs, surely some percentage of those holders will end up deciding to go elsewhere.
And it seems like the stock market has already made this comparison, because the market isn’t looking favorably on Tesla’s unveiling, whereas other EV truck makers are up quite a bit (with Rivian leading the pack, since trucks are their whole thing). The whole market is up intraday, but Rivian is up quite a bit more than the sector as a whole.
Given the poor reaction to the Cybertruck’s unveiled specs, maybe it’s time for those production plans to ratchet up a little further, since we think there will be some customers out there looking for something else to buy.
FTC: We use income earning auto affiliate links.More.
There you are, motoring along in your Volvo XC90 PHEV with the Pilot Assist engaged alongside a big 18-wheeler at a comfortable 70 mph cruise when the interstate starts to slowly sweep left. From the drivers’ seat, that semi on your right looks awfully close. As the steering wheel turns itself in your hand, you start to wonder if that truck’s a bit too close. The car isn’t doing anything wrong, but it’s too close for your comfort and you give the wheel a little nudge to hug the inside of the lane just a bit more.
These deeply personal preferences are tough to quantify, and highlight a simple fact about Advanced Driver Assistance Systems (ADAS) that the industry at-large hasn’t yet to come to terms with: when it comes to self-driving cars, one size does not fit all.
The Volvo experience I outlined above was very real, happening just as the wife and I were arguing about the relative merits of our very different choice in running shoes. She prefers the supportive, cushion-y ride of the HOKA Clifton 9s, which I’ve become convinced are The Devil™, preferring instead the zero-lift, no-cushion feel of my Xero Prio runners. The intervention with the Volvo interrupted that particular argument and started another. Namely, the one about why I had chosen that moment to “interfere” with the Pilot Assist.
“It was too close to that truck,” I explained. “Freaked me out.”
Advertisement – scroll for more content
“That’s how I feel in the Honda,” she said. “I’m always afraid that it’s going to try and put me into oncoming traffic.”
That’s when the idea for this post came to me. Because, as a car brand, it’s really not possible to just say that your car has ADAS or doesn’t have ADAS in a binary sense. That’s because these systems are not just proprietary to a given brand, they can vary from vehicle-to-vehicle within that brand, and each one can have distinct lane centering behavior, steering feel, lane change aggressiveness, braking distances, timing for its hand-off warnings, and probably a bunch of other stuff that I haven’t even thought of depending on what kind of cameras, sensors, and software the specific vehicle you are in is equipped with.
It’s a bit of a mess, in other words.
Opinion: Honda Sensing gets it right
I first experienced Honda’s ADAS in 2014, driving a then-new CR-V between Chicago and Bay Harbor, Michigan for an Acura press drive. Even in its early generations, I was impressed with the way it handled stop-and-go traffic, the way it guided you through turns, but didn’t do the turning for you, and the speed and intensity it used in braking very much mirrored my own.
Last month, I had a chance to test out the 2025 Honda Civic Sport Touring Hybrid for a week on Cape Cod. I picked the car up at PreFlight Parking outside Boston Logan, jammed it with luggage, and immediately hit heavy traffic, where the Honda Sensing Low-Speed Follow function took me right back to 2014, ratatouille-style, when my experience in that car had led me to believe that self-driving cars were right around the corner.
In the decade-plus since experiencing that first autonomous Acura, I’ve had the chance to experience Ford BlueCruise, Tesla Autopilot and FSD, and Mercedes-Benz DRIVE PILOT. And all, interestingly enough, in and around the Circuit of the Americas in Austin at one time or another over my three years of hosting Electrify Expo events there.
Each different OEMs’ system had its strengths and quirks. I remember Mercedes DRIVE PILOT as impressively precise, even clinical. The Ford system faded into memory. I couldn’t tell you anything about it, which is probably high praise. The Tesla systems, though, stood out — but for all the wrong reasons. Lane changes came too quickly, it accelerated too late, and too aggressively, and I often found myself bracing for collisions that (in fairness) never came.
More than once in those years I’ve wondered if maybe I’d just got it wrong back in 2014. That the tech was so new, and I had been so wow’ed by it initially, that I had got swept up in the hype of self-driving cars … but that drive in my wife’s XC90, back-to-back as it was with the Civic Hybrid, showed me that wasn’t it. Instead, I just didn’t like the way those other cars drove. Just like I don’t like the way HOKAs feel. And, just like my wife isn’t wrong for liking her gross marshmallow shoes (probably), I’m not wrong for preferring a more restrained digital co-pilot.
It’s a matter of fit, not fact — and that’s going to be a tough sell.
Everyone but me is wrong
Classic Carlin bit.
As the great George Carlin once asked, “Have you ever noticed that anyone who is driving slower than you is an idiot, and anyone driving faster than you is a maniac?”
ADAS systems live squarely in that same subjective space occupied by other drivers. If the bots brake too hard, steer too sharply, or get too close to the car head before changing lanes, they might not be technically doing anything wrong, but they’re maniacs – and right now, there’s no real way to know how one car’s ADAS is going to behave until you’ve spent some significant time behind the wheel. Like, “Uh-oh. I bought a thing and I hate it,” amounts of time.
That’s a problem for both buyers and sellers (to say nothing of manufacturers and software developers), because why would you risk demonstrating a system that might scare someone? How do you sell “confidence” and “convenience” when what feels confident and convenient to one driver feels reckless to another, and milquetoast to a third?
Lucky for you guys, I have a solution.
