The Driverless Future

Autonomous Risk

When self-driving vehicles meet the sharing economy, numerous industries will need to brace for change; insurance among them.
By: and | October 1, 2016

Outside of the rare Google car sighting, autonomous vehicles, or A.V.s, have mostly been the domain of futuristic action movies. That fictional future, though, is now.

Advertisement




A limited number of autonomous Volvo XC90s will be on the road this year. Audi’s self-driving A8 limousine is due out in 2017. Tesla plans to have fully autonomous vehicles available in 2018. Top executives at Ford and GM expect to launch fully autonomous vehicles by 2020. The U.S. Secretary of Transportation said at the 2015 Frankfurt Auto show that he expects driverless cars to be in use all over the world by 2025.

The twist, however, is that by 2025, the dominating force driving sales of those vehicles will most likely be commercial or public fleet operators, rather than private individuals.

VIDEO: Uber driverless takes to the streets in Pittsburgh.

While carmakers have been perfecting all this autonomous technology, the rest of the world was getting cozy with Zipcar, Car2Go, Uber and Lyft. Now those two disruptive concepts are merging, and the result is a seismic shift in the world as we know it.

Because driverless cars eliminate human error and human foibles like road rage and texting while driving, accident rates are expected to decline. Combined with an overall decrease in ownership, some say that will — eventually — cause a drastic upheaval in the auto insurance industry.

“Right now … everybody’s wringing their hands,” said John Lucker, global advanced analytics & modeling market leader with Deloitte Consulting.

John Lucker, global advanced analytics & modeling market leader, Deloitte Consulting

John Lucker, global advanced analytics & modeling market leader, Deloitte Consulting

“Half of all premiums are going to disappear! The industry’s going to blow up! The personal lines market is going to disappear! And we’re saying, ‘Well, time out everybody.’ ”

The likely reality is more nuanced. Deer will still dart into the road in the blink of an eye. Stray softballs and golf balls will still take out windshields. Car theft isn’t going away. And of course, machines can malfunction or be hacked. Insurance will remain as necessary as ever, but coverage needs will shift in a variety of ways.

“There’s going to be a blurring of the distinctions between various types of commercial insurance and personal insurance,” said James Guszcza, U.S. chief data scientist for Deloitte Consulting.

Sorting through it all is going to be incredibly complicated for carriers, as personal ownership slowly gives way to increased commercial ownership, in the form of driverless “Uber armies” or large A.V. fleets. There will also be a long period when non-autonomous, semi-autonomous, and fully autonomous vehicles will all be in the mix to varying degrees.

“Until all of the vehicles on the road are autonomous, how do you mix and match this?” asked Lucker.

“Your car is acting on its own, then it crashes into somebody who’s driving. How do you figure out who’s at fault? And then, is it a software issue? Is it your liability because you [own the car]? Was it the fault of the [human driver]?

“I don’t think that these things are going to go away,” Lucker said. “There’s obviously a blending of commercial auto, personal auto, potentially E&O for software developers, potentially some sort of product liability risk if a sensor didn’t work properly on the autonomous vehicle … there’s a lot of stuff going on here that traditionally has always been assumed to be a simple personal auto coverage.”

“This is a spectrum,” said Guszcza, “and we’re moving along that spectrum.”

How Quick a Shift?

Opinions vary widely on how fast this transformation will come to pass. Looking out to 2025, “you are going to find universally available cheap automotives, but those are probably going to be bought by self-drive fleet-manager-as-a-service type companies,” said Chris Smedley, CEO of Digital Habitats Corp. and longtime technology entrepreneur.

“Most of us are going to pay for transportation on a per-use basis.”

The rise of mobility as a service (MaaS) is taking hold quickly, and human drivers are indeed being pushed out. Uber CEO Travis Kalanick expects the entire Uber fleet to be driverless by 2030. The foundation is already being laid.

“I can’t imagine how many years it’s going to take for there to be mechanisms where police or road construction can be updated in real time to the point where true autonomous vehicles can navigate.” —John Lucker, global advanced analytics & modeling market leader, Deloitte Consulting

In August, the MIT-born company NuTonomy launched a test fleet of six self-driving taxis in a small section of Singapore, with human drivers on board to take over if necessary. Uber and Volvo also announced plans to launch a fleet of 100 self-driving Uber vehicles in Pittsburgh by the end of summer. The goal is to eliminate human drivers altogether — eventually.  Once that happens, Uber’s Kalanick said, its service will be so inexpensive and broadly available as to make personal car ownership obsolete.

