As I sit in sweltering Northern California -- where it's 105 degrees as I write this -- I realize that I barely have anything to complain about, given the crazy weather seemingly everywhere else.

The biggest problem for the moment is Hurricane Beryl, which devastated parts of the Caribbean and made landfall in Texas early Monday. The storm killed at least four, knocked out power for millions of people and caused vast damage to homes and other property. 

But there are plenty of other catastrophes, too. Wildfires have created much more devastation than normal for this time of year. Heat domes have caused severe distress throughout North America, India and other parts of the world. Floods have hit the Middle East and parts of the U.S. Unusually severe storms have also crushed areas with heavy winds, hail and lightning. 

And we seem to just be getting started. In particular, Hurricane Beryl's appearance as a Category 5 hurricane so early in the season -- drawing energy from water in the Atlantic that is far warmer than normal -- suggests that all the dire forecasts about hurricanes this year may, sadly, play out.

As a recent headline in Bloomberg read, "The Era of Super-Wild Weather Is Already Here."

The Bloomberg article, from June 19, opens:

"Wildfires in Canada that burned continuously for over a year. Floods that brought Dubai to a standstill. Deadly heat blanketing the streets of New Delhi.... 

"Florida is in its second week of battling torrential rainfall so intense near Sarasota that it has odds of occurring just once in 500 to 1,000 years. Damages could top $1 billion."

The article continues:

"The weather is no longer an even roll of the dice. It’s more like throwing loaded dice that have sixes on three sides — or sevens and eights, says Katharine Hayhoe, a distinguished professor at Texas Tech University who studies climate impacts. The term 'global warming' itself suggests a kind of predictability that may no longer suit the times. 'These days I think it’s much more appropriate to call it ‘global weirding'.... Wherever we live, our weather is getting much weirder.'”

Since that article came out, headlines like these have landed in my inbox: "California's Early Explosive Wildfire Season Is Nearly 1,500% Ahead of Last Year," "Extreme Heat Deadlier Than Wildfires, California Insurance Regulator Says" and "Much of New Mexico Is Under Flood Watch After 100 Rescued from Waters Over Weekend," The Triple-I reported that claims paid for lightning strikes, which surged in 2023, totaled $1.27 billion in the U.S.  

As bad as those headline are, what really worries me is what comes next. 

The Washington Post tallied records set by Hurricane Beryl:

"Beryl became the Atlantic’s strongest June storm on record. Its intensification from tropical depression to a Category 4 storm within 48 hours was unprecedented for the time of year. When it gained Category 5 strength July 1, it did so earlier than any other Atlantic hurricane on record.

"And when it struck Texas, it became just the 10th hurricane on record to make landfall there during the month of July, according to Colorado State University hurricane researcher Philip Klotzbach. No other Atlantic hurricane in the past decade has made landfall on U.S. shores so early in tropical cyclone season, which began in June."

An article in Carrier Management says:

“'Beryl is unprecedentedly strange,' said Weather Underground co-founder Jeff Masters, a former government hurricane meteorologist who flew into storms. 'It is so far outside the climatology that you look at it and you say, ‘How did this happen in June?''”

And worse can be expected because the water in the Atlantic is so warm that it provides unprecedent fuel for major storms. 

The Bloomberg article warns that we should start thinking beyond individual catastrophes and start imagining "compound events," "where multiple disasters — natural and manmade — occur at the same time or place, exacerbating their combined impact. A prime example can be found in Texas, where high temperatures contributed to the state’s largest-ever wildfire. Abnormally dry conditions in the Canadian province of Alberta translated into an early start to fire season.

"In other cases, impacts spread across borders. In March, Saharan dust storms blew north, turning skies yellow and orange in Sicily and degrading air quality from Greece, through Italy, to France, which also saw intense rain. Spiking food and energy prices have also overlapped with harsh weather conditions, for example, magnifying the consequences of the years-long drought in Syria, Iraq and Iran."

The only short-term advice I can muster comes from my sailing days: Batten down the hatches. 

But in the medium- and long-term we need to do some serious thinking to prepare ourselves and our clients for more "global weirding."

Cheers,

Paul

We had a doozy of red ink recently that is now ready to be smothered in black.

As we get ready for 2Q2024 earnings season, which I expect to be a whopper for personal auto, let’s take a look at an excerpt from the NAIC March 4, 2024, news release focusing on personal auto: “Total private passenger auto insurance has the largest amount of direct premiums written reported as of March 4th, 2024, at $314,788,570,644, which is about 33% of all written premiums.”  

