Tag Archives: robots

The Future of Work: Collaborative Robots

Recent developments in robotics and artificial intelligence have changed the playing field for automated technologies. (Here is an earlier blog on the topic.) Historically, automation was beyond the reach of small and medium-sized companies. Robotics were costly, required highly sophisticated programming expertise, took months to integrate and could only perform single, discrete tasks.

In 2012, the advent of artificial intelligence (AI) was a game changer. AI brought collaborative robots to the market — robots that see and feel like humans, learn (including integrating new data sets and information) and perform multiple tasks. These collaborative robots are also more cost-effective and easier to integrate, making them available and attractive to small and medium-sized businesses.

AI and robotics are now transforming many traditional labor-intensive industries, such as farming, construction, factories and fast food. While Amazon continues to be a global leader in leveraging AI and smart robots, there are plenty of examples of smaller businesses across the country embracing these new automated technologies.

Agricultural farms are using automated tractors and drones to help with growing their crops. Construction firms are purchasing automated brick-laying machines (to lay 3,500 bricks per day). Restaurant owners are investing in new automated machines that can store, prep and cook fast food in a highly controlled environment without any human intervention.

If the adoption of these new automated machines continues, there will be fewer jobs and payrolls in these industries. Over time, the job and payroll loss will affect insurance carriers that specialize in writing workers compensation insurance for these industries.

Historically, technology’s disruption was limited to blue-collar workers; however, AI technology now has its sights set on white-collar workers, including insurance underwriters, claims executives and legal professionals. The insurance industry, which has not been easy to disrupt, is primed for transformations due to developments in AI and automation.

Two years ago, Cambridge University predicted that insurance underwriters were vulnerable to automation. Since that time, we have seen a greater demand among U.S. carriers to invest in new AI technologies that allow them to automate the underwriting and settlement of claims for small commercial insureds. Given the shortage of new talent available to fill expected insurance and claims executives retirements, coupled with new AI technologies, we expect this trend to accelerate.

See also: Measuring Success in Workers’ Comp  

Developments in AI and automation are already changing the U.S. legal profession, one of the most regulated and specialized professions in the U.S. — McKinsey estimates that 22% of lawyers’ and 35% of paralegal tasks can be automated today. A recent HBO documentary, “The Future of Work,” supports this prediction. It highlighted how LawGeex, a new AI-driven computer software, performed against skilled corporate lawyers on a common task — analyzing complex legal documents. LawGeex proved its ability to review and interpret the documents, identify potential legal issues and provide substantive advice to a client in half the time — and with much greater accuracy — than the corporate lawyer.

While LawGeex and other AI technologies will not displace lawyers in the short term, it will exert pressure on lawyers to shift their time to more highly skilled work – such as negotiating and deal structuring – and away from research, writing and reviewing documents. The result could significantly change law firm practices and economics.

Have you considered how robots, AI and automation will change the workplaces of your insureds – and your own company? Stay tuned for my next blog, “Navigating the Fourth Industrial Revolution,” for ideas on how to navigate AI and developing technologies.

What Robots Mean for Workers’ Comp

History provides interesting insights into the debate around automation and employment. In 1632, King Charles I of England banned casting of buckets, for fear that allowing it would ruin the livelihood of the craftsmen who were making the buckets the old-fashioned way. In 1811, the Luddites in England started a movement where they smashed machines that they viewed as threats to employment. These examples have occurred with increasing frequency since the industrial revolution began. Not coincidentally, the per capita income in the world doubled every 6,000 years prior to the revolution and every 50 years afterward.

According to a Pew study, 52% of Americans think that much of our work can be done by robots, but only 38% believe that it could replace the type of work that they do. Additionally, 76% of Americans believe that robots would increase the inequality between the rich and poor.

But standing in the way of change, when viewed through the lens of history, has rarely worked. The key is to focus on the dislocated individuals and provide training to make sure that they can move into new positions. Historically, new positions tend to be more highly compensated, fueling an upward cycle.