Advertisement




Researchers seem to agree. A 2015 Barclay’s report concluded that auto sales will dip by as much as 40 percent within the next 25 years, as traditional ownership gives way to family autonomous vehicles, shared autonomous vehicles, and autonomous vehicle pooling (think Waze Carpool with no drivers). Barclays estimated that one shared, pooled vehicle could do the work now accomplished by 17 vehicles.

But there’s a silver lining for car manufacturers. A January 2016 McKinsey report suggested that the decline in private vehicle sales will be offset by increased sales of commercially and publicly owned shared fleet vehicles. Those vehicles will need to be replaced far more often due to heavy usage. The report concluded that overall global car sales would drop from the current annual growth rate of 3.6 percent to 2 percent by 2030.

The bottom line is that until somebody figures out how to make Star Trek’s transporter beam a reality, cars and trucks will still be the lifeblood of commerce and society. It’s the way we use them that’s about to shift in a variety of ways, and every manner of industry will need to assess how the shift will affect them.

Like falling dominoes setting off chain reactions, the mobility revolution will strike blows across multiple industry sectors.

Impact Across Industries

The beleaguered petroleum industry is in for more pain, because a steadily growing percentage of new autonomous cars will be electric. The taxi industry, already ailing, appears to be doomed.

James Guszcza, U.S. chief data scientist, Deloitte Consulting

James Guszcza, U.S. chief data scientist, Deloitte Consulting

The parking industry will take heavy hits — who needs a parking space when you Uber to work? Even those who do choose to buy their own autonomous cars won’t need to park — they can just send their cars home to wait.

Prevailing wisdom says that accidents will decline sharply as A.V. usage increases, which could send the $62 billion repair industry into a tailspin. Automotive computer repair will move to the fore as collision repair needs shrink.

Car-sharing and ride-hailing services, now the disruptors, will become the disrupted unless they evolve as Uber is already trying to do. The traditional car rental industry will also be forced to adapt or die.

The same changes will drive a new dynamic in public transportation as well. In fact, it’s already begun.

In April, the City of Beverly Hills, Calif., unanimously passed a resolution to develop a public transportation system consisting of driverless municipal shuttles. The idea is spreading. All seven finalists in the U.S. Department of Transportation’s 2016 Smart City Challenge submitted proposals that included semi- or fully autonomous vehicles.

Columbus, Ohio, winner of the $50 million challenge, plans a fleet of connected, electric, autonomous shuttles to ferry people around its business district. Other cities’ proposals included plans to shuttle people between transit hubs and airports autonomously, or to have self-driving vehicles handle deliveries and municipal transportation of materials.

For the average business, this shifting dynamic will be a mixed bag. Eliminating drivers will obviously have a significant impact on the cost of transporting or delivering products. However, as new players enter the autonomous on-demand fleet space, trying to grab for market share, choices for business leaders could become cloudy.

“There’s going to be a blurring of the distinctions between various types of commercial insurance and personal insurance.” — James Guszcza, U.S. chief data scientist, Deloitte Consulting

How do you properly vet your vendor? How much do you know about their parts and programming and how safe or reliable their driverless units are? It remains to be seen whether a company could be held responsible if a fleet vendor’s car were to cause life or property damage while delivering its products.

It will likely take decades of litigation to achieve any clarity on these types of liability issues.R10-1-16p28-30_1Driverless6.indd

Deep, Deep Data

The upside for both insurers and insureds is that the newest automotive technologies — and the data they collect — will create transparencies to a degree barely ever imagined before.

The new wave of vehicles will be equipped with ever-more-sophisticated sensors and telematics, enabling an unprecedented degree of precision in adjusting as well as pricing, said Lou Brothers, senior manager, West Monroe Partners.

In the event of a crash, adjusters will simply download both vehicles’ “black box” data and know instantly what happened and why, improving accuracy and eliminating the need for investigations.

“The device knows,” said Brothers.

Lou Brothers, senior manager, West Monroe Partners

Lou Brothers, senior manager, West Monroe Partners

“It knows what it did, it knows what the other driver did, it knows what you did. And if the other car’s a smart car, we know all the components … the friction on the road, the speed, the position of the gas pedal and brake pedal, deceleration patterns … now there’s no question about who’s at fault.”

On-board technology will give underwriters access to more varied and deeper layers of data, potentially enabling them to fine-tune premiums specific to actual usage, said Brothers.

“Lou’s driving on I-78 heading west out of New York through the Holland Tunnel. Say we know that on the entry point on the Holland Tunnel, he has a 2 percent greater chance of a side-by-side collision because of the funneling that happens there. Therefore for that period of time, instead of paying $3.00 for insurance he’s going to pay $3.50 for insurance — just for that 10 minute period.

