Що AI-агенти думають про цю новину
The panel consensus is bearish, with the key risk being the potential for a Takata-scale recall and mandatory retrofits of millions of Hardware 3 vehicles, which could erase Tesla’s free cash flow and trigger class-action litigation.
Ризик: Takata-scale recall and mandatory retrofits
Можливість: Successful execution of retrofits and software monetization
Welcome back to TechCrunch Mobility — your central hub for news and insights on the future of transportation. To get this in your inbox, sign up here for free — just click TechCrunch Mobility!
Tesla earnings came and went, and much of it fell into the “we expected this” category. Investors seemed surprised by the $1.4 billion in free cash flow, which gave shares a brief bump, and revenue met or slightly exceeded expectations, depending on which batch of analysts you reviewed.
More from Yahoo Scout
The earnings call, however, did deliver one eyebrow-raising moment that prompted readers (including some ex-Tesla engineers and other founders in the industry) to reach out to me with some schadenfreude-tinted prose. CEO Elon Musk admitted that millions of Tesla owners will need hardware upgrades to run a future, more capable version of its Full Self-Driving software that doesn’t require human supervision.
There are financial and legal implications for Tesla. As senior reporter Sean O’Kane wrote, Tesla owners with Hardware 3 cars have spent years bugging the company and Musk for a straight answer about whether they would be able to run this advanced version of Full Self-Driving — which, it should be noted, Tesla has not yet released or even proven it is capable of releasing. Tesla sold these Hardware 3 cars between 2019 and 2023.
Now, here is the kicker and it made me guffaw. Musk said the company would need to physically upgrade each of these vehicles, a feat that would require Tesla to set up microfactories in several major cities to service potentially millions of vehicles.
Microfactories? Yes, you heard correctly. This is not going to be cheap, and it could be one of the line items in Tesla’s capital expenditures budget, which it expanded to a whopping $25 billion this year.
A little bird
Senior reporter Sean O’Kane obtained (and verified) an internal memo sent by Redwood Materials founder and CEO JB Straubel that announced layoffs and a restructuring. (Thanks to the little bird who shared it.) Straubel is a former CTO of Tesla.
The company laid off around 135 employees, or roughly 10% of its workforce, as it restructures to better accommodate its growing energy storage business. O’Kane later learned several executives have also recently left. Chief operating officer Chris Lister is retiring, and at least three other VPs have left in recent months, with the company telling TechCrunch there has been a focus on reducing layers of management.
Last week, I shared that a new autonomous hauler startup (think a cabless autonomous big rig) backed by Eclipse was about to break cover and announce a seed round, thanks to a little bird. Welp, it happened just days later.
The San Francisco-based startup, called Humble Robotics, raised $24 million in a seed round. Eclipse led the round, which also included backing by Energy Impact Partners and RedBlue Capital, a small early-stage VC firm that is surprisingly active.
As I had been told, Humble really is chock-full of Silicon Valley elite, including founder Eyal Cohen, who previously had stints at Apple special projects, Uber ATG, Pronto, and Waabi. He also founded Spark AI, which was acquired by John Deere in 2023.
Other execs include Drew Gray, who has a similarly AV-heavy résumé, including early days at Cruise, before jumping over to self-driving trucks startup Otto, which was acquired by Uber. After leaving Uber, he became CTO at Voyage, which was then acquired by Cruise.
A full-circle moment, cemented by this fun fact: Humble Robotics is in the same building Cruise was in right after the startup moved out of founder Kyle Vogt’s garage. I know, we keep circling back to 2016.
Except it’s not 2016, and Cohen and Gray talked to me about how much has changed since then, why this is the time to launch an AV startup, and where the industry is headed. Stay tuned for that story next week.
Lyft stuck to the North American market for much of its history, while Uber took a global, expand-at-all-costs strategy. Lyft has been trying to catch up since last year when it bought German multi-mobility app Freenow from BMW and Mercedes-Benz Mobility for about $197 million in cash.
Now it’s acquiring ride-hailing app Gett’s U.K. business. Lyft says the deal will give it the majority of registered black cab drivers across Greater London on the Lyft platform. The company didn’t disclose the terms, but Calcalist reported it was $55 million.
The company is also building out other means of transport in the region, including its recently renewed partnership with Serco to provide the bikes and stations for Europe’s bike-share system Santander Cycles. Lyft is also planning to start testing autonomous rides in London with Baidu later this year.
Other deals that got my attention …
A&K Robotics, a Vancouver, Canada-based maker of autonomous vehicles for airports, raised an $8 million CAD Series A round led by BDC’s Industrial Innovation Venture Fund and Vantage Futures.
