AVision: Rideshare and Autonomous Vehicles

Full Disclosure: For the last few months, I've been driving for Lyft and Uber. While it's not quite the lucrative gig-economy lifestyle promised by the advertisements that lured me in, I appreciate it for the low barrier to entry and flexibility.  I do this with the knowledge that five to ten years from now, driving for rideshare services may no longer be an option.  In today’s post we will explore why.

The Rideshare Disruption

Rideshare as we know it began six years ago with two entrepreneurs, a Mercedes S Class, a driver and an iPhone app, and has since exploded in popularity. Transport network companies like Uber and Lyft have positioned themselves as low-cost alternatives to traditional transportation options, and the market has responded.  Emerging as an often controversial answer to widespread frustration with inadequate public transportation infrastructure and disillusionment with the taxi industry, rideshare companies have expanded their presence across the globe and rewritten the rules of urban transit using only an app and contracted drivers offering rides in their own vehicles.  This disruption has shifted how people get around cities in numerous ways, creating new opportunities for movement in urban transportation networks that have long been static.  Capitalizing on this new service, consumers are now using rideshare as a temporary substitute, complement, or even complete replacement for other transit options.

When public transit systems break down or are otherwise disrupted, its users turn to rideshare services to get to work on time, sometimes at a steep price.  And, despite strained relationships with numerous local governments, some American cities have started partnering with Uber to bolster their public transit systems, particularly in the first and last miles of urban commutes.  Customers are also turning to Uber and Lyft for mid-range commutes, which can be cheaper than buying a car

Many users also rely on rideshare for their weekend activities, allowing even those with cars to avoid the hassle of parking downtown and the danger of drinking and driving.  This allows passengers to circumvent some of the cost and safety considerations that plague the taxi industry, while also beginning to address race and destination discrimination issues perpetuated by some taxi drivers.

Rideshare is now allowing residents of lower-income communities to access affordable and reliable transportation instantly with a smartphone.  As smartphone ownership continues to increase among the American adult population, including 50% of Americans earning $30,000/year or less and 71% of Americans earning $30,000-49,000/year, rideshare services have the potential to compensate for the gaps in public transportation throughout metropolitan areas.  This is especially important following a recent report from the Brookings Institute that revealed only 30% of jobs in the country are accessible within a 90 minute ride using public transit.  Now, limited only by the time it takes for a driver to arrive, low-income communities have the potential for (but not the guarantee of) being reconnected with the larger city, empowering residents to travel where bus and rail lines fail to adequately serve them. 

Smartphone Ownership Highest Among Young Adults, Those With High Income/Education Levels


But an even more significant change in transportation is to arrive in the form of self-driving cars that will be hailed through rideshare apps like Uber and Lyft.  As the technology behind self-driving cars (also referred to as “autonomous vehicles” or “driverless cars”) matures, urban transit will likely shift toward these automated rideshare services, revolutionizing riding in cars just as much as Uber and Lyft revolutionized getting into them.

Automated Rideshare

Under this new model, rideshare companies will replace drivers using their own cars with fleets of company-owned self-driving cars.  Following Elon Musk's prediction that Tesla could be producing fully autonomous vehicles in three years, Uber CEO Travis Kalanick reportedly offered to buy 500,000 of them in 2020, perhaps entirely replacing its current 160,000 drivers within one phase of deployment. 

This model has been foreshadowed by a slew of public-private partnerships and corporate investments in autonomous vehicle research and development, principally led by tech companies and car manufacturers.  Recent developments include Google’s famous self-driving car, Uber’s controversial partnership with Carnegie Mellon University to create the Uber Advanced Technologies Center in Pittsburgh, and stirrings in the analyst community suggesting that Tesla is weighing a shot at developing its own autonomous vehicle rideshare system, hypothetically dubbed Tesla Mobility.  And, on August 25th, Uber announced yet another public-private partnership with the University of Arizona, facilitated by Arizona Governor Doug Ducey, which entails a $25,000 donation to the College of Optical Sciences in return for cooperation in mapping and optics research. 

The motivations behind Uber’s interest in autonomous vehicle technology are clear: self-driving cars will not demand higher wages or benefits, are less prone to accidents, will drive themselves to and from the gas/charging station, and, most importantly, eliminate the largest operational expense of rideshare services: paying the driver.  Furthermore, Uber’s brand and global presence will allow customers to continue using the same e-hailing app and will reduce barriers to product engagement.  This would be a boon for Uber and its customers alike, slashing the costs of operation and labor while increasing profitability and lowering fares for passengers. 

Uber CEO Travis Kalanick seems to agree, having stated at Code Conference in May 2014:

"The reason Uber could be expensive is you're paying for the other dude in the car. When there is no other dude in the car, the cost of taking an Uber anywhere is cheaper. Even on a road trip."

But positive attitudes toward autonomous vehicles are not universal.  While chatting with my rideshare passengers, I often have the opportunity to hear their opinions about this technology and the future of rideshare.  Many suggest they are still concerned about putting their lives in the hands of robots, saying that it would be “unnerving” or “dangerous.”  This attitude prevails in spite of the relatively safe track record of Google’s self-driving cars which, according to Google’s August 2015 report, have been involved in 16 minor accidents in 6 years and 1.5 million miles of autonomous driving—in each of these cases, Google’s cars were not found to be at fault.  

