There is an elephant in the AV room, although most reporters pretend not seeing one. I’m an analyst not a reporter, so when I see one, I see it's my job to call it out...
There is an elephant in the automated vehicle room, although most reporters pretend there is none.
I’m an analyst not a reporter, so my learning curve about the journalistic profession has been steep. I find writing great fun — which is why I have been doing so much of it recently — but the experience of hitting “send” on a finished article turns out to be both exhilarating and oddly terrifying. Author and researcher Brené Brown calls this a vulnerability hangover.
But that is “send” not “publish” and in my case each piece goes off to a great person I have come to regard as my editor: Junko Yoshida. Writing and editorial judgement are two vastly different skills, so I just focus on the former and leave Junko to have the final (entire) say on the latter. This arrangement is working out really well for me so far.
When I heard Junko was featuring on a virtual panel from the Partners for Automated Vehicle Education (PAVE) called “Behind the Byline: Journalists on the Automated Vehicle Beat”, I signed up and watched it live. You can see the whole video below and the panel also features Tech Crunch’s Kirsten Korosec, Automotive News’ Pete Bigelow and is hosted by PAVE’s communications director Ed Niedermeyer.
I really enjoyed watching it and have a whole new understanding of the challenges of modern-day journalism in our 24/7, always-on, information-based economy. More than anything, I now understand the race to publish first and that being an electronics-educated analyst who has covered tech for over 25 years gives me an insight and grounding which most reporters writing about automated vehicles (AVs) can only dream of.
I once heard someone define news as: “Something that happened today that didn’t happen yesterday” and this has been a busy couple of days for news reporters covering AV. Let’s take a look:
Kirsten quotes Ola Källenius, head of Mercedes, saying: “Many people talk about the modern car, the new car as kind of the smartphone on wheels. If you want to take that approach you really have to look at source software architecture from a holistic point of view,” he said. “One of the most important domains here is the driving assistant domain. That needs to dovetail into what we call software-driven architecture, to be able to (with high computing power) add use cases for the customer, this case the driving assistant autonomous space.”
Can anyone please tell me what Ola is talking about, because that quote just sends my BS meter to 11. I expect that junk from Tesla, but Mercedes? I understand perfectly why Xpeng has licensed Nvidia for its AV technology, but Mercedes?
Kirsten’s prolific capacity for writing and reporting can further be found in Thursday’s reveal of the 2021 Ford F-150 which shows that the Blue Oval plans to launch a hands-free Active Drive Assist in 2021 via software updates sans-Nvidia.
As a tech analyst, the Mercedes story doesn’t seem to me to be anything to do with the capability of Nvidia’s chips — even though that made all the headlines — but that Mercedes seems to have no more intellectual property for automated driving than Xpeng or any of the other Chinese new energy vehicle makers that have signed licensing deals with Nvidia.
I really hope I am wrong, because otherwise Mercedes looks to be heading for a future as an Nvidia white-label box maker with a legacy internal combustion engine business — but who wants to write that headline?
Yes, you’ve guessed it, Kirsten also has a piece on the Waymo-Volvo deal. Seriously, when does she sleep? Let’s even the score a little with a link to a piece from Andy Hawkins in The Verge. Full marks to both for generating lots of industry buzz. Now let’s look behind the byline at the elephant in the AV room.
Who is going to buy the Waymo-enabled cars to operate as a robotaxi fleet? Who is liable for the statistically inevitable deaths resulting from these vehicles being deployed in ever greater numbers? I’m fantastically impressed that Waymo has signed-up another automaker alongside Fiat Chrysler Automobiles (FCA), Jaguar Land Rover (JLR) and Renault/Nissan. I am certain more will follow. But let’s look at some cold, hard numbers.
If we take the 62,000 Chrysler Pacifica Hybrid minivans and add it on to the 20,000 all-electric I-Pace vehicles, we get 82,000 units. Say the average car price, plus sensor suite, compute processing, and Waymo license fee is about $100K, and we have an up-front investment from the robotaxi operators of over $8 billion — for an unproven business model. All of the investment, depreciation, liability and execution risk would appear to be on them, so who are the operators? Forget about L4 autonomy, I want to know where the money is coming from?
Waymo’s valuation in March was down to $30 billion and that was pre-Covid-19. The bottom has fallen out of the automotive market this year and robotaxi VC investments have dropped too — Zoox was just sold to Amazon for about $1.2 billion having raised about $800 million.
There’s only one entity left buying everything at current nosebleed valuations, and that is the Federal Reserve. So maybe the expectation is that the Fed is going to become a robotaxi operator sometime in the next couple of years? Or maybe the sunny announcements being made by the AV tech companies are the industry’s last hurrah before the money runs out for most. I don’t know, but I suspect there are some unfortunate headlines up ahead.
Don’t believe everything you read
The PAVE virtual panel was a hugely insightful deep-dive into the challenges facing journalists writing about AV. This has been an incredibly busy week for news and I have written three pieces for EE Times. I need to go and lie down in a darkened room to recover, so I don’t know how professionals like Kirsten Korosec, Pete Bigelow and Junko Yoshida do it. I salute them.
The most important lesson: Be seriously skeptical. If your instincts tell you something doesn’t seem all that it appears, chances are they are right. Don’t read a story and look at what is written — read it and ask yourself what is missing. That’s much more informative.