Uh-Oh: Millions Of Teslas Might Not Be Hardware-Compatible With Unsupervised FSD

Uh-Oh: Millions Of Teslas Might Not Be Hardware-Compatible With Unsupervised FSD

Tesla has long been promising an autonomous future. Since 2016, in fact, when it promised that all cars were being built with hardware capable of full self-driving. It turns out, that might not have been true, because now Tesla isn’t sure if cars built as early as last year will be capable of achieving fully-autonomous driving due to hardware limitations. Uh-oh.

Welcome back to Critical Materials, your daily roundup for all things EV and automotive tech. Today is a Tesla-centric day—we’re chatting about Tesla’s gamble on Hardware 3, the now-profitable Cybertruck, and the reason why Tesla killed the $25,000 EV (officially, this time). Let’s jump in.

30%: Millions Of Teslas On The Road Might Not Get Unsupervised FSD

Tesla Hardware 4.0 (HW4) Autopilot And Self-Driving Computer (FSD2)
@greentheonly (Twitter)

Tesla Hardware 4.0 (HW4) Autopilot And Self-Driving Computer (FSD2)

Tesla CEO Elon Musk isn’t so sure that the automaker will be able to deliver unsupervised Full Self-Driving to millions of Teslas already on the road.

In a statement made during Tesla’s third-quarter financial call, Musk let slip something that owners of Hardware 3-equipped Teslas have been fearing for some time. It turns out that the automaker can’t provide a clear answer on whether or not cars manufactured in early 2023 and beyond will actually be able to achieve driverless autonomy.

Here’s Musk answering a question from an investor regarding Hardware 3:

So the answer is we’re not 100% sure [if HW3 will work with unsupervised FSD]. Hardware 4 has really several times the capability of Hardware 3. It’s easier to get things to work, then it takes a lot of effort to squeeze that [into] Hardware 3. And there is some chance that Hardware 3 does not achieve the safety level that allows for unsupervised FSD. There was some chance of that.

Musk did, however, commit to upgrading some HW3-equipped cars—not all—to a newer version of Tesla’s Full Self-Driving computer, despite the automaker previously noting on a recently-deleted blog post that all vehicles built after October 2016 have the necessary hardware to achieve Full Self-Driving.

“If that turns out to be the case, we will upgrade those who bought hardware 3 FSD for free. And we have designed the system to be upgradable,” said Musk, despite previously calling the upgrade ‘not economically feasible’ in 2022.

“So it’s really just to switch out the computer. The cameras are capable. But, we don’t actually know [if HW3 will work with unsupervised FSD]. But if it does turn out, we’ll make sure we take care of those who have bought FSD on Hardware 3.”

Here’s the conundrum—Tesla has committed to a no-cost upgrade for the customers who actually bought FSD on Hardware 3, but not for all HW3 vehicles.

Keep in mind that FSD, at its peak, was priced at $15,000. Today, it’s available either as a subscription for $99 per month or an outright purchase of $8,000. Tesla’s CEO marketed its cars with FSD as “an appreciating asset,” meaning that a customer can expect to purchase FSD at any time—whether it be through the subscription model or by outright purchasing the software—and expect the same feature functionality as those who may need an upgrade to HW4.

The world has known that HW3 has been nearing its limits for some time. And if Tesla is now unable to meet its commitments, it could land the brand in hot water, perhaps even triggering Dieselgate-level consumer protection efforts by government agencies if enough consumers lodge legal complaints about what was delivered versus the intent of what was promised.

And, if we’re being frank, it’s not a great look for a company that is betting its future on the public trusting its ability to deliver autonomy.

Stay tuned on this one because it could get a lot more complicated in the coming months.

60%: Tesla Cybertruck Has Turned A Profit

Tesla Cybertruck at a Tesla Supercharging station

It’s been less than a year since Tesla’s shiny electric pickup has hit the streets, yet it’s already managed to turn a profit.

Tesla revealed in its quarterly earnings report that the Cybertruck has “achieved a positive gross margin for the first time” during the third quarter of 2024. A rather impressive feat, if you think about it, considering the fairly small number of units sold compared to the 3 and Y program—and in the midst of a cut-throat American truck market where the stainless steel cheese wedge sticks out like a sore thumb.