“We know everything about the car, where it is, how long it’s there for, we know the accident statistics of all of those different areas — you could really drive yourself into a very detailed, nuanced view of this one driver, in this one car, in this one scenario, at this moment.”

Admittedly, said Brothers, that capability is still a long way off. “The amount of computation, the amount of analysis and of thought that would have to be put into those risk models would be insane,” he said.

“We’re talking about … driving down to the statistic of one.”

But when it happens, it presents some very attractive possibilities for the businesses that will be entrusting their products to autonomous on-demand fleets.

If allowed access to the data collected from commercial fleets, companies would be able to make better decisions about the risk level of each provider, and even about the routes traveled by vehicles being used to conduct their business.

It could also drive a conversion to usage-based automotive insurance, giving commercial fleet owners more power to control their premiums and offer more competitive rates to customers.

Imagine a carrier sending a message that said, maybe you should consider these alternate routes because they’re safer. Or maybe suggesting taking a longer route because it’s better paved or lowers the overall risk of the route.

In the end, money talks, said Brothers. If your carrier told you that if you were to leave 10 minutes earlier every day, your premiums would go down $80, chances are good that you’d leave 10 minutes earlier.

“That kind of real-time alert could be a very positive thing that I think people would welcome.”

Infrastructure Lacking

Risk managers trying to map out how the mobility shift will impact their organizations still have some time to get a handle on it. Despite the current race to put autonomous vehicles on the road, some believe the pace of change will be slow because the infrastructure to support a driverless world doesn’t exist, and few are talking about how to create it.

Advertisement




That’s why projects like NuTonomy’s Singapore test and the Uber/Volvo Pittsburgh venture make sense, because they’re constrained to areas that can be mapped precisely.

The trickier problem is that even keeping vehicles to proscribed routes won’t eliminate unknowns, experts said.

“Let’s say a sinkhole gets formed because of a massive rainstorm,” said Lucker. “How are the police going to alert this global GPS repository that Maple Street is now closed, so every single car knows not to turn down Maple Street?

“Worse, what if there’s an obstacle or accident halfway down a road, the cars don’t know that, and a whole string of cars goes down the road and they form a traffic jam. How will the cars sort out how to get out of there?

“There is no mechanism to create the global GPS database repository of changing road conditions and changing obstacles that an autonomous car is going to have to have access to in real time in order for something to be truly autonomous,” Lucker said.

“I can’t imagine how many years it’s going to take for there to be mechanisms where police or road construction can be updated in real time to the point where true autonomous vehicles can navigate.” &

Michelle Kerr is workers' comp editor for Risk & Insurance. She can be reached at [email protected]

More from Risk & Insurance

More from Risk & Insurance

Risk Scenario

The Betrayal of Elizabeth

In this Risk Scenario, Risk & Insurance explores what might happen in the event a telemedicine or similar home health visit violates a patient's privacy. What consequences await when a young girl's tele visit goes viral?
By: | October 12, 2020
Risk Scenarios are created by Risk & Insurance editors along with leading industry partners. The hypothetical, yet realistic stories, showcase emerging risks that can result in significant losses if not properly addressed.

Disclaimer: The events depicted in this scenario are fictitious. Any similarity to any corporation or person, living or dead, is merely coincidental.

PART ONE: CRACKS IN THE FOUNDATION

Elizabeth Cunningham seemingly had it all. The daughter of two well-established professionals — her father was a personal injury attorney, her mother, also an attorney, had her own estate planning practice — she grew up in a house in Maryland horse country with lots of love and the financial security that can iron out at least some of life’s problems.

Tall, good-looking and talented, Elizabeth was moving through her junior year at the University of Pennsylvania in seemingly good order; check that, very good order, by all appearances.

Her pre-med grades were outstanding. Despite the heavy load of her course work, she’d even managed to place in the Penn Relays in the mile, in the spring of her sophomore season, in May of 2019.

But the winter of 2019/2020 brought challenges, challenges that festered below the surface, known only to her and a couple of close friends.

First came betrayal at the hands of her boyfriend, Tom, right around Thanksgiving. She saw a message pop up on his phone from Rebecca, a young woman she thought was their friend. As it turned out, Rebecca and Tom had been intimate together, and both seemed game to do it again.

Reeling, her holiday mood shattered and her relationship with Tom fractured, Elizabeth was beset by deep feelings of anxiety. As the winter gray became more dense and forbidding, the anxiety grew.

