AI policy & ethics, Tesla layoffs, Detroit Auto show highlights - SAI Newsletter #3
Updated: Jan 31, 2019
Just a brief introduction today. Not much going on at the Detroit Auto Show, but some important announcements from Tesla and a few other auto companies. An expose from CNN about Uber's Investigative unit that should be interesting to quite a few of you. If there are any topics you'd like me to focus future newsletters on, please do make the request and I'll try to accommodate. Too much going on, but my goal is to keep this short and informative.
My name is Tu Le and I am the founder and managing director of Sino Auto Insights.
This weekly newsletter is a collection of articles I feel best reflect the happenings of the week or important trends that have effects on the automotive and mobility sectors here and in the U.S. I also provide a point of view that I hope educates and sparks debate about how I look at the issues. We will mostly divide our articles into these buckets: AI, Mobility/Ridesharing/Ride-hailing/Bikesharing, OEMs, EVStartups, Investments, and Other. If you know of anyone who would like to sign up for this newsletter please have them visit: www.sinoautoinsights.com. Thanks for reading.
The Sino Auto Insights team
On the heels of the restructuring news coming out of GM and Ford, the OTHER big news coming out of the auto show was that it’s moving from freezing January to sun-soaked June starting in 2020.
This should help encourage some of the players that haven’t participated lately, namely the Germans, Tesla, and a few other major nameplates from having to brave the cold to intro some of their cool new cars.
Eerily absent from the show are two legends, Sergio Marchionne, who passed away earlier in 2018, and Carlos Ghosn who is stuck in a Japanese jail for allegedly under-reporting his salary for the last number of years.
- Toyota Supra (when compared to the concept, am thoroughly disappointed although, at that price point (~$50K), they should be able to sell as many as they make.
- Nissan’s IM Concept
- Ford Explorer and Mustang (two cars that will survive the chopping block)
- Cadillac XT6 (doesn’t have Super Cruise, arguably Cadillac’s best new tech, which is a shame)
- GAC Entranze EV concept (GAC will be selling vehicles in the U.S. by 2020, maybe)
- Lexus LC convertible concept (I like it …for a Lexus)
It was announced that Cadillac will be the brand that leads the EV transition for GM. There’s a lot riding on Cadillac succeeding including being able to finally get out from behind the shadows of Daimler, Bimmer, Lexus and all the other luxury automakers it currently sells less than.
Will they be able to do it while still being a ‘team’ player for GM and supporting their other brands? Some are skeptical and point to Infiniti, Lincoln as other examples of luxury carmakers that struggle under the weight of large multi-brand OEMs. That argument doesn’t hold weight if you make the comparison to Lexus, Audi or Porsche though.
One thing that Cadillac does have going for them is that they’ve sold more cars in China than in the U.S. since 2017 so the Chinese are gravitating towards the brand more and more. What we will likely see is that a lot of design influence will come from Asia, specifically China to the U.S. bound products, the opposite of what currently occurs.
The question is, can they maintain that momentum and design EVs that will compete with the likes of Tesla, Audi, Mercedes as well as the new Chinese EVStartups? Can’t wait to find out.
When you include this layoff with the layoff of 9% from the previous June, Tesla should be a pretty lean company. This means they’re taking the cost side of the profit equation pretty seriously and to the dismay of the current employees that get to keep their jobs. Their responsibilities likely increased significantly.
They are of course hoping that they’ve gotten past most of their self-inflicted wounds when it comes to manufacturing for mass production and can ramp the Model 3 for Europe, which they’ve just launched, and China which they will launch later this year. That is on top of launching their promised $35K version of the Model 3 later this year in the U.S., then China.
That should lead to a pretty busy plant, suppliers, and employees but if they plan to squeak out a small profit on those $35K Model 3’s, this was a necessary, you could almost stay strategic move. I say this a bit tongue-in-cheek because most traditional OEMs would’ve kicked the layoff ‘can’ down the road until they were literally forced to cut staff.
This article highlights some of the lingering issues from prior management that are still being cleaned up by Dara and the new management team.
The numbers being thrown around sound like a lot but there was little context given relative to the other sectors that have similar investigative teams. The Uber investigative team does seem small and the author does highlight how underpaid Uber’s investigative team is on average, compared to other sectors which is a lot, but it does seem to be improving and a lot of the issues surrounding this team do not seem to be unique to Uber.