That implies that at a 65% pure loss ratio target, there is more than $100 billion allocated to running the business and profits.  After massive layoffs and writing restrictions in the past several quarters, profits will be eyepopping this summer, again.

It’s no secret that insurance companies are for-profit operations, whether they are organized as private companies, public companies, mutuals, captives or any other form of risk transfer. And anyone who makes a living on a percent-of-premium basis is wondering when there’s going to be a knock on their door, asking, “What are you doing for me next”?  

[Full disclosure, I help insurance companies improve profits and experiences.] 

See also: Modernizing Commercial Auto Insurance

As an industry veteran agent-of-change, my constant quest to unravel old ways of working, to hunt down and incorporate new data and to segment existing levels of analysis to create threads with deeper understanding and transparent explanation of contribution to loss has made me simultaneously respected and reviled, not always in equal parts. Everyone hates change, even when change is the only solution. 

A big culprit in the red ink phase we have experienced was not so much that used car values soared. It is that the industry pricing for vehicle physical damage is uncoupled from the actual cash value of a vehicle while industry claims practices are tied to actual cash value.

Since the 1950s, the tradition has been to use the base value of the manufacturer’s suggested retail price (MSRP) for setting prices, using a starting suggested value at risk that treats all the makes and models (and more recently trim levels and other standard equipment) as though there are no possible upgrades.  

The tradition goes on to predict the future value of every vehicle using a single depreciation factor table as though every vehicle depreciates in the same manner. That depreciation prediction slowly goes down from year 1 until flattening out typically between 45% to 35% of base price new between 10 and 15 years after the vehicle was originally available for retail sale. 

After 10 to 15 years, that flat prediction of value will last as long as the vehicle is insurable, unless that specific vehicle becomes “collectable” and is required to be a scheduled valuable like a collector car.

We witnessed historic red ink when actual vehicle values sharply diverted from predicted values at risk - it was like a stock market meltdown where “naked call options” were suddenly cashed in and money to cover the spread was due on demand. The obvious analytic pursuit then is to question both the current process to set the value at risk and the way we predict the future value at risk.

Who would guess a $300 billion-plus industry with over 250 million vehicles under risk management would write insurance without keeping exacting tabs on the cash value of those assets?  As we dig into the structure of the existing framework, we see many areas of being data poor (the 1950s era) and what that means in regard to recent rate taking and what comes next for the change agenda.

See also: Reducing Auto Claims by Embracing Sustainability

Spoiler alert: 

• For newer vehicles with optional installed equipment, policyholders get “free insurance” -- the predicted scheduled insurance value is less than the actual value. 
• For older vehicles still in operation at the flat part of the factor table, as the actual cash value trends to zero, the ratio of insurance value to actual value becomes large.

What this means in respect to the recent base rate increases:

• Vehicles with total MSRP above base MSRP still get “free insurance.”
• Older vehicles pay higher rates than their value suggests they should.
• Price accuracy declines as the predicted versus actual values diverge up/down.
• Price accuracy can be radically improved by just looking up the actual value as needed. (Homeowners carriers use insurance-to-value accuracy methods already).

What’s next?

• Breakthrough innovations revolve around using data more than technology.
• Every data process can be segmented for capturing more value -- more accurate predictions, easier task accomplishment, improved choices, cheaper costs and improved margins.
• Newer data frequently creates the largest scalable impact, but deeper levels of refined analytics are the catalyst for the most rapid value creation.

No good deed goes unpunished when there are hundreds of billions of dollars in play, yet ultimately what is good for customers ends up being good for companies and shareholders. Especially at auto insurers -- they have unique traditions from other lines of insurance, but 100 years of tradition unhampered by progress is coming to an end.

Can the insurance industry rise to the challenge of giving businesses a safety net for their AI usage? In a recent World Economic Forum report, nearly 1,500 surveyed professionals identified AI as their organization’s biggest technology risk. Insurers, however, could view AI risk mitigation as a meaningful business opportunity. Deloitte projects that by 2032, insurers can potentially write around $4.7 billion in annual global AI insurance premiums, at a compounded annual growth rate of around 80%.

To get there, many insurance firms will likely need to build and expand their capabilities—and soon. To put things into perspective, there are estimates that AI could add over 10%—or roughly $12.5 billion—to global GDP by 2032. In the next few years, society may be hard-pressed to find any aspect of daily life that does not have an AI engine in the background.