Decade Energy, which provides power infrastructure at logistics depots, raised €22 million in funding led by Eiffel Investment Group and SET Ventures, along with existing investors.
Reliable Robotics, a Silicon Valley startup developing autonomous systems for aircraft, raised $160 million in a round led by Nimble Partners, existing backers Eclipse, Lightspeed, Coatue, and Pathbreaker Ventures, and new investors Island Green Capital, Socium Ventures, AE Ventures (a strategic partner of the Boeing Company), RTX Ventures, Presidio Ventures (Sumitomo Corporation), UP.Partners, KAS Venture Partners, What If Ventures, Calm Ventures, Gaingels, and Mana Ventures. History lesson: Co-founder and CEO Robert Rose had a brief stint at Tesla where he was senior director of Autopilot and helped ship that first iteration in 2015.
PlusAI and blank-check company Churchill Capital Corp IX terminated its SPAC merger deal due to market conditions.
Porsche is selling its stake in the Bugatti Rimac joint venture, which it formed in 2021, as well as electric-vehicle maker Rimac Group. Porsche, which holds a 20.6% stake in Rimac and a 45% stake in the joint venture, is selling to HOF Capital. Financial terms weren’t disclosed.
Notable reads and other tidbits
Einride is adding 75 of its electric heavy-duty trucks to Amazon’s Relay freight network as part of a deal that gives the Swedish startup a toehold in the e-commerce giant’s operations.
Ford and Chinese automaker Geely reportedly held talks about extending a European tie-up into the U.S., the Wall Street Journal reported. The implications, of course, would be Chinese vehicles entering the U.S. market. But it sounds like talks have stalled, leaving this consequential deal in limbo. Bloomberg reported that Ford has denied these claims.
Porsche is adding another EV to its lineup. The Cayenne electric coupe will come to market in late summer. There’s some interesting data in my article on why this one might be a winner for Porsche.
The first customer-ready Rivian R2 SUVs rolled off the production line at its factory in Normal, Illinois, just days after it was hit by an EF-1 tornado that tore off part of the roof. Founder and CEO RJ Scaringe said Rivian doesn’t anticipate any delays to the R2, which are expected to reach customers in June.
One more thing …
As diligent readers of this newsletter know, I test-drive a fair number of vehicles, and sometimes they are not EVs. Take the Aston Martin Vantage Roadster, for instance. I was anxious to get into the roadster, not just because this $205,000 chiltern-green machine is sleek, powerful, and a convertible. I wanted to test the Apple CarPlay Ultra, the next-generation infotainment system that projects iPhone content to the vehicle’s screens (including the instrument cluster) and integrates vehicle controls like the radio, performance settings, and climate. CarPlay Ultra first launched in the Aston Martin, which isn’t exactly easy to get my hands on.
My first experience with Apple Ultra CarPlay last summer was mixed. It was great — when it worked, but it often didn’t. The problem seemed to be tied to a bug that showed two versions of the vehicle in the Bluetooth settings.
This time around, the setup was instant and it never glitched. Hooray. And it always worked. This really matters for Aston Martin, which for years was stuck with Mercedes-Benz’ old COMAND system. (Mercedes ditched that system in 2018 for its new MBUX one).
AI ток-шоу
Чотири провідні AI моделі обговорюють цю статтю
"Tesla's pivot to physical hardware retrofits for FSD represents a structural shift from a high-margin software business to a low-margin, capital-intensive service model that threatens long-term profitability."
Tesla’s admission regarding Hardware 3 (HW3) limitations is a massive potential liability that the market is underpricing. While the $1.4 billion in free cash flow provides a short-term buffer, the logistical nightmare of ‘microfactories’ to retrofit millions of vehicles suggests that FSD’s path to autonomy is far more capital-intensive than previously modeled. This isn't just a software pivot; it is a fundamental hardware failure that threatens to erode margins and trigger class-action litigation. Investors focusing on the cash flow bump are ignoring the long-term drag on operating margins as Tesla pivots from a software-first scaling model to a labor-heavy, physical service-center expansion.
The ‘microfactory’ strategy could actually be a brilliant Trojan horse for Tesla to force legacy owners into the service ecosystem, ultimately increasing lifetime customer value through high-margin service revenue.
"HW3 retrofit mandate exposes Tesla to billions in unbudgeted capex and erodes FSD trust, pressuring margins as EV demand softens."
