Autonomous Driving Stocks Decline, Argo AI shuts down, CARIAD Software Delays Porsche EV, Qualcomm Might Win Big In Vegas
Dear Reader,
Last week, the story broke that Argo AI is shutting down:
The AV/robotaxi startup co-owned by Volkswagen and Ford will be partially resorbed into the two OEMs' ADAS teams, as both of them seemingly abandon their Level 4 efforts in favor of better L2/L3 systems for consumer vehicles.
Suddenly, a lot of voices in the industry dared utter the 'forbidden question' - is autonomous driving going to fail? Sort of yes, but also no. Let's dive into a number of news items from this month which are all very much connected to each other and to answering that question:
These figures compiled by Crunchbase tell us one thing for sure: Automated driving has entered the age of accountability. Gone are the days when you could buy cheap money, media goodwill and public favor with little more than fancy PowerPoint slides and ambitious promises about driverless vehicles.
Automated driving is still a massive value driver within ADAS - but unlocking it will require proof that you can actually deliver on those promises: With Tesla under criminal investigation by the US Department of Justice and rigorous type approval regulation announced by the European Commission, the days of simply deploying something not yet fully legislated and learning from whatever happens seem all but gone.
As I wrote last month, we may have to get used to less flashy PowerPoints and expect incremental AD progress instead of revolutionary leaps. Doing so seems much safer, also from an economical point of view: ADAS is an industry worth tens of billions of dollars yearly and with healthy growth rates - it seems a safe bet to say that supervised automation has a future.
For those at Argo, this may not be all that comforting in the near term, though. Not all US employees will receive a new offer; and it also seems unlikely that all of their German colleagues will want to join Volkswagen/CARIAD. As a matter of fact, I'm starting to worry a bit about the future of CARIAD itself:
领英推荐
Porsche Macan EV delayed due to VW software woes - via Autocar
Porsche has been vocal about its discontent with the new CARIAD software platform in the past, when Herbert Diess was still in charge at Volkswagen. Together with Audi, the sports car brand pushed strongly for an interim release, because the wait for the eventual "E3 2.0" platform seemed too long. Now it seems that even the compromise they negotiated, dubbed "E3 1.2" is late - to such a degree that Porsche has to delay the launch of its next Macan EV model.
This year, two relevant things have changed in the picture:
If such a software platform brake-away actually happens, it would be quite a blow for Volkswagen's CARIAD. This is special in the sense that it seems like Porsche-Blume undermining VW-Blume. It may, however (and this is speculation, of course) be a calculated move to make CARIAD the perceived 'bad bank' within Volkswagen - you could see this as Blume laying the groundwork for an eventual dismantling of unloved Diess legacy. And there are other related developments in the works, with Qualcomm at the center:
Winners, Losers in the Fall of Argo AI - via The Ojo-Yoshida Report
In May of this year, Qualcomm won a contract to supply its Snapdragon Ride SoC to the Volkswagen brands through CARIAD. The new Snapdragon Ride Flex, which was announced just this September, could presumably serve as the basis for all things ADAS and infotainment - and might just replace E3 2.0 entirely.
Colin Barnden, Principal Analyst at Semicast Research and author of this article actually expects VW to be named as one of the first OEMs to commit to Flex at CES in January - along with their former Argo co-owners Ford as well as BMW and General Motors.
As the article quotes Blume, it would seem that Volkswagen is ready to re-embrace consumer vehicles, as well as the traditional OEM principle of outsourcing economical development risks to suppliers: "Our goal is to offer our customers the most powerful functions at the earliest possible time and to set up our development as cost-effectively as possible."
All in all, it looks like we're in for a hot winter season ...
Senior Data Product Owner ?????? #ML #AI | Experience in Purchasing, Cost Engineering, Cost of Sales and Automotive Industry ????
2 年I would say that having nice deployment of L3/L4 will open doors for L5. It means that even more important investments are needed ??????. Stakes in terms of efficiency and safety are so big that continuing to invest makes sense. #automotive #autonomousdriving
Leading Autonomous Driving @ NEOM
2 年Spot on analysis Tom. And also a very interesting take on Blumes (possible) strategy for CARIAD, as his split seating might very much indicate a conflict of interest that he is trying to resolve. Thanks!
AV Safety Consultant
2 年The month is not over yet... https://finance.yahoo.com/news/tusimple-probed-fbi-sec-over-201603623.html