When I share these statistics, my passengers often say that they still like having human drivers because they trust the driver will want to arrive at their destination safely.  They are not alone in their hesitation: a recent study by Nerdwallet showed that only 50% of men and 37% of women are interested in eventually owning a self-driving car, and 50% of people would not pay more for full autonomous vehicle technology when purchasing their next car. 


Furthermore, people are far less comfortable about putting their kids in autonomous vehicles, indicative of a deep public confidence issue:

“As another measure of trust in autonomous car technology, we asked whether people would put a child in a self-driving car alone to go to school or a friend’s house. Only 6% gave a thumbs-up to that idea. Most people (76%) said no, and the rest were unsure.” (Nerdwallet)

The increased media coverage of self-driving cars has also fueled skepticism and opposition from pundits and consumers alike, drawing attention to the potential shortcomings of the technology.  In fact, a recent study by Gartner has shown that autonomous vehicles have begun to mature in the public eye to the point that they have peaked in their “hype level” of inflated expectations and, according to Gartner, autonomous vehicles will be met with deflated expectations by consumers in the coming years before the public attitude finally shifts toward more stable outlooks.

In this sense, the uptake of self-driving cars and automated rideshare is not just contingent on technological development, but also a well-coordinated campaign that reassures customers of the safety and superiority of automated vehicles.

The Driverless Elevator

Convincing consumers to adopt technologies that replace humans with automated systems is not a new challenge.  Planet Money recently discussed concerns about self-driving cars in the context of the widespread implementation of automated elevators in the 1940s.  Up until then, elevators were hand-operated by elevator attendants who manually operated the doors and controlled the elevator speed.  And although we take elevators for granted now – as well as running into them at the last minute – it was once dangerous to jump through the closing doors since human operators could not always react in time to prevent serious injury. 

However, safety concerns alone were not enough to trigger a massive transition to automated elevators, which were invented over 50 years before properties began to buy them.  Instead, the shift was triggered by the volatile relationship between elevator operator unions and management, including a strike in 1945 that virtually shut down 2,015 Manhattan office buildings, bringing the workday to a grinding halt as an estimated 1.5 million employees flooded into the streets and spent the day at bars, cinemas, and shopping centers.

To address the issue of both safety and labor stability, companies began to install automatic elevators that reacted almost instantly to prevent humans from getting caught in the closing doors and were unable to strike or form labor unions.  Of course, people were at first confused when they did not find attendants in the elevator – Planet Money mentions stories of people wandering around lobbies, looking for an elevator attendant.  Elevator companies started running campaigns promoting the safety and efficiency of these “driverless elevators” and the public eventually forgot there was ever “another dude” in the elevator.  

August 1952 Advertisement for the Otis Autotronic Elevator that mentions “Savings for each non-attended elevator average $5,500.” (Vintage Ads)

The key difference here is that we can pick our preferred mode of transportation much more readily than an elevator.  Those who are skeptical of self-driving cars will have the option to continue buying manual vehicles or riding in traditional rideshare services—that is, until riding in self-driving cars becomes unavoidable, either because manual vehicles are outlawed for safety reasons or are no longer manufactured (or licensing becomes prohibitively expensive or difficult).    

The Uber Playbook

Uber’s objective in deploying the first wave of its self-driving car service will be to convince its passengers that they are not using a new service, but simply an upgraded version of Uber that is safer, more reliable, and more efficient.  This initial driverless service will likely be designed to comfort passengers by emulating some of the aspects of riding in a traditional Uber: the seats will face forward, an embedded AI system will respond to directions and voice commands as a driver would, and an emergency stop function will be easily accessible.

Yet Uber’s greatest challenge will be that people in urban areas, particularly millennials who are notoriously averse to car ownership, likely will not be exposed to autonomous vehicles by someone they know in the few years following their release.  Instead, many people will probably experience their first ride in an autonomous vehicle through an automated rideshare service—particularly since Uber is currently exploring its own production of autonomous vehicles to make the technology cheaper to implement and because Uber, currently valued at $51 billion, could likely field its own fleets of autonomous vehicles.  Much like Uber and Lyft in their startup phases, these services will have to rely on their earliest customers to generate interest and trust in their safety through word of mouth (however, it may also be helpful if regulators did more than say that they might be legal)

Furthermore, Uber is likely to encounter significant resistance from its current drivers who – as we mentioned earlier this week – have already filed lawsuits regarding their status as contractors rather than employees with full wages and benefits.  As we will discuss in future articles, this will erupt into a larger political battle and international dialogue about the nature of work in the twenty-first century, and whether there is a social responsibility to keep people employed, even when automation objectively becomes cheaper, safer, and more effective. 

Uber CEO Travis Kalanick already came under fire last year for his aforementioned “other dude” speech lauding the potential for autonomous vehicles to replace drivers, after which he backpedaled and tweeted:

The backlash was fierce, and the public railed against Kalanick’s cold calculation and lack of empathy for the drivers that helped to build his company. 

