Let’s be honest: the Tesla Cybertruck is great as a technological showcase, and for making viral YouTube videos. But $100,000 stainless steel trucks aren’t going to get Tesla to the promised land. That would be the holy grail of electric vehicles: one that can be built profitably and sold for around $25,000. And it may be coming sooner than we think.

News of Tesla’s long-awaited “Redwood” project leads this midweek edition of Critical Materials, our morning news roundup. Also on tap for today: we may get fewer Volkswagen ID.Buzz vans in the U.S. this year than we expected, and a look at why automakers struggle on the software front. Let’s dig in.

30%: Tesla ‘Redwood’ Comes To Light

We have known for a while that Tesla wants to go even lower than the Model 3 and Model Y’s price point to make a truly affordable EV for everyone—a vehicle that will possibly be, or share a platform with, a fully automated robotaxi, per CEO Elon Musk’s demands.

Today, we finally know a little bit more about this project’s future, and it may be happening soon. Reuters has a heck of a scoop, reporting Tesla is telling suppliers it wants to start production of a new compact crossover in the summer of 2025. 

Would this vehicle be the affordable model Musk and co. have talked about for so long? The Reuters dispatch doesn’t say that, exactly, but it is very heavily implied. Also, what else could it be? We’re not talking about the Roadster here, clearly: 

Tesla has told suppliers it wants to start production of a new mass-market electric vehicle codenamed “Redwood” in mid-2025, according to four people familiar with the matter, with two of them describing the model as a compact crossover.

Tesla CEO Elon Musk has long whetted fans’ and investors’ appetites for affordable electric vehicles and self-driving robotaxis that are expected to be made on next-generation, cheaper electric car platforms.

[…] Tesla sent “requests for quotes,” or invitations for bids for the “Redwood” model, to suppliers last year, and forecast weekly production volume of 10,000 vehicles, two of the sources said.

Production would begin in June 2025, three of the sources said. All spoke on condition of anonymity because the matter is confidential.

There’s a lot to speculate on here, but also a lot to unpack. 

We know that affordability is one of the biggest barriers to EV adoption right now, and outside of the Chinese automakers, nobody has really been able to pull of a truly great, higher-range sub-$30,000 EV without taking an absolute bath on profits. We also know from Walter Isaacson’s biography and Musk’s own repeated statements that Tesla wants to chase even more volume than it has now, and that it’s been mulling a $25,000-ish vehicle—possibly with Cybertruck-like visual cues—for years now. Musk has also said this car could be built in Texas, though many have speculated the planned Mexican factory is also in the cards. 

And this interesting tidbit from Reuters:

Tesla in recent years tore down a Honda Civic, whose price starts at $23,950 in the United States, to study how to make cheaper cars, two separate sources said. The next-generation Tesla architecture, internally called “NV9X,” will include two or more models, said the two people and one of the initial sources.

Is it possible one of those models is also the fully automated Tesla robotaxi? That’s unclear. But according to Isaacson’s book, Musk resisted putting resources into the $25,000 car, saying he wanted the team to focus on a vehicle that didn’t need pedals or a steering wheel—until they convinced him to build these two cars on the same platform and assembly line.

It’s also worth noting the obvious: Tesla is notoriously bad when it comes to product delays. A mid-2025 production date seems a little optimistic, to say the least. But if this report is accurate, things are moving. 

I’d also say that Tesla is unique among the non-Chinese automakers (for now, anyway) in its ability to produce such a car at scale, and if it can pull this off, it’s due to be an industry game-changer on a level beyond even the Model 3 and Model Y. 

60%: Only 20,000 VW ID.Buzz Models May Make It Stateside In 2024

Speaking of delayed products, it has now been seven (!!!) years since we saw the Volkswagen ID.Buzz’s original concept vehicle. Now, the U.S. may even just get the retro-styled electric van in 2024 after waiting so long; it went on sale in Europe last year.

But VW’s dealer council leader let it slip to Automotive News that this won’t be some volume-seller vehicle—at least, not right away: 

2024 seems like a big year from a product standpoint with the ID Buzz set to arrive in the second half of the year. What does the arrival of that vehicle mean for Volkswagen’s U.S. dealers?

Finally is one word to recap it. We’ve all been eagerly awaiting and we all have a list of customers. Some dealers have taken deposits already. The initial demand is absolutely there and the excitement is there to match it. While we thought the vehicle would be out maybe as soon as the very end of ’23, we’re pretty darn excited to have that car in our showrooms by the end of ’24.

Has Volkswagen given dealers any guidance on what type of volume it expects initially with the ID Buzz?

I think it will be in the 20,000 range. It sounds like we may have the opportunity for additional European capacity if we need it, but we’re still trying to sort out where the demand is going to be globally.

For context, that’s almost half as many ID.4 models were sold in the U.S. in 2023. I still think VW should’ve led with the van instead of the competent but unexciting ID.4, but hey, VW knows what it’s doing here. (Right?)

90%: Why Automakers Struggle With Software

Finally, CES this year was very much a car show, but a radically different one than in years past. The future of cars is likely to be far more defined by smartphone-like software features and autonomy than horsepower and mechanical specs. But very few automakers are truly great at making that stuff. Why is that?

Besides the obvious—the auto industry has spent 100 years doing one thing and now it must do a radically different thing—Wards Auto has a good explanation of what’s off here. Here’s longtime journalist John McElroy to explain: 

Traditional automakers are struggling with software because organizationally, they’re doing it all wrong, said Jan Becker, the CEO of Apex.ai. He says you need a small, highly efficient team of only eight to 20 exceptional software engineers (SWEs). And you’d better be ready to pay to get that rock-star talent: at least $1 million a year per engineer, because that’s the going rate for top talent in The Valley. Outsourcing software development to low-cost countries is not the answer, he says, adding that traditional OEMs focus too much on cost and need to focus more on value. Moreover, those SWEs need to operate outside the existing corporate organization. They need short development cycles. They need a single depository of all the data for all the vehicle lines an OEM has, i.e., no more “data lakes” strewn across the company. Also, they need to write in C++ or Rust, but definitely not in C.

Here’s another example!

ETAS, a subsidiary of Bosch, has software that allows OEMs and their tiers to develop software together. Called DEV OPS, it uploads everything to the cloud at the end of every day and evaluates it to eliminate bugs. This is the way the tech industry does it, ETAS says, developing software in slices. Traditional OEMs, on the other hand, tend to build the entire stack and start testing it. But that’s too slow. And by putting everything in the cloud, you look at a vehicle from a total systems standpoint, not just at individual components or systems.

That story is worth a read in full. Even the headlights are about to be software-defined! 

100%: How Do You Get Average Drivers Into The Software-Defined Vehicle?

We’ll have more to say about the Tesla “Redwood” project on InsideEVs today. 

For now, let’s talk about the third item: the wild software-driven features the automakers want to put in your next car, from downloadable apps to AI integration. I get why the car companies are trying to do this. And yet, when I tell normal, ordinary folks about it—you know, the kind of person shopping for a Subaru Crosstrek or something—they rightfully look at me like I am insane. And I think that will be a big hurdle for the car companies to overcome.

So how do they do that? How do they justify all this crazy tech stuff coming to cars in the next few years, and get people into it?