There’s no doubt that, over the long term, autonomous vehicles (better known as self-driving cars), combined with electrification and shared mobility, have the potential to massively change society.
However, almost every other question is up for debate. How long until we reach “true autonomy” (level 4 and 5)? Will autonomy act as lighter fluid for shared mobility services? What will the downstream impact be, from traffic to travel to retail?
My colleagues and I like to say that these long-term effects depend on how we get “from 1 to 100” — that is, from the launch of self-driving cars to the day when autonomous mobility is ubiquitous. But while prognostication is fun, the map to that destination is
still far from clear.
The pieces of the first leg of the journey, however, are becoming slightly clearer. We call this going “from 0 to 1” — from where we are today, to the successful commercialization of the first fully self-driving vehicle. Going from 0 to 1 requires solving three key challenges:
- Technological: Achieving full autonomy (i.e., Level 4 or 5 on
the SAE International automation scale)
- Regulatory: Creating the conditions for safe and effective operation
- Industrial: Discovering and organizing the right business model to produce a commercially viable product
The first two challenges are extraordinary, but will someday be overcome. More interestingly, the path forward on #3 — creating a business model to produce a sellable product — is just becoming clear.
In short, I believe there are two possible business models, which have deep parallels in the tech industry:
Many companies are placing their bets on what I call the “Apple model” — imagining an integration of hardware (the vehicle, sensors, etc.) and software (the AI “brain” making decisions to operate the car). In fact, most of the automakers — fearful of being reduced to commodity hardware players — are betting heavily on this model.
However, a division of labor may be more likely. I call this the “Android model” — where the key decision-making challenges are solved by an AI and machine learning platform — such as the one Alphabet (Google)’s Waymo unit is working on — and then licensed across numerous hardware-makers. There is increasingly evidence to suggest that industry leaders like Waymo, Apple, and Baidu are pursuing this approach.
My colleagues Alan Lewis, Rob Haslehurst, and I explored this further in an Executive Insights article, which can be found here (online PDF): Mapping the Road to Autonomous Vehicles.