However, this revolutionary technology is not without both anticipated and unforeseen risks. Consider the following scenario: In the not-too-distant future, a person could take their self-driving car to a doctor’s appointment to get an AI-assisted diagnosis; a few weeks later, they could have AI-assisted surgery and eventually file an insurance claim through an AI chatbot. A lot of things can go wrong in this scenario; the autonomous car could bump into another vehicle, the initial diagnosis could be incorrect, or the chatbot could reject the valid claim outright. The risks stemming from AI in this example could range from a significant financial loss to a potential fatality. And while some of these risks may seem futuristic, they are already starting to materialize.

Liabilities arising from use and development of AI can potentially be both significant and unpredictable. However, in today’s competitive market, business leaders may feel pressure to adopt AI technology, despite the risks of diving into unknown territory. Consequently, leaders are often seeking security against unforeseen events. 

See also: How AI Is Shaking Up Insurance

Currently, a number of AI solution vendors are providing some safeguards for their AI products, including indemnification from legal claims made against output of their generative AI tools. However, the current and future anticipated velocity of AI development could affect the magnitude and variety of risks that can unfold and may go beyond what can be managed by a few corporations on their own, particularly those that may have already been on the receiving end of lawsuits. 

From generative AI alone, businesses could face losses from risks such as cybersecurity threats, copyright infringements, wrong or biased outputs, misinformation or disinformation and data privacy issues. Having an insurance policy to protect against such issues could help assuage concerns and even encourage further AI adoption at scale. 

Regulators globally are likely to soon demand safeguards and risk management practices around AI use, which will likely include insurance coverage.

The European Union is developing the world’s first comprehensive set of regulations governing AI and has provisions for fines up to $38 million. Several U.S. states have also introduced bills or resolutions governing AI. At the federal level, President Biden issued the Executive Order on Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. Even if these regulations do not mandate insurance, provisions for hefty fines may drive companies to seek insurance coverage for these risks. Another factor that could compel businesses to seek insurance coverage would be if there were an increase in the severity and frequency of AI damages and losses. 

A few large reinsurers are already participating in the AI insurance market. Munich Re rolled out a specific AI insurance product, primarily meant for AI startups in 2018. They also launched coverages for AI developers, adopters and businesses building self-developed AI models. Several insurtech startups are also beginning to operate in this space. Armilla AI launched a product that guarantees the performance of AI products. 

That said, the lack of historic data on the performance of AI models and the speed at which they are evolving can make assessing and pricing risks difficult. Insurers entering the market are developing in-house expertise and proprietary qualitative and quantitative assessment frameworks to better understand the risks inherent to these AI systems.

See also: Cautionary Tales on AI 

Most insurers are expected to follow a wait-and-watch approach, looking at large global carriers as they establish some pricing and loss history. Examining lessons learned from cyber insurance, carriers will likely demand stringent risk management practices and guardrails to limit their liabilities. Carriers may also rely on model audit, attestation firms and other outside AI expertise for help in understanding the “black box” better before pricing it. 

As the world continues to evolve, new risks will emerge. In their role of providing coverage for a wide range of risks, insurers will be called on to design protection and trust in a society where AI is pervasive. 

The 2023 Global Wealth Report has predicted that global wealth will increase by 38% over the next five years, with the number of high-net-worth individuals (HNWIs) likely to increase to 85 million by 2027. That would be a 71 million increase from the start of the century. The ultra-HNWI (UHNWI), with wealth exceeding $50 million, is projected to rise to 372,000 by 2027, with more than half living in the APAC region. The growth of this market presents a significant opportunity for insurance companies to reshape and reimagine how they optimize their clients' wealth.

One way advisers can help clients is through life insurance. Life insurance can be an intelligent and reliable wealth management strategy that can help mitigate risk for HNWIs at a time dominated by uncertainty, continued inflation concerns and significant financial complexity. It can allow HNWIs to maintain their focus on diversifying and extending their portfolios while allowing for increased asset protection and risk mitigation. 

A variety of different approaches may be of interest to them, depending on their needs and circumstances. Life insurance can be used as a financial vehicle that helps preserve wealth while building and securing it for future generations. It is a form of estate planning that allows for HNWIs to help ensure their assets are distributed and managed in accordance with their wishes. Depending on the type of insurance plan, life insurance can also accumulate a cash value over time that can be accessed while the client is still living to use as supplemental income or even cover unexpected costs. 