 

Part of the profitability could be Tesla’s push for the Foundation Series, which tacked on a hefty $20,000 early adopter’s tax. Between November 2023 and October 2024, Tesla sold around 30,000 of these vehicles, meaning it raked in somewhere around $600 million thanks to just the Foundation Series branding alone.

Tesla officially scrapped the Foundation Series earlier this month, lowering the price of the Cybertruck to $79,990 for the All-Wheel Drive variant or $99,990 for the performance-oriented, tri-motor Beast trim.

Tesla also made a smart move by securing what was essentially an interest-free loan crowdsourced by folks who put $100 down on the truck when it was announced back in 2019. Reportedly, 2 million people forked out the cash to reserve the truck, which gave Tesla $200 million in interest-free funding for the program.

As for actual earnings, well, Tesla may have raked in anywhere between $3 billion and $3.6 billion in Cybertruck sales in the past 11 months. That’s around 4.6% of its automotive revenue since the truck launched. Keep in mind that Tesla sold more than 1.7 million Model 3 and Y since Q4 2023, meaning that the Cybertruck—which sold just 1.7% of that volume—carries a significantly higher margin than Tesla’s more affordable mass-market cars.

Here’s the downside: conversion rates aren’t exactly great. It’s estimated that only 2.5% of reservation holders are actually picking up a truck, meaning that the automaker has likely burned through its entire reservation list already. The remaining buyers could be waiting for Tesla to launch its more affordable version of the truck, which was initially expected to be under $40,000. It’s not clear when Tesla will launch a more affordable version or how much money it could rake in, but if Tesla is at least profitable on its truck at this point in the game and buyers aren’t biting at current pricing, it shouldn’t be too long before we see some sort of movement.

90%: Autonomy Killed The $25,000 Tesla

Tesla Model Y Slimmed Down $25,000

Remember back in April when Elon Musk said Reuters was lying when it reported that the $25,000 affordable Tesla EV was dead? It turns out that ol’ Musky boy may have been overstating things a bit.

As we learned in the quarterly earnings report, Tesla won’t be making a new standalone, human-operated $25,000 EV. The outcome, according to Musk, would be “pointless” and “silly.” So the dream of a new, non-Robotaxi, sub-$30,000 EV is officially dead at Tesla.

Let’s look back at the history of what happened here to piece things together. Back in February 2023, Musk’s lieutenants held a meeting where they pitched the cheap “Model 2.” Codenamed NV91 (or, “New Vehicle 91”), the car was described as a slimmed-down Model Y and would target that coveted $25,000 price bracket for mass affordability.

During a follow-up meeting that same month, the same staff shared another conceptual product, NV93, or as it’s better known today: the Robotaxi. The idea wasn’t to have the company focus on the product, but instead to satisfy Musk’s appetite for future products. But it backfired, because Musk enamored by the idea and greenlit the project. This killed the NV91.

When investors learned of the Reuters report claiming that the affordable EV was cancelled, they voted with their wallets. Musk stopped the bleed by claiming that the outlet was lying, though yesterday’s investor call made it clear that Tesla has no intentions of delivering the product after all, despite investors clearly seeing a need to compete with low-cost alternatives entering the market from China.

So, what killed the car? It turns out the fatal blow was delivered by the promise of something that Tesla has yet to deliver on: full autonomy.

Musk says that its goal is focusing on reducing the cost per mile of transportation however possible. In typical Tesla fashion, this means slimming down a car with the fewest number of parts possible.

The robotaxi is a great example of this. Likely a tiny battery, no physical charge port, no pedals or steering wheel. It’s basically an ode to cost-cutting. And at the forefront of everything comes the promise of convenience—of getting in a car and controlling it from your phone alone. An effortless mode of transportation delivering on the promise of solving self-driving, which Tesla has been promising to deliver “next year” since 2016. But it’s really happening in 2025, according to Musk during yesterday’s quarterly earnings call. Really, this time. Really.

