When Rivian's R1 pickup hit the market in 2021, RJ Scaringe says it launched with technology he now considers a dead end. The category includes software architecture and electrical decisions the company has spent years working around. Four years, 175,000 vehicles, and nearly $25 billion in cash burn later, the MIT-trained mechanical engineer with a PhD focused on internal combustion engines is staking everything on a second chance: the R2, a smaller, more affordable SUV designed to be the volume model the R1 never was.
In a long on-record sit-down with WIRED, Scaringe framed the moment in unusually blunt terms. According to WIRED's reporting, Rivian lost $3.6 billion in 2025 alone, on top of roughly $25 billion in cumulative cash burn across the company's life. The 2021 IPO was the largest of that year worldwide and briefly pushed Rivian's valuation past $100 billion. The stock has since collapsed from a high of about $130 to roughly $16. The company delivered its first vehicles in 2021, debuted publicly at the 2018 LA Auto Show after nearly a decade in stealth, and has since inked joint ventures that have changed its cost structure.
The R2 is the bet. It is Rivian's planned smaller SUV, priced below the R1T pickup and R1S SUV that launched the brand, and it is the vehicle the company says it has been designing around the lessons of the first one. Scaringe's central admission, as he put it to WIRED, is that the R1 shipped with electrical and software architecture the company now treats as a mistake. The years since have been spent building around that foundation rather than on top of it, and the R2 is meant to be the first Rivian that does not carry the inheritance.
That framing is more useful than the competitive theater around it. Scaringe is blunt about Tesla, about the Cybertruck's commercial failure, about the Ferrari Luce and the design world's recent attempts to rethink the car interface. None of those comparisons are the spine of the story. The story is that the founder of a publicly traded automaker, eight years and $25 billion into his attempt, is willing to say on the record which of his early calls he would make differently, and to bind that admission to a specific product launching in a specific window.
The competitive reality, as Scaringe lays it out, is mostly China. Tesla still sells the most volume in the United States, with roughly 8 million vehicles delivered in the same window that Rivian sold its 175,000, but the threat Scaringe names is Chinese EV makers scaling into global markets at a cost basis that legacy automakers have not matched. The R2 has to be the vehicle that demonstrates Rivian can build at a price and a cadence the company has not yet hit.
The buttons-versus-touchscreens debate is the most human-scale window into how Scaringe thinks about that problem. The industry has spent the last decade moving cabin controls into touchscreens. Scaringe argues, in the interview, that the pendulum has overshot and that physical controls for the most-used functions are a usability and safety choice rather than a nostalgia choice. It is a small debate on its face, and a useful one for a reader who is not tracking electrical architectures. It is also a way of hearing how Scaringe talks about product decisions in general. He is willing to name a consensus, say the consensus is wrong, and ship the opposite. The R2's claim is that the same willingness is being applied to architecture that no one outside Rivian's engineering org will ever see.
The financial scaffolding behind the bet has changed shape in the last two years. In 2024, Volkswagen Group committed up to $5.8 billion to co-develop software and electrical architecture with Rivian through a joint venture, as WIRED reports. This year, Uber committed up to $1.25 billion to build and deploy up to 50,000 fully autonomous robotaxis with Rivian. Both deals are real money against the $3.6 billion 2025 loss and the eight-year cumulative burn, and both shift Rivian's center of gravity away from being a pure consumer-truck company. The robotaxi deployment, in particular, gives the R2 platform a second life as a fleet vehicle that the R1T pickup never had.
What to watch is what Scaringe says the R2 has to do, in his own words. The vehicle has to land at a price that opens a market the R1 never reached, build at a rate Rivian has not previously sustained, and do both while the stock trades for a fraction of its 2021 peak. Scaringe's argument, as he put it to WIRED, is not that Rivian has earned the right to a second chance. It is that the company has named, specifically, what it got wrong the first time, and has built the next product to make the second attempt on different terms. The R2 is the test of whether that accounting is honest or promotional. Either way, the founder is now on the record about which it has to be.