Life insurance can be incorporated into a wealth transfer strategy using an irrevocable life insurance trust (ILIT) that designates the trust as both the owner and the beneficiary of the insurance policy. As HNWIs have a significant estate, they may be subject to estate taxes. Mitigating the estate tax burden is a core value of a well-structured life insurance policy, as it delivers exemption from estate taxes on the death benefit and ensures that the estate has the necessary funds to cover estate taxes and other expenses. An ILIT is a smart strategy that delivers immense value.

Individuals looking to keep the insurance out of their estate may also consider alternatives such as a beneficiary defective inheritance trust (BDIT), which is a type of irrevocable trust; life insurance limited partnerships (LPs), which limit the personal liability; or a dynasty trust, which enables wealth to be passed on from generation to generation. 

See also: Is 2024 the Year of Digital Health?

HNWIs can also better manage the financial future of their estate and beneficiaries by placing a percentage of their premium into a cash-value account that accumulates on a tax-deferred basis. These funds can also be accessed on demand without incurring taxes. However, assets in retirement accounts will be taxed, so HNWIs should consider leveraging low-income tax rates by shifting to life insurance to help offset the risk of increases in taxation as of 2025. 

Life insurance can also help to ensure equal distribution of the estate among all beneficiaries. If there is a family business or real estate that isn’t easily divided among recipients, life insurance can be used to distribute inheritance fairly among specific, named beneficiaries. This means that business continuity is maintained posthumously and that there is a smooth transition of ownership while ensuring that surviving family members are provided with financial stability. 

The same approach can be applied to charity commitments. If an HNWI wants to nominate a specific charity as a policy beneficiary, life insurance can be structured to provide continuing contributions for organizations they're passionate about.

Having an adequate policy in place can provide peace of mind for clients and their beneficiaries. This also allows the HNWI to spend more time focusing on other financial plans and goals they may have while they’re still living.

The payouts from life insurance can be extensive with HNWIs, so avoiding unnecessary taxation is invaluable. However, there may be federal and state taxes added to life insurance payouts in some instances. It’s key that HNWIs have a trusted, expert financial professional to help them navigate the risk and structure of their policies and help to ensure that the policy complies with all state and federal tax regulations. 

See also: Revolutionizing Life Insurance Uptake in Younger Markets

A specially designed life insurance developed by a financial adviser with clear visibility into tax-advantaged growth, flexibility, estate and wealth management and policy growth can help ensure HNWIs maintain visibility and retain value within their estates. The key is to collaborate with a company that has a proven, in-depth understanding of the nuances of this environment, the legal requirements and the structuring of ILITs. 

While clients bring their portfolios to advisers who have the professional expertise to manage their wealth, they are also looking for partners who genuinely have their best interest in mind. Building trust and a good rapport with a client can be just as important as having a professional background. 

The right company can help ensure that an HNWI’s insurance policy is tailored to individual needs and align with broader financial goals and estate planning objectives. HNWIs can optimize and strategize within clear frameworks that deliver a lasting impact and allow for them to navigate the future with confidence as they have a robust financial plan in place to address their unique circumstances.

With the advent of artificial intelligence (AI), there is a significant opportunity to enhance the efficiency and accuracy of workers' compensation processes. This is particularly true for injury care assessment and predictive outcomes, determining time to maximal medical improvement, settlement pricing and tracking patient clinical recovery.

Predictive Outcomes in Injury Care Assessment 

The assessment of injury care in workers' compensation has traditionally been fraught with subjectivity and variability. Different medical practitioners may provide varying diagnoses and treatment plans for similar injuries, leading to inconsistencies. Additionally, delays in obtaining appropriate care and treatment authorizations can hinder recovery and prolong the workers' compensation process.

AI-driven solutions offer a way to overcome these challenges. Machine learning algorithms can analyze vast amounts of patient data, identifying patterns and trends that may not be apparent to human evaluators. Predictive analytics can forecast injury outcomes and recovery paths based on historical data and individual patient characteristics gleaned from hundreds of thousands of patient exams and workers’ compensation claims.