It is clear that the pace of change and automation is increasing. In January, the parent company of Giant, Martin’s and Stop & Shop said it would introduce 500 robots to its supermarkets this year. Sure enough, if you Google “Marty the Robot” – a large, grey cone with a bright smile and “googly” eyes — you will find out that he is hard at work at 40 Stop & Shop’s in New Jersey, finding and reporting spills in the aisles and calling for a mop.

It will be interesting to see what retailers follow suit. Walmart has given robots a thumbs up. Target? A thumbs down.

The pros and cons of automation are widely written about. The pros: eliminating mindless tasks, saving money on employee costs, having a safer working environment. The cons: reducing human contact with the customer, eliminating jobs for people who need then and decreasing flexibility in the workplace as automated tasks occur at programmed times.

See also: What’s Beyond Robotic Process Automation  

As providers of workers’ comp insurance, we are watching the rise of automation in the workplace closely. One of the ways we do this is to analyze actual claims that are submitted by our insureds, which are most often small to medium-size businesses. In the restaurant sector, we analyzed over 84,000 claims. and in the retail area we looked at more than 20,000.

One area where we are convinced automation could help reduce worker injuries is in coffee shops. Workers who operate espresso machines eight hours a day are reporting repetitive motion injuries akin to “tennis elbow.” In fact, so-called “Barista Wrist” is now a recognized medical condition. Our study of workers’ comp claims in the restaurant industry found that cafés had more lost time due to injuries than any other restaurant type. And the cause of the highest number of days needed to return to work in cafés – 366 days – was due to wrist injuries.

In a parallel from the retail sector, workers in hair salons are reporting hand, wrist and arm injuries from drying hair with a blow dryer, setting the stage for a new condition that could be called the “Brazilian Blow Out Arm.” Perhaps an innovative, automated “Robot Blow Out” could eliminate these repetitive motion injuries.

Among the most dangerous and expensive injuries in our retail analysis (which includes some wholesale) came from workers engaged in the preparation of meat, poultry and fish, which involves cutting hazards caused by sharp tools and machinery. The average paid claim for a worker who sustains a cut ranges from $4,200 – $7,800, depending on whether it was caused by a non- powered tool, by a powered tool or by being caught in or between machinery.

But once again, repetitive motion injuries in meat, fish and poultry preparation are by far the most expensive at $16,200 for the average paid claim. Clearly, this is an area where more automation would be helpful.

See also: How Robotics Will Transform Claims  

All of this gets us back to our original thesis that history has shown that automation is a net positive for workers which, over time, leads to people taking higher-paying jobs. Yes, jobs are eliminated, for sure. With machines, come risk and injuries, that’s undeniable. But it is also clear that robots will take over the mindless, thankless (and dangerous) jobs and likely lead to a workplace that is safer overall.

With all that said, there is one robot that I don’t want to see and that’s: “Matt, The Workers’ Comp Insurance Executive.”

The Question That Insurtech Is Avoiding

There’s a lot of it about. Insurtech and technology, that is. New ways of doing stuff. Breaking traditional distribution models and deconstructing established supply chains. Who could not be excited?

But there’s another side to this coin, and that’s the issue of established practice. Insurance isn’t a new gig, like telematics, but something that’s been around for three centuries. Some might argue even longer, as there are records of even the ancient Egyptians sharing and aggregating risk. Protecting the few by collaborating with the many.

Over the centuries, insurance hasn’t been an easy ride. What do we mean by appropriate compensation, or, in insurance parlance, by the principle of indemnity? How to deal with those at fault, or, in insurance language, the matter of subrogation.

See also: Where Will Unicorn of Insurtech Appear?

But in the old way of doing things, we all knew where we stood. Insurance contracts had evolved over decades, and where there had been differences in interpretation the legal system had sorted things out for us. There was a sort of certainty and framework to our business and a more certain relationship, even if the topic of trust remains contentious — the level of trust between policyholders and carriers has always been low, despite a degree of contractual certainty.