It just seems odd that Tesla really wants to focus on pushing this path with the sub-$30,000 Robotaxi. If the future is autonomous, and Tesla can make more money by ditching more interior parts, why not delete them from the Model 3 and Y since the original idea behind those cars were to deliver mass-market transit at an affordable price? It just seems wrong to completely kill off a potential line of customers in what seems like an effort to prove a point to the public. The $25,000 Tesla could have been so much more.

100%: Would You Have Bought A Driveable Cybercab?

Tesla Cybercab Robotaxi
InsideEVs

It’s really a shame. A few of us here at InsideEVs spoke about the potential of Tesla having to absolutely knock it out of the park with a vehicle built on the Cybercab platform. I mean, it’s hard to deny that the thing looks pretty cool—like a Cybertruck, but without the sharp edges and major DeLorean vibes.

Tesla also doesn’t have a coupe offering on the market right now, and the next-gen Roadster (whenever that comes out) won’t be in a reachable price bracket for most folks. Come to think of it, no automaker has something like this today. So offering something like this up to the mass market could have been a huge win for Tesla that the automaker is just giving up on. And that feels like a damn shame.

Here comes the question: would you have bought a $25,000 version of the robotaxi if it had a steering wheel and pedals? Let me know in the comments.

Uh-Oh: Millions Of Teslas Might Not Be Hardware-Compatible With Unsupervised FSD

The⁤ world of autonomous driving technology has been evolving at‍ a rapid pace in recent years, with Tesla leading the charge with⁤ its Full⁣ Self-Driving (FSD) capabilities. ‌However, a recent revelation by the electric car manufacturer has raised concerns among Tesla owners and enthusiasts alike.

Tesla CEO Elon Musk ‍recently announced that the company’s new unsupervised Full Self-Driving (FSD)‍ software, which ⁣is set⁣ to be ⁢released later this‍ year,⁢ may not be compatible ‌with millions of older Tesla ​vehicles. The reason for this incompatibility lies ⁣in⁢ the hardware‌ limitations of the older​ vehicles, particularly those built before 2019.

The hardware in‍ question ⁤is Tesla’s Full Self-Driving Computer (FSD Computer), which is‍ responsible for processing the massive amounts of data gathered by the car’s sensors and cameras to enable autonomous driving capabilities. The older vehicles are equipped with the Hardware⁢ 2.5 or even the earlier Hardware 2.0 computers, which‍ may not ⁣have the processing power⁣ or capabilities ​required for the unsupervised ​FSD software.

This revelation has sparked concerns among ⁤Tesla owners⁢ who have already purchased the ⁣Full Self-Driving package, as ​they may not be ⁢able to fully ​utilize the features of the new⁣ software. ⁤Many are ‍worried that they may have to upgrade⁣ their hardware, which could come ​at a hefty⁤ cost.

Furthermore, this development raises questions about Tesla’s promises of full‌ self-driving capabilities for all its vehicles. If millions of older Tesla vehicles are not compatible with the new unsupervised FSD software, then the promise of autonomous driving for everyone may⁢ not be achievable in the near ⁤future.

In light of these concerns, Tesla has yet to provide a ‌concrete solution​ for owners ⁤of older‍ vehicles who may not be​ able to take‍ advantage of the unsupervised FSD software. The company has stated that it ⁣is exploring potential options, such as offering hardware upgrades at⁣ a discounted ⁤rate or providing alternative‍ solutions ​for affected customers.

As‍ the automotive industry continues to push the boundaries of autonomous driving‌ technology, it‍ is crucial for companies⁢ like Tesla to ensure that their promises align with the capabilities​ of ⁢their existing hardware. The⁢ revelation that millions of Teslas may⁢ not be hardware-compatible with unsupervised FSD serves as a stark ‌reminder of the ⁤challenges and​ limitations that come ⁢with pushing ⁣the ‌boundaries of technology.

In conclusion, the news that millions ‌of ‍older Tesla vehicles may not ‌be compatible with ​the new unsupervised Full Self-Driving software raises concerns about the company’s​ promises ‍of‍ autonomous driving for all. It is imperative⁣ for Tesla to address these issues promptly and‌ provide solutions for ⁤affected customers to ensure that the promise of full self-driving capabilities remains a reality for ‍all Tesla ⁢owners.