For example, AI systems can assess the severity of an injury and suggest the most effective treatment plans based on numerous cases involving injuries of a similar nature to the body part in question, thereby improving the accuracy and consistency of assessments. Real-world implementations have shown that AI can significantly enhance the speed and precision of injury evaluations, leading to better patient outcomes. One doctor assembling one assessment may or may not produce an evaluation that all parties can agree on, However, “averaging out” the outcomes of several dozen adjacent cases give all parties a better chance at arriving at a logical, sensible and fair conclusion. 

The benefits then cascade to all stakeholders from there. Insurers and employers achieve more accurate and consistent injury evaluations, reducing the risk of disputes and litigation. Healthcare providers can develop more effective treatment plans, improving patient recovery times and satisfaction. Ultimately, injured workers receive quicker and more effective care, facilitating their return to work.

See also: How AI Could Set Premiums in Real Time

Arriving at MMI Using AI

The time to reach maximal medical improvement (MMI) varies widely based on several factors, including the body system affected, geographic location and job duties. These variations present significant challenges in setting realistic and achievable recovery goals. Additionally, recovery rates can differ based on the quality and availability of medical care in different regions. 

Again, the odds of one physician nailing the appropriate time to MMI are smaller than a report that removes the subjectivity of a single individual by synthesizing the results of thousands of like cases.  AI can analyze data from historical cases and identify key factors influencing recovery timelines. Custom algorithms can consider geographic and occupational variables to provide more accurate predictions.

For instance, AI tools can analyze data from previous similar injuries to predict how long it might take for a worker to reach MMI, factoring in their specific job duties and location. This information can help in setting target recovery goals that are both realistic and achievable.

Armed with this data analysis, employers and insurers can develop more personalized and efficient recovery plans, ensuring that resources are allocated appropriately. Healthcare providers can set more realistic expectations for patients, leading to better planning and care delivery. Overall, the efficiency in reaching MMI improves, benefitting all parties involved.

The (Settlement) Price Is Right

Settlement pricing in workers' compensation involves determining the financial value of an injury claim, which can include stipulations for future care and compromise and release agreements. Accurately pricing settlements is challenging due to the variability in injury severity and long-term care needs.

AI can significantly enhance the accuracy of settlement pricing by evaluating historical settlement data and using predictive models to forecast future care costs. These models can analyze various factors, such as the type of injury, the affected body system and historical treatment costs, to provide more precise pricing. 

For example, AI tools can generate predictions for future medical expenses based on current trends and past data, ensuring that settlements reflect the true cost of care. Thus, we may already be able to determine from similar previous cases the associated costs of surgically repairing and rehabbing an MCL tear in the elbow that results from repetitive heavy lifting of boxes. This leads to more equitable and fair settlements for all parties involved.

AI-driven settlement pricing benefits stakeholders by providing more accurate and equitable financial evaluations. This reduces the likelihood of disputes and facilitates quicker resolution times. Insurers and employers benefit from increased transparency and trust in the settlement process, while injured workers receive fair compensation reflecting their future care needs.

AI  Paves the Road to Clinical Recovery

Continuous monitoring of patient recovery is crucial in workers' compensation to ensure that injured workers receive the appropriate care and support throughout their recovery journey. Traditional methods of tracking patient progress often fall short in providing real-time insights and comprehensive evaluations.

AI can revolutionize the tracking and interviewing process by offering real-time return on investment (ROI) analysis in treatment care. Voice packet sampling, an advanced AI technique, can classify a patient's recovery momentum, gauge their opinion of care and assess their satisfaction with the recovery process.

For instance, AI can analyze voice recordings of patient interviews to detect nuances in their tone and language, providing insights into their emotional state and recovery satisfaction. This data can help healthcare providers tailor their approaches to better meet patient needs.

AI-driven tracking tools can also predict the likelihood of litigation or long-term complications by analyzing patient data and recovery patterns. This information is valuable for determining future care needs and supporting decisions regarding return-to-work expectations or alternative job placements.

The benefits of AI-enhanced tracking and interviewing methods are substantial for all stakeholders. Once again, employers and insurers can anticipate and manage potential complications and reduce litigation risks. Healthcare providers gain real-time insights into patient progress, enabling more responsive and effective care. Patients benefit from personalized attention and improved satisfaction with their recovery journey.