Now, here we are in a Brave New World of insurance. Things will never be the same because of technology, the experts say. Some say insurtech is mainly just about new distribution channels, customer management and operational efficiency, but that leaves the rest of the insurance proposition.

It feels like we’re throwing a ball onto a sports field and asking the two competing teams to sort out the rules for themselves.

Will there be winners and losers? Of course. The winners will be the legal profession, which will spend years, perhaps, discussing where the liability for death rests as a result of a driverless vehicle incident. Was it the manufacturer – as a product liability issue? Was it the occupant of the vehicle – extending the concept of occupiers liability? Was it the system administrator, which ran the system and which surely must be involved somehow? Maybe even the victims themselves: “Don’t you know you need to be more careful, with all these unmanned gadgets all around us?’”

We can’t all just contract out of responsibility. The proverbial buck must rest somewhere.

Think forward a few decades. Let’s accept that the insurance industry will have been re-engineered and reimagined, with robots, chatbots and wobots. Let’s assume that physical risk is calculated in a more granular way and that underwriting risk management is absolutely aligned to the risk appetite of a carrier. And we have somehow managed to be proactive, to have better responsiveness to climatic change and everything else. And ubiquitous devices provide us with bottomless barrels of information, from which our systems draw insight through advanced analytics.

See also: 3-Step Approach to Big Data Analytics

Someone, somewhere, will need to address the question — what does all this mean contractually to the insurance industry? After, all isn’t insurance just no more than a contract, between two parties? Or was that concept somehow lost, somewhere inside the Innovation Hub, or among the bits and bytes of technology?

Isn’t it time that someone slowed the momentum of change and had a real hard think about the legal implications for insurance?

Robots and AI—It’s Just the Beginning

You’ve probably had regular help from a virtual assistant by phone or online, assisting you with basic tasks such as directing your call or giving you your bank balance. Helpful, right? The companies that employ the virtual assistants think so, too, and are applying these AI/robotic processes to more and more of their everyday business operations.

Often called out for being slow to change, the insurance industry is beginning to catch up quickly. It’s making sweeping changes across organizations and core systems because of the disruptive emergence of insurtech. Carriers like Celina and USAA are using AI in their daily operations and reaping the benefits.

As a result, insurers are now either delivering — or are in the process of delivering — a great digital experience to consumers. Once complete, this transformation will entail an entirely new way of doing business and servicing customers.

See also: Strategist’s Guide to Artificial Intelligence

There are four main technologies to keep in mind:

Robots

Robotics is the branch of technology that deals with the design, construction, operation and application of robots, virtual or physical. They are autonomous or semi-autonomous machines or systems that can act independently.

Artificial Intelligence

AI is the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making and translation. AI is software that learns and improves. Some robots can use AI to improve their capability by learning, but that is optional.

Cognitive Computing

Cognitive computing technologies are a subset of AI. Cognitive computing “refers to computing that is focused on reasoning and understanding at a higher level, often in a manner that is analogous to human cognition,” writes Lynne Parker, director of the division of information and intelligent systems for the National Science Foundation, in Computerworld. “This is a subset of AI that deals with cognitive behaviors we associate with ‘thinking’ as opposed to perception and motor control.”

Robotic Process Automation

Insurtech consultant Celent defines robotic process automation (RPA) as a set of technologies that can automate processes that currently require human involvement. Robots replicate human behavior to conduct the tasks as a human would; robots also optimize the tasks. RPA can yield benefits when applied to the right roles. It does well supporting repetitive tasks in various environments where there is little change, often back-office support roles and tasks.

Accenture found that cost savings after deploying RPA can reach as high as 80% and time saved on tasks as high as 90%. Automating repetitive processes means tasks are completed quickly with fewer errors, opening up new opportunities for employees to focus on more customer-centric tasks.

But RPA is not the answer to everything. It does not think, reason or predict. It completes simple, repetitive tasks quickly, but it does not learn or self-improve. Developing an enterprise-wide strategy to determine where RPA provides the most value and to anticipate the organizational change that may result is the prudent approach.