Makes me wonder how Didi tackles this, the number of complaints they deal with on a weekly basis, as well as the strategy on dealing with them and the size of their investigative team. This would seem very relevant to them since they had those two tragedies last year with their two female passengers. An article I was interviewed for, written earlier this year by Wired discussed in detail the challenges that Didi had to deal with because of those tragedies.
I get the feeling that Lyft, Didi, Grab or any other ride-hailing company isn’t that keen to share these numbers nor how they resolve these types of complaints so kudos to Uber for being more open kimono about this than other companies in the space.
Anyone who has been a part of this sector for a fair bit of time, and TRULY understands all that’s involved with getting autonomous vehicles on the road; policy, technology, public acceptance, data, safety, investment, competitors, infrastructure, products, etc. (please let me know if I’ve missed anything), kinda sorta knew in the back of their minds that this would not happen quickly.
That didn’t stop them from boasting about their ability to put a driverless vehicle on the road. We are still a good minimum 15 years away from cars, infrastructure, other types of mobility all communicating via V2X and moving the accident rate to zero after all isn’t that one of the true main goals?
I’ve had many discussions with folks about the morality/ethics of AI, or more importantly lack thereof. I’ve also read about the examples of some of the blindspots AI has that this author mentions in the article. Bias that need to be considered prior to creating the system.
I’ve even thrown around the idea with colleagues of creating a set of AI norms that can be posted on Wikipedia until they reached the point that there is some consensus that can be used across AI applications, sectors, functions, etc. It would basically be a list of ‘Don’t Be Evil’ type guidelines that would cross-cultural & moral boundaries and truly be universal and sets the table moving forward for how AI should work and plug in with society.
This is all very well and good, but if we haven’t been able to eliminate bias in products, services, policies, and people up till now, why do we think we’d be able to eliminate it from AI?
This article and the prior article touch on a similar concern, but in this case, a team of individuals, likely policymakers or diplomats, from each of the 36 OECD countries gathered at MIT recently to discuss and develop consensus on AI policy for those countries.
In theory, this should help level the playing field with the goal or wrestling away some of the heavy influence from the two countries that plan to benefit from AI the most, you guessed it China and the U.S.
The goal of the group is to foster responsible economic development, balancing innovation and social protections. This group of people is specifically gathering with the goal of making the world safe for AI while allowing it to flourish.
Regulations will be forthcoming so that rules of competition and engagement can be agreed to. A good headline from the article: AI policy is data policy
I always get a bit concerned when a bunch of people that probably don’t have any idea about the types of struggles their everyday fellow citizens (read: poor, mostly minority) go through, nonetheless think they’re wise enough to dictate how the world should work. I just stepped off my soapbox.
I don’t really subscribe to this author’s view that the traditional OEMs aren’t concerned about Tesla or that Tesla doesn’t keep them up at night since the folks I’ve spoken with and the actions that many OEMs have recently taken point to the exact opposite.
He uses as examples some of the one in a lifetime, disastrous problems that have inflicted FCA and Nissan-Renault. When does, in the normal course of business for ANY company, a CEO pass away or get thrown in jail??
Should they be thinking of anything else than trying to calm the troops, have their companies stay the course until they find an appropriate next leader for their company?
And for ONLY selling 250K units in 2018, a rounding error for most large OEMs, how is Tesla able to have such a passionate, opinionated following and such a large voice when compared to their minuscule footprint in the industry?
I know for a fact that most automakers, startups or otherwise would pay lots of money for such a following for one of their vehicles, let alone an entire brand.
The author also glosses over the difficulty of designing and manufacturing a successful EV as something the automakers could do pretty quickly and easily. This is NOT true, they do NOT have the personnel to design and develop a vehicle, that’s why they’ve pushed everything out to 2022. When Ford initially integrated MS Sync into their Focus, it was a huge hit! One version later, it was a HUGE source of complaints from customers and brought their JD Powers initial quality down.
Not to say that Sync was a bad product but how it was integrated and designed into the user experience, something at the time Ford didn’t prioritize, as an afterthought really stood out to its end users.
Finally, he brings up Apple as an example of great design and refinement of user experience. One could argue that Apple is THE best at hardware-software integration and interaction design and I am pretty sure that any car company would love to have that type of reputation.
Sino Auto Insights is a Beijing, China-based market research and advisory firm that specializes in assisting companies analyze, strategize, and develop products and services that will shape the future of mobility and transportation. Members of our team have experience working in Detroit, Silicon Valley as well as here in China across multiple sectors and functions as entrepreneurs as well as working at larger companies like Apple, Google, Amazon, GM and FCA, and many others.