See also: How to Enhance Workers' Comp Outcomes

Conclusion: AI Improves the Workers’ Comp Lifecycle From End to End

The integration of AI into workers' compensation processes holds tremendous potential to enhance outcomes and efficiencies for all stakeholders. From improving injury care assessments and predicting recovery timelines to accurately pricing settlements and tracking patient progress, AI offers tools and solutions that address long-standing challenges in the workers' compensation system. By leveraging AI technologies, insurers, employers, healthcare providers and injured workers can achieve better results, reduced costs and improved satisfaction across the board. The future of workers' compensation is undoubtedly brighter with AI at its helm, promising a more efficient, fair and patient-centered approach to managing workplace injuries.

When I spoke at an AI event at the Insead campus in San Francisco not long ago, a fellow panelist told the scariest story I've yet heard about an AI-based scam. He said a man had gone to the door of a friend's parents and said their son was in jail and needed $8,000 to post bail. As proof, the man pulled out his phone so the parents could hear their son's voice in a frightened voicemail. The frantic parents rushed to the bank, withdrew $8,000 and handed it to the man to go bail out their son. 

The son was not, in fact, in jail. The man had somehow gotten a sample of the son's voice and had run it through a rather inexpensive AI system that generated the voicemail. He counted on the parents -- sophisticated people in their 70s, according to my fellow panelist -- to lose their bearings long enough for him to scam them out of $8,000. And they did.

As we all salivate about how generative AI can make insurance radically more efficient and effective, let's take a moment to appreciate how malevolent hackers are transforming their businesses, too, in ways that endanger not just cyber insurers but all of us.

As a recent Wall Street Journal article reports:

"Artificial intelligence is making scammers tougher to spot. Gone are the poorly worded messages that easily tipped off authorities as well as the grammar police. The bad guys are now better writers and more convincing conversationalists, who can hold a conversation without revealing they are a bot....

"ChatGPT and other AI tools can even enable scammers to create an imitation of your voice and identity. In recent years, criminals have used AI-based software to impersonate senior executives and demand wire transfers....

"Criminals today are faking driver’s licenses and other identification in an attempt to open new bank accounts and adding computer-generated faces and graphics to pass identity-verification processes. All of these methods are hard to stave off."

The improved tactics by hackers arise amid generally good news for cyber insurers. Risk & Insurance reports that "cyber insurance loss ratios have steadily declined... from a peak of 66.9 in 2020 to 41.6 in 2023. Improved cybersecurity practices by insureds and refusing to pay ransoms in over 70% of cases have reduced claims severity, more than offsetting the higher frequency of ransomware attacks."

But no one is declaring victory. Risk & Insurance also reports that Marsh clients in the U.S. and Canada reported a record 1,800 cyber claims in 2023 and that "the median ransom demand soared to $20 million in 2023 from $1.4 million in 2022,... Similarly, the median extortion payment skyrocketed to $6.5 million in 2023 from $335,000 in 2022."

An article in Wired said ransomware victims paid more than $1 billion to hacker gangs in 2023 and raised the prospect that they could escalate their attacks into threats of physical violence against those who refuse to pay. Hacking has already led to the deaths of dozens of patients in hospitals, the article said, and the gangs have all the personal information they need to locate those they're threatening. 

Although cyber insurers have greatly improved their modeling, the models haven't really been put to the test yet, so insurers need to be careful about relying on them. Cyber insurers also know they are heavily dependent on reinsurers, who could back off if the market faces some major shock.

So we can congratulate ourselves on the growing sophistication of the cyber insurance market -- but should probably do so quietly and tentatively. Plenty could still go wrong, both for the insurers and for us as individuals, as hackers use AI to become ever more sophisticated.

The Wall Street Journal article quotes a cybersecurity expert as offering this caution: 

“'Your spidey senses are no longer going to prevent you from being victimized.'" 

Cheers,

Paul

Paul Carroll

In the U.S., a backlash against telematics may be taking shape, along the lines of, “Your car is spying on you.” How big a problem do you think that will be?

Matteo Carbone

Let's separate perception from reality. If we look at the perception in the insurance community, in the OEM community [the big car makers, known as original equipment manufacturers], sure, this is perceived as a big issue. But if we look at the reality, I'm not convinced that the issue is so big.

First, after the problem with General Motors was highlighted in the New York Times, GM said it discontinued providing data on drivers to data brokers. And Verisk shared in their earnings call that this service only accounted for $1 million in revenue, so we’re talking about nothing.

In my research, the number of policyholders who share driving data through an opt-in program is about 19 million. How many of them did it through OEMs? Less than a million. Almost everyone is sharing data through an insurer’s own mobile-based app after opting in to its UBI [usage-based insurance] program. There is no comparison. One is an elephant. The other is a small insect.