The Future Is Here

IBM’s Watson and Amazon’s Alexa are early examples. Insurers already have joined the revolution. Celina Insurance Group uses an analytics-based agency prospecting tool to appoint agents in high-potential underserved areas. USAA’s “Nina” is an AI virtual assistant that chats with customers on the USAA website. It’s designed to respond to 120 questions, from reporting stolen payment cards to changing a PIN.

See also: The Big Lesson From Amazon-Whole Foods  

There will inevitably be lessons to learn from successes and failures of this first wave of robotics and AI. However, early adopters of these technologies also risk success. Investing in innovation is what will allow insurers to stay ahead of disruption and, in some cases, create it.

As robots evolve, their capabilities and applications will no doubt be vast. Just as we could not have predicted how the internet — and now the Internet of Things — would evolve, robotics and artificial intelligence will likely follow the same course.

Welcome to the Robot Revolution

The robots are coming. In fact, in many places, they have already arrived. Some consider software automation such as robo-advisors to be robotics, but there is also considerable progress in the world of the physical, tangible robots that are very similar to the ones made popular by a century of science fiction stories.

We are headed toward a future of robots all around us – on land, sea, and air; in our homes, businesses, and communities. Today, we are witnessing the first glimpses of this robot revolution. The rate of robot proliferation and adoption is astounding, which means that a future with billions of robots may not be that far off. The International Federation of Robotics reports that there are expected to be 31 million household robots in service by 2019. There are already millions of industrial robots in use; over a quarter of a million were sold just last year. Add that to the millions of drones being sold and robots in business, agriculture, and other settings, and it becomes clear that robots are delivering good value and market acceptance. Recent examples of robotics pilots and implementations demonstrate some of the future potential:

  • Takeout food delivery: There are current pilots underway in which small mobile robots deliver takeout orders from restaurants in Washington, DC and Hamburg, Germany.
  • Robotic prostheses: 3D printing and AI advances have enabled low costs and customization of robotic arms, hands, and legs. Wearable robotic gloves that allow the disabled or elderly to have hand function are now available.
  • Robotic kitchen assistant: A robot called Flippy has been proven to cook burgers at a fast food restaurant more efficiently and at less cost than humans.
  • Insurance sales: Meiji Yasuda Life will use 100 humanoid robots in branch locations to answer sales and service questions and support sales personnel.

See also: Next Big Thing: Robotic Process Automation  

At issue now is how the new wave of robots will alter risks in the world and what that means for the insurance industry. It’s easy to jump to a vision of the world of the future controlled by robots, à la the Terminator movie series. But right now, Elon Musk, among other prominent tech figures, is truly worried about an AI apocalypse in which robots and other AI driven devices run amok and destroy the world and civilization as we know it. While it is advisable and even imperative to think about these long-term possibilities and establish the right governance today, the truth is that robots are already affecting risks – both positively and negatively. Insurers should consider these aspects of a world with more and more robots:

  • Job loss: Robots are likely to replace human workers in many different professions and in many different industries.
  • Worker safety: Robots can operate in dangerous environments, where there may be toxic chemicals or otherwise unsafe conditions for humans. Robots can also work alongside humans, handling tasks that could help to reduce workplace injuries and accidents.
  • Elder/disabled care: Robots in homes and healthcare settings may allow more individuals to live independent lives and reduce the need for assisted living facilities.
  • Increased cyber exposure: Robots will collect and create vast amounts of data about the world around them. Like any other environment with electronic data, these will be subject to hacking and criminal abuse.
  • Robot-caused injuries: Malfunctioning robots in industrial or residential settings could inadvertently cause injury or death to humans. There are examples of this already, and the potential increases as robots become pervasive.

See also: Of Robots, Self-Driving Cars and Insurance  

It should be evident from these few examples that insurance coverages will need to evolve correspondingly. In some cases, the use of robots will decrease risks and can be leveraged for loss control. In other cases, new risks will emerge and will demand insurance solutions for individuals and businesses. No one can predict with accuracy how rapidly robots will be adopted and spread across the world. But wise insurers will begin planning for a robot revolution today.