Fed up, she broke up with Tom just after Christmas. What looked like a promising start to 2020 now didn’t feel as joyous.

Right around the end of the year, she plucked a copy of her father’s New York Times from the table in his study. A budding physician, her eyes were drawn to a piece about an outbreak of a highly contagious virus in Wuhan, China.

“Sounds dreadful,” she said to herself.

Within three months, anxiety gnawed at Elizabeth daily as she sat cloistered in her family’s house in Bel Air, Maryland.

It didn’t help matters that her brother, Billy, a high school senior and a constant thorn in her side, was cloistered with her.

She felt like she was suffocating.

One night in early May, feeling shutdown and unable to bring herself to tell her parents about her true condition, Elizabeth reached out to her family physician for help.

Dr. Johnson had been Elizabeth’s doctor for a number of years and, being from a small town, Elizabeth had grown up and gone to school with Dr. Johnson’s son Evan. In fact, back in high school, Evan had asked Elizabeth out once. Not interested, Elizabeth had declined Evan’s advances and did not give this a second thought.

Dr. Johnson’s practice had recently been acquired by a Virginia-based hospital system, Medwell, so when Elizabeth called the office, she was first patched through to Medwell’s receptionist/scheduling service. Within 30 minutes, an online Telehealth consult had been arranged for her to speak directly with Dr. Johnson.

Due to the pandemic, Dr. Johnson called from the office in her home. The doctor was kind. She was practiced.

“So can you tell me what’s going on?” she said.

Elizabeth took a deep breath. She tried to fight what was happening. But she could not. Tears started streaming down her face.

“It’s just… It’s just…” she managed to stammer.

The doctor waited patiently. “It’s okay,” she said. “Just take your time.”

Elizabeth took a deep breath. “It’s like I can’t manage my own mind anymore. It’s nonstop. It won’t turn off…”

More tears streamed down her face.

Patiently, with compassion, the doctor walked Elizabeth through what she might be experiencing. The doctor recommended a follow-up with Medwell’s psychology department.

“Okay,” Elizabeth said, some semblance of relief passing through her.

Unbeknownst to Dr. Johnson, her office door had not been completely closed. During the telehealth call, Evan stopped by his mother’s office to ask her a question. Before knocking he overheard Elizabeth talking and decided to listen in.

PART TWO: BETRAYAL

As Elizabeth was finding the courage to open up to Dr. Johnson about her psychological condition, Evan was recording her with his smartphone through a crack in the doorway.

Spurred by who knows what — his attraction to her, his irritation at being rejected, the idleness of the COVID quarantine — it really didn’t matter. Evan posted his recording of Elizabeth to his Instagram feed.

#CantManageMyMind, #CrazyGirl, #HelpMeDoctorImBeautiful is just some of what followed.

Elizabeth and Evan were both well-liked and very well connected on social media. The posts, shares and reactions that followed Evan’s digital betrayal numbered in the hundreds. Each one of them a knife into the already troubled soul of Elizabeth Cunningham.

By noon of the following day, her well-connected father unleashed the dogs of war.

Rand Davis, the risk manager for the Medwell Health System, a 15-hospital health care company based in Alexandria, Virginia was just finishing lunch when he got a call from the company’s general counsel, Emily Vittorio.

“Yes?” Rand said. He and Emily were accustomed to being quick and blunt with each other. They didn’t have time for much else.

“I just picked up a notice of intent to sue from a personal injury attorney in Bel Air, Maryland. It seems his daughter was in a teleconference with one of our docs. She was experiencing anxiety, the daughter that is. The doctor’s son recorded the call and posted it to social media.”

“Great. Thanks, kid,” Rand said.

“His attorneys want to initiate a discovery dialogue on Monday,” Emily said.

It was Thursday. Rand’s dreams of slipping onto his fishing boat over the weekend evaporated, just like that. He closed his eyes and tilted his face up to the heavens.

Wasn’t it enough that he and the other members of the C-suite fought tooth and nail to keep thousands of people safe and treat them during the COVID-crisis?

He’d watched the explosion in the use of telemedicine with a mixture of awe and alarm. On the one hand, they were saving lives. On the other hand, they were opening themselves to exposures under the Health Insurance Portability and Accountability Act. He just knew it.

He and his colleagues tried to do the right thing. But what they were doing, overwhelmed as they were, was simply not enough.

PART THREE: FALLING DOMINOES

Within the space of two weeks, the torture suffered by Elizabeth Cunningham grew into a class action against Medwell.