Second, I don’t think drivers are all that concerned. They only get concerned about privacy when the value proposition is not interesting enough. When you do customer surveys and ask someone, “Are you concerned if a company sells data about you?”, they say, “Yes.” But if you ask, “Are you interested in communicating with a friend?” or “Would you like a discount, but to get it you have to consent to having data about you used?”, people are not concerned any more. They want the benefit. This is why so many people use social media and a growing number of policyholders are opting in to UBI programs when they switch carriers.

Obviously, there are different personas, but I'm talking about the large majority of people.

A lot of OEMs are becoming more prudent. In the last few months, I’ve heard less talk about how they want to be a data provider. But in the future many data sources will be available.

Paul Carroll

Tell me more about what that future will look like.

Matteo Carbone

Progressive has demonstrated that the data works super well for pricing sophistication. if you look at the delta between the loss ratio for Progressive and the loss ratio for the rest of the market, it was five percentage points 10 years ago, was between 10 and 15 points for the following seven years and has been consistently above 20 points over the past two years. Over these years, they have increased the relevance of UBI in their portfolio and raised the sophistication of the telematics tariff.

But we can be seen as too inquisitive and as the bad cop. You remember when the Wall Street Journal did an article five or six months ago about homeowners insurance companies using aerial data about rooftops. We’re chasing the policyholder and saying, “I may penalize you.”

It is legal. Insurance commissioners allow this behavior. Insurance companies are one of the few categories of companies that can use consumers’ credit scores. This allows carriers to use data as a way to provide more affordable insurance to many.

But I foresee a future where insurers will go beyond the inquisitive bad cop and be more friendly. Any carrier can obtain pricing sophistication by offering an insurance proposition such as, “You provide me verified information that you are better than the average policyholder, and you will receive great benefits.”

Paul Carroll

I know you have an example of a program that you think really gets the approach to customers right.

Matteo Carbone

If you look at the first article in the New York Times, there were many comments from readers. Some said that GM’s tracking of driving behavior was just the first step and that, before long, my health insurer will be looking at what I purchase at the grocery store.

The funny thing is that in the Vitality program in many markets, one of the verified types of information you can provide is from swiping your loyalty card at the grocery store. You receive benefits if you show that a large part of your basket consists of healthy products.

In Vitality programs around the world, people buy the health or life insurance product, then pay a fee to enroll in the reward program. Then you need to show verified information to obtain benefits.

Vitality launched more than 25 years ago in South Africa and today is in use in more than 40 markets, from Japan to continental Europe. In the U.S., John Hancock is their partner. In South America, it’s Prudential.

The Vitality approach is the right one: “If you want to share verified information with me, I may be able to offer you benefits.” This makes policyholders crave to share the data.

But there needs to be a big change in the marketing of UBI. Let's be honest, the storytelling for UBI products hasn’t changed a bit in 15 years.

Paul Carroll

Beyond the shift in storytelling, what else needs to change?

Matteo Carbone

Telematics and UBI have been used in the U.S. as a booster for new business for the past 20 years. But only between 10% and 15% of auto policyholders change insurance each year, and frequently it’s the same customers who move around. So, by design, you are focusing on a fraction of the market.

If you look at the total number of policies that share data with an insurer compared with the total number of policyholders in the U.S., we are talking about less than 8%. That’s not bad. But if you look at how many new customers choose UBI when buying insurance through the direct channel for the top 10 providers of personal auto insurance, the figure is more than 40%, for some up to 70%.

But the direct channel is not the main one for many insurers. They use agents, and the way UBI is currently designed doesn’t offer them any benefits. Insurers are telling agents, “We have a great program for you that can offer your clients a 30%, 40%, 50% discount if they drive well,” but that means a 30%, 40%, 50% drop in the commission to the agent.

If you expose the customer to the offer, they like it, but why would agents promote the UBI offer?

Insurers need to provide an incentive to agents.

Insurers also need to think about the vast majority of clients who don’t switch every year. Why limit UBI to new customers? Why not offer it to everyone? Without touching the pricing, only to provide a better claim experience and behavior change programs?

Paul Carroll

Terrifically insight, as always. Thanks, Matteo.

According to a recent report, 38% of insurance CEOs are embarking on generative AI initiatives, and 34% of surveyed companies have already incorporated this technology into existing workflows. While the rapid development of large language models leaves room for discoveries, this survey and others present a clear picture of generative AI’s potential value for the insurance industry.