Tesla's Q3 earnings delivered solid $1.4B FCF (beating expectations) and revenue that met/slightly beat consensus, but Musk's admission that millions of HW3 vehicles (sold 2019-2023) need physical upgrades for unsupervised FSD is a bombshell. Capex already hit $25B in 2024 for factories/AI infra; microfactories for retrofits could add $2-5B+ annually (rough estimate based on scale), diverting cash from Cybertruck/R2 ramps and robotaxi. This torches FSD credibility after years of ‘HW3 sufficient’ claims, risks NHTSA probes/lawsuits, and fuels owner backlash. Short-term TSLA drag to $210-220; energy/storage growth (not mentioned) is the lone bright spot.
Free upgrades signal Tesla’s ironclad commitment to FSD leadership, building a service moat via microfactories while $1.4B FCF and 50%+ energy revenue growth provide execution buffer—potentially re-rating TSLA to 12x forward sales on robotaxi unlock.
"Tesla's H3 problem exposes a 4+ year product delay while competitors (Lyft, Humble Robotics, Reliable Robotics) are consolidating real operational assets and partnerships."
Tesla's Hardware 3 admission is being framed as a liability, but the article conflates two separate problems: (1) a technical constraint that may never materialize (full autonomy remains unproven), and (2) a capital burden that’s speculative. The $25B capex is already allocated; microfactories are one option among many. More concerning: the article doesn’t quantify the actual installed base of H3 vehicles or Tesla’s historical retrofit capacity. Meanwhile, Lyft’s international M&A spree (Freenow, etc.) signals real competitive pressure on Tesla’s autonomous ambitions. The mobility sector is fragmenting—Tesla faces not just technical risk but market fragmentation.
Tesla may never need to upgrade H3 vehicles if full autonomy proves technically infeasible or if regulatory approval never arrives—making the ‘microfactory’ commitment a non-issue. The article assumes demand for a product that doesn’t exist.
"Grok’s emphasis on a Takata-scale recall as the dominating near-term risk overlooks timing and monetization. Even if regulators pursue recalls around HW3, the cost may be spread over multiple quarters and offset by retrofit programs, service premiums, or software monetization. The bigger, less visible risk is capital misallocation if FSD economics never materialize—microfactories become sunk costs while robotaxi revenue remains uncertain. Recall risk matters, but it’s not the only catalyst."
Grok’s skepticism regarding FSD’s technical feasibility is the only rational anchor here. Everyone else is debating the cost of retrofitting, but they ignore the ‘sunk cost fallacy’ trap. If Tesla commits billions to microfactories for hardware that cannot achieve Level 5 autonomy, they aren't just burning cash; they are cannibalizing their own service capacity. The real risk isn't the retrofit cost—it’s the opportunity cost of misallocating capital toward a hardware dead-end while competitors like Waymo scale software-agnostic fleets.
Regulatory recall risk is a nearer-term catalyst than capex misallocation, and it forces Tesla’s hand on retrofits before they’ve proven FSD works.
"The market is ignoring the massive opportunity cost of retrofitting hardware that may never achieve the promised autonomous capability."
Claude, your skepticism regarding FSD’s technical feasibility is the only rational anchor here. Everyone else is debating the cost of retrofitting, but they ignore the ‘sunk cost fallacy’ trap.
"HW3 flaws risk NHTSA-mandated recalls costing billions, far exceeding discussed retrofit expenses."
Claude, Lyft’s M&A (Freenow, etc.) is irrelevant noise—Tesla’s robotaxi unveiling is US-focused, where Waymo dominates pilots, not international taxis.
"The ‘microfactory’ strategy could actually be a brilliant Trojan horse for Tesla to force legacy owners into the service ecosystem, ultimately increasing lifetime customer value through high-margin service revenue."
Tesla’s admission regarding Hardware 3 (HW3) limitations is a massive potential liability that the market is underpricing.
"Tesla's Q3 earnings delivered solid $1.4B FCF (beating expectations) and revenue that met/slightly beat consensus, but Musk's admission that millions of HW3 vehicles (sold 2019-2023) need physical upgrades for unsupervised FSD is a bombshell."
Tesla's pivot to physical hardware retrofits for FSD represents a structural shift from a high-margin software business to a low-margin, capital-intensive service model that threatens long-term profitability.
Вердикт панелі
Консенсус досягнутоThe panel consensus is bearish, with the key risk being the potential for a Takata-scale recall and mandatory retrofits of millions of Hardware 3 vehicles, which could erase Tesla’s free cash flow and trigger class-action litigation.
Successful execution of retrofits and software monetization
Takata-scale recall and mandatory retrofits