Jojo’s ADAS scorecard *
System
Lane centering bias
Lane change distance (car lengths)
Follow distance (default)
Braking force (max Gs)
Hands-off time allowed
Overall “feel”
Ford BlueCruise
Centered
~3.5
Moderate
0.30 G
Medium
Stable
Honda Sensing
Slight left bias
~2.5
Safe
0.35 G
Short
Balanced
Mercedes-Benz DRIVE PILOT
Centered
~3.5
Moderate
0.40 G
Long
Confident
Tesla Autopilot
Centered
~1.5
Close
0.45 G
Long (varies)
Aggressive
Volvo Pilot Assist
Slight right bias
~3.0
Moderate
0.30 G
Moderate
Cautious
NOTE: THESE ARE NOT REAL VALUES
That asterisk (*) is there because these are completely made up, imaginary values. They’re simply there to illustrate one way for manufacturers and dealers to share objective, quantifiable information about how their different ADAS systems behave. If it’s done right, it might help a car shopper get a better feel for how their next car might drive, and prevent them from spending their hard-earned cash on a car that drives like an idiot. Or a maniac.
That’s my take, anyway – what’s yours? Head down to the comments and let us know what values you’d like to see represented on an ADAS scorecard, and how much you’d be willing to base your next car buying decision on how it drives.
As for me, my X handle might be VolvoJo, but if I’m shopping for a car that’s going to drive me instead of the other way around, I might have to see if “HondaJo” is available.
If you’re considering going solar, it’s always a good idea to get quotes from a few installers. To make sure you find a trusted, reliable solar installer near you that offers competitive pricing, check out EnergySage, a free service that makes it easy for you to go solar. It has hundreds of pre-vetted solar installers competing for your business, ensuring you get high-quality solutions and save 20-30% compared to going it alone. Plus, it’s free to use, and you won’t get sales calls until you select an installer and share your phone number with them.
Your personalized solar quotes are easy to compare online and you’ll get access to unbiased Energy Advisors to help you every step of the way. Get started here.
FTC: We use income earning auto affiliate links.More.
Elon wants the US military to start buying Tesla Cybertrucks – and now they are! The Air Force has ordered two Cybertruck testers for target practice to determine how easy they are to blow up, while Jo makes up a whole new conspiracy theory on today’s explosive episode of Quick Charge!
Today’s episode is brought to you by retrospec—makers of sleek, powerful e-bikes and outdoor gear built for everyday adventure. Electrek listeners can get 10% off their next ride until August 14 with the exclusive code ELECTREK10 only at retrospec.com.
An it doesn’t stop there. We’ve also got exciting new home battery backup and V2X options for Tesla owners, and one Texas EV driver that decided to conquer the Texas floodwaters by harnessing the awesome combined powers of electrons and stupidity (it’s pretty awesome).
New episodes of Quick Charge are recorded, usually, Monday through Thursday (most weeks, anyway). We’ll be posting bonus audio content from time to time as well, so be sure to follow and subscribe so you don’t miss a minute of Electrek’s high-voltage daily news.
Got news? Let us know! Drop us a line at tips@electrek.co. You can also rate us on Apple Podcasts and Spotify, or recommend us in Overcast to help more people discover the show.
If you’re considering going solar, it’s always a good idea to get quotes from a few installers. To make sure you find a trusted, reliable solar installer near you that offers competitive pricing, check out EnergySage, a free service that makes it easy for you to go solar. It has hundreds of pre-vetted solar installers competing for your business, ensuring you get high-quality solutions and save 20-30% compared to going it alone. Plus, it’s free to use, and you won’t get sales calls until you select an installer and share your phone number with them.
Your personalized solar quotes are easy to compare online and you’ll get access to unbiased Energy Advisors to help you every step of the way. Get started here.
FTC: We use income earning auto affiliate links.More.
Tesla’s Dojo supercomputer project is reportedly over. Bloomberg reports that CEO Elon Musk is killing the project after a mass exodus of talent from the Dojo team to a competing startup.
Dojo was the name of Tesla’s in-house AI chip development to create supercomputers to train its AI models for self-driving.
Tesla hired a bunch of top chip architects and tried to develop better AI accelerator chips to rely less on companies like NVIDIA, AMD, and others.
For the last few years, Peter Bannon, who worked with Keller for years, has been leading Tesla’s chip-making programs, but he is now reportedly also leaving the automaker.
Bloomberg reports that Musk has “ordered the effort to be shut down.”:
Peter Bannon, who was heading up Dojo, is leaving and Chief Executive Officer Elon Musk has ordered the effort to be shut down, according to the people, who asked not to be identified discussing internal matters. The team has lost about 20 workers recently to newly formed DensityAI, and remaining Dojo workers are being reassigned to other data center and compute projects within Tesla, the people said.
DensityAI is a new startup currently in stealth mode, founded by several former Tesla employees, including Venkataramanan.
It reportedly plans to build chips for AI data centers and robots, much like the Dojo program.
The company recently hired 20 former Tesla employees who worked on Dojo.
While the program appeared to be lagging behind for years as Tesla increasingly bought more compute power from NVIDIA, Musk has been claiming progress.
The CEO said in June:
Tesla Dojo AI training computer making progress. We start bringing Dojo 2 online later this year. It takes three major iterations for a new technology to be great. Dojo 2 is good, but Dojo 3 will be great.
During Tesla’s quarterly conference call in late July, the CEO claimed that Dojo 2 will be “operating at scale sometime next year.”
Electrek’s Take
It’s unclear whether the report is accurate or if it’s an extrapolation from the talent exodus to Elon killing Dojo, or if Elon was lying just a few weeks ago.
Alternatively, this development may be so recent that Elon went from being confident in Dojo a few weeks ago to disbanding the team working on it now.
Either way, I think it’s clear that the project has been lagging, and Tesla has been extremely dependent on chip suppliers rather than making its own.
I think Dojo being likely dead is not a big loss for Tesla.
When it comes to chip making, developing its own inference compute for onboard “AI computers” was always the more important project.