In workers’ compensation, generative AI has the power to transform claims management by improving decision-making accuracy, enhancing documentation and, most importantly, freeing time for adjusters to focus on helping injured workers get the care they need quickly. This will ultimately accelerate return to work and protect an employer’s brand. Generative AI can equip staff with more robust data, leading to enhanced predictive models to optimize and reduce emerging risks. 

Let’s explore where generative AI makes the greatest impact in claims management and how AI-friendly workplaces can augment human insights and empower claims professionals to work at their highest professional standard. 

See also: The Dawn of Gen AI In Insurance

Revolutionizing Efficiency

A chief insight of the report is the belief held by nearly 80% of respondents that generative AI will significantly increase operational and process efficiencies. This capability stems from the power of large language models (LLMs) to digest vast amounts of information, detect patterns often hidden to the human eye and generate insights and reports. As adjuster workloads have historically been composed of manual processes, the power of generative AI to streamline claims processing represents the transformation of an entire industry. 

When adjusters receive a claim, the information found in the accompanying medical records tells a story that will determine the direction of the claim and the resulting next steps. In current cases, generative AI tools can read and digest this complex medical information, produce concise summaries of key highlights and propose actions. This utility is taken one step further by the power of LLMs to extract unstructured data from medical documents for quick updates on claims management platforms. This helps ensure that all stakeholders are apprised of the status of the current claim. In addition, adjusters can leverage generative AI’s rapid processing capabilities to answer queries in context, sidestepping searching what could be copious amounts of documents for answers. 

The efficiencies realized from incorporating generative AI tools into claims workflows translate into time savings, increased productivity and more accurate decision-making. By automating repetitive tasks, professionals can focus on more complex cases, enhance customer service and satisfaction and ultimately improve injured worker outcomes. 

Improving Outcomes 

When adjusters automate and streamline high-volume, manual claims processing tasks with generative AI, they can devote additional time to strategic work, such as directly engaging with injured workers, conducting comprehensive investigations and delivering personalized assistance throughout the claims journey. These personalized interactions and faster claims processing times enable injured workers to be quickly connected with care resources and embark on their recovery journey. In addition, by leveraging generative AI’s power to derive insights from medical documentation, adjusters can equip clinicians and insurers with pertinent information about workers’ injuries and health history, further removing communication silos between key stakeholders on workers’ care teams. 

Experiencing delays in getting an appointment and navigating insurance processes are common challenges many encounter within the U.S. healthcare system. For injured workers who may be unfamiliar with the workers’ compensation process, these poor experiences can easily be compounded, leading to fragmented care and delayed recovery. By accelerating claims processing and streamlining communications, generative AI places injured workers at the center of the claims journey and influences better experiences, enhanced care outcomes and faster return to work.

See also: How Gen AI Changes Everything in 2024

Transforming the Role of Claims Adjuster   

Generative AI not only has the power to transform injured worker outcomes but can also revamp and expand the role of the adjuster. As adjusters leverage advanced technology to equip injured workers with care, these professionals are empowered to understand their role as more than just processing documents; they can effect real change in the lives of injured workers. Over the next five years, as adjusters increasingly rely on generative AI to handle routine tasks and streamline processes, we will see the role of the adjuster focus on highe-value tasks such as strategic decision-making and customer relationship management. 

To support adjusters’ evolving role, third-party administrators (TPAs) and insurers will be called to prioritize generative AI training and education to ensure the effective use of these tools. Learning models such as hands-on workshops, online courses and continuous skill development programs will enhance understanding and proficiency in using AI for claims management. The key to promoting AI best practices is highlighting the importance of using AI as a supportive function rather than a worker replacement. This perspective, known as “augmented AI," refers to using AI technologies to enhance human capabilities rather than replace them entirely. While AI has the benefit of streamlining processes and providing actionable insights, human oversight and empathy remain critical to claiming success. 

Considerations and Looking Ahead

As with any new process or implementation, TPAs and insurance leaders can expect to encounter challenges with integrating generative AI into claims processing workflows. These challenges include data privacy concerns, ensuring the accuracy and reliability of AI-generated insights, employee resistance to change and regulatory compliance considerations. Leaders can navigate these obstacles by investing in robust data security measures, providing comprehensive training and support, collaborating with regulators to ensure compliance and, most importantly, proactively addressing concerns. 

The future is bright for