Flying Cars' Oculus Moment

This past Monday, Joby Aviation raised a $590 million Series C round of funding, with the stated goal of developing an electric air taxi service. The round was led by Toyota, who is joining as a strategic partners to help the startup scale up manufacturing operations for its 5-seater electric vertical-takeoff-and-landing (eVTOL) aircraft . This is roughly the same amount of money that Lilium, who is developing a similar 5-seat eVTOL, was reportedly looking to raise in late 2019.

This is the largest publicly-known investment in the nascent eVTOL industry. In my opinion, in the future this will be seen as what “Facebook buys Oculus” was to VR, and “GM buys Cruise” was to self driving cars: not a guarantee that the technology is (or ever will be) ready for mass adoption, but the beginning of massive R&D investments by industry incumbents looking for the Next Big Thing™️. Uber Elevate has been banging the drum and releasing renders for a few years now, but Toyota has finally put its money where its mouth is.

Not All Mega-Rounds are Created Equal

Honestly, it’s an easy time to be skeptical of Japanese conglomerates investing massive amounts of money into startups. Softbank, the Japanese holding company whose $100 billion Vision Fund has been fueling capital-thirsty startups for the past several years, has seen its portfolio companies announce waves of layoffs and valuation haircuts over the past 9 months. The schadenfreude is sure to repeat itself in a few months once we start to get trailers for the WeWork movie and Uber TV series.

$590 million is a huge, Softbank-sized investment which almost certainly places Joby in unicorn territory. Though the birth of a flying car unicorn (AKA Pegacorn) seems to be a perfect symbol for tech industry hubris, I’d wager that this is more fundamentally sound than the high-profile Softbank investments.

Most high-profile tech company implosions in the past couple years, with the notable exception of Theranos, were a result of relying on unprofitable customer acquisition to fuel rapid growth, or “losing money on every sale, but making up for it in volume.” WeWork was the most egregious example of scaling up a business with negative unit economics to dizzying heights, but gig economy businesses like Uber and Doordash have relied on heavy promotional discounts to juice growth.

Hard tech investments like Joby and Cruise take the opposite approach. Instead of spending on customer acquisition, the money goes towards expensive R&D which will theoretically lead to a business with good unit economics provided that the companies are able to overcome the challenges necessary to bring a novel technology to market. Obviously, this comes with a different set of risks.

Classifying Risk

Risk for startups has traditionally been categorized into two categories: market risk (Do people want this?) and execution risk (Can a given team build this?). For hard tech, I like to split execution risk into two: operational risk and technical risk.

Operational risk is the question of whether a company can accomplish the hard, but not unprecedented, work required to build a particular business. Take SpaceX for example: many doubted that SpaceX could successfully build a company that could reliably launch rockets, but 30 years after the Apollo program, no one doubted that it was possible on a theoretical level. In Tesla’s case, it was obviously possible to build a car company, but it remained a daunting challenge: Tesla was the first American automaker to go public in more than 50 years.

Technical risk is what most people think when they hear “hard tech”: the risk that something cannot be built at all, limited either by biology (Theranos, Google’s glucose sensing contact lenses), physics (uBeam), or today’s technology (maybe: most self-driving car companies in 10 years). Helion Energy, a startup working on nuclear fusion, is a good example of a company still working on solving for technical risks.

Hard tech companies tend to have less market risk; it’s easy to imagine someone wanting a self-driving Uber ride for half the cost, or an affordable 10 min eVTOL ride from San Francisco to Palo Alto. The steep technological and operational risks answer the question of “why hasn’t anyone else already done this?”

Which Risks Remain?

Sebastian Thrun, founder of Waymo turned CEO of Kitty Hawk, has declared that we’ll have self-flying cars before self-driving cars (Full disclosure: I work for Kitty Hawk).

I think he’s correct here, and most of the risks of deploying eVTOL aircraft are operational, rather than technical. Unlike self-driving-cars, where there’s still significant questions if the machine learning-based approaches can reach the safety levels of a human driver, electrifying aircraft more closely resembles the transition from gasoline cars to EVs. Instead of one, possibly insurmountable technical “great filter”, there are a handful of difficult, but obviously possible operational challenges that stand between the present and widespread adoption.

Make no mistake: the operational risks are still massive. Certifying an aircraft, especially a new category of transition vehicles, is a massive undertaking, that will require large skills in working with regulators, systems engineering, and huge amounts of testing. This has all been done before by incumbent aerospace companies, but it is a wickedly difficult problem for newcomers.

Joby and other eVTOL companies are also in for their own version of Elon Musk’s infamous “manufacturing hell.” All of these designs rely heavily on advanced carbon fiber structures to reduce weight; and the story of Icon Aircraft provides a great case study at how that can add huge amounts of complexity and cost to the manufacturing process. This is on top of the challenges of transitioning a company from a prototyping organization to one with enough process, documentation, and testing to enable repeatable, safe assembly. Money aside, this is clearly Toyota’s biggest value-add to Joby.

Random Thoughts

  • One perk of the fundraising announcement is that Joby’s website now has some good photos of their vehicle. Unfortunately, still no flight test videos. The team photo gives a good sense of the aircraft’s scale.

  • They have some pretty bold performance claims: 200 mph speed and 150+ miles range. For comparison, Lilium claims a 186 mile range at 186 mph for their similar 5-seater, and Kitty Hawk advertises a 100 mile range for its single-seat Heaviside.

  • Speaking of eVTOLs, Uber and Hyundai are still talking up their eVTOL partnership originally announced at CES. I take all of Uber Elevate’s partnerships with a heavy grain of salt, but it’s interesting to see all the Asian automakers announce make flying car announcements in such a short timespan. Maybe Honda will end up leading Lilium’s $500 million mega-round?

  • EASA released a proposal for electric propulsion unit certification. Obviously, very important for eVTOLs. Thanks to Xin Gou for bringing this to my attention; he’s a great aggregator of electric aviation news. (On the topic of Twitter, here’s mine. Please feel free to DM me any interesting news!)