In the coming century, such tone-deaf statements by executives will be a stumbling block for tech companies that serve broad demographics, particularly if people they know are being unceremoniously replaced through automation.  To combat this, companies such as Uber will need to show not only that their automated services are safe and desirable, but also that they are conscious of their social impact – a difficult task when the financial stability of as many as 160,000 contractors is on the line. 

For this reason, Uber will need to put a human face on their autonomous vehicles, espousing their safety and virtues, much like Google’s disarmingly adorable self-driving car and its feel-good video ads that emphasize mobility for seniors, children, and the visually impaired. 

The Human Element

Currently, 55% of women and 37% of men are concerned that self-driving cars would not be safe to use.  This is indicative of a larger obstacle to widespread use of autonomous vehicles: although the technology has been relatively safe throughout its testing and may eventually eliminate up to 90% of car accidents, it will be important to pair empirical evidence of their safety with our instinct to trust other humans with driving our vehicles.

We tend to trust that the humans controlling our cars, trains, and airplanes share our instinct for self-preservation.  We intuitively know that our airline pilots want to land safely and that our subway operator will not plow into a train at the next station because they also want to return home safely to their loved ones each night. 

What we will come to realize – and what automated rideshare companies will likely tell us – is that pilots, drivers, and train operators are already aided every day by automated technologies such as autopilot, crash avoidance features, and positive train control (PTC).  From this perspective, it naturally follows that we continue automating transportation to make it safer.  Automated rideshare companies will likely emphasize the human motivations behind this technology: how the engineers wanted to save the 33,000 lives lost to car accidents every year, how a child can get home from school safely in a self-driving car, and the value of the time recovered from the frustrations of driving like sitting in traffic, looking for parking, and taking the car to a mechanic.

What Happens to the Drivers?

When cars become fully automated and no longer need humans to operate them, the 233,000 taxi drivers and hundreds of thousands of rideshare drivers in the United States will find their jobs in danger.  Driving taxis, long known to be a reliable source of work for new immigrants and unskilled labor, will become an even more endangered occupation than it is today. 

As mentioned previously, rideshare drivers will not be spared from this seismic workforce shift either.  Some – though far from all – drivers have already voiced their opposition to Uber and Lyft’s current model designating them as 1099 contractors rather than employees entitled to full wages and benefits.  But this may just be the automatic elevator paradox repeating itself: by demanding to be conferred better compensation and working conditions, drivers may actually be encouraging Uber to rush ever faster toward simply doing away with human drivers and deploy autonomous vehicles as soon as possible.  We have already seen this take hold in the service industry, with companies looking to replace their workers with automated check-out stands, e-waiters, and even burger-flipping robots that can make a burger every 10 seconds.

 The Future

The rideshare model flourished under Uber and Lyft as an effort to connect people and share resources in order to travel faster, cheaper, and safer than previous systems allowed.  Despite our initial concerns about riding in a stranger’s car that we summon with a smartphone, we have come to recognize rideshare as a safe way to travel in cities around the world.  Presenting an alternative to driving our own vehicles, these rideshare services have challenged the mainstay of car ownership by using smartphones to find freedom and connect passengers with “Everyone’s Private Driver” and “Your Friend with a Car.”

In the same spirit, we may choose to adopt autonomous vehicles – and rideshare services powered by them – if we find that the benefits of autonomous vehicles exceed those of using human drivers. 

Although the potential for autonomous vehicles to be used in conjunction with rideshare is tremendous, it is by no means inevitable.  As mentioned previously, considerations remain in terms of opposition by labor, regulatory issues, complex insurance policies, consumer uptake of the technology, and a variety of other factors.

As we will discuss in upcoming posts, there is also a chance that the public will choose not to adopt autonomous vehicles, opting instead to merely use highly advanced driver assistance features that intervene only when collisions are imminent.  Or perhaps the transition to self-driving vehicles will happen so gradually through small, incremental leaps that the transition to full automation is barely noticed, and removing the operator of the vehicle seems to be a foregone conclusion.

Nevertheless, we have before us an unrivaled opportunity to democratize urban transit through automated rideshare, which will eliminate location discrimination, reduce the cost of urban travel, and virtually guarantee safe movement across cities and throughout the world. 

In the next few decades, the driver license may become a relic that is replaced by the smartphones in our pockets and a rideshare subscription service.  The financial and logistical burden of owning a car in a metropolitan area will no longer be necessary, as automated rideshare services will have a fleet of vehicles perpetually circulating the city, ready to pick up passengers at a moment’s notice.  If this comes to pass, one day our gridlocked city streets will look like this:

(Created by Kurt Dresner and Peter Stone from the University of Texas at Austin Computer Science Department)

And, with any luck, I will never have to teach my children how to drive in that.


Thank you for following The Autonomer!  Please sign up for our newsletter below and keep up with us on social media for the latest in automation news and policy.  Please let us know if you have any questions about the topics we have covered this week and send any requests for future articles to TheAutonomer@gmail.com