In addition to the violation of her privacy, the investigation by Mr. Cunningham’s attorneys revealed the following:

Medwell’s telemedicine component, as needed and well-intended as it was, lacked a viable informed consent protocol.

The consultation with Elizabeth, and as it turned out, hundreds of additional patients in Maryland, Pennsylvania and West Virginia, violated telemedicine regulations in all three states.

Numerous practitioners in the system took part in teleconferences with patients in states in which they were not credentialed to provide that service.

Even if Evan hadn’t cracked open Dr. Johnson’s door and surreptitiously recorded her conversation with Elizabeth, the Medwell telehealth system was found to be insecure — yet another violation of HIPAA.

The amount sought in the class action was $100 million. In an era of social inflation, with jury awards that were once unthinkable becoming commonplace, Medwell was standing squarely in the crosshairs of a liability jury decision that was going to devour entire towers of its insurance program.

Adding another layer of certain pain to the equation was that the case would be heard in Baltimore, a jurisdiction where plaintiffs’ attorneys tended to dance out of courtrooms with millions in their pockets.

That fall, Rand sat with his broker on a call with a specialty insurer, talking about renewals of the group’s general liability, cyber and professional liability programs.

“Yeah, we were kind of hoping to keep the increases on all three at less than 25%,” the broker said breezily.

There was a long silence from the underwriters at the other end of the phone.

“To be honest, we’re borderline about being able to offer you any cover at all,” one of the lead underwriters said.

Rand just sat silently and waited for another shoe to drop.

“Well, what can you do?” the broker said, with hope draining from his voice.

The conversation that followed would propel Rand and his broker on the difficult, next to impossible path of trying to find coverage, with general liability underwriters in full retreat, professional liability underwriters looking for double digit increases and cyber underwriters asking very pointed questions about the health system’s risk management.

Elizabeth, a strong young woman with a good support network, would eventually recover from the damage done to her.

Medwell’s relationships with the insurance markets looked like it almost never would. &

Bar-Lessons-Learned---Partner's-Content-V1b

Risk & Insurance® partnered with Allied World to produce this scenario. Below are Allied World’s recommendations on how to prevent the losses presented in the scenario. This perspective is not an editorial opinion of Risk & Insurance.®.

The use of telehealth has exponentially accelerated with the advent of COVID-19. Few health care providers were prepared for this shift. Health care organizations should confirm that Telehealth coverage is included in their Medical Professional, General Liability and Cyber policies, and to what extent. Concerns around Telehealth focus on HIPAA compliance and the internal policies in place to meet the federal and state standards and best practices for privacy and quality care. As states open businesses and the crisis abates, will pre-COVID-19 telehealth policies and regulations once again be enforced?

Risk Management Considerations:

The same ethical and standard of care issues around caring for patients face-to-face in an office apply in telehealth settings:

  • maintain a strong patient-physician relationship;
  • protect patient privacy; and
  • seek the best possible outcome.

Telehealth can create challenges around “informed consent.” It is critical to inform patients of the potential benefits and risks of telehealth (including privacy and security), ensure the use of HIPAA compliant platforms and make sure there is a good level of understanding of the scope of telehealth. Providers must be aware of the regulatory and licensure requirements in the state where the patient is located, as well as those of the state in which they are licensed.

A professional and private environment should be maintained for patient privacy and confidentiality. Best practices must be in place and followed. Medical professionals who engage in telehealth should be fully trained in operating the technology. Patients must also be instructed in its use and provided instructions on what to do if there are technical difficulties.

This case study is for illustrative purposes only and is not intended to be a summary of, and does not in any way vary, the actual coverage available to a policyholder under any insurance policy. Actual coverage for specific claims will be determined by the actual policy language and will be based on the specific facts and circumstances of the claim. Consult your insurance advisors or legal counsel for guidance on your organization’s policies and coverage matters and other issues specific to your organization.

This information is provided as a general overview for agents and brokers. Coverage will be underwritten by an insurance subsidiary of Allied World Assurance Company Holdings, Ltd, a Fairfax company (“Allied World”). Such subsidiaries currently carry an A.M. Best rating of “A” (Excellent), a Moody’s rating of “A3” (Good) and a Standard & Poor’s rating of “A-” (Strong), as applicable. Coverage is offered only through licensed agents and brokers. Actual coverage may vary and is subject to policy language as issued. Coverage may not be available in all jurisdictions. Risk management services are provided or arranged through AWAC Services Company, a member company of Allied World. © 2020 Allied World Assurance Company Holdings, Ltd. All rights reserved.




Dan Reynolds is editor-in-chief of Risk & Insurance. He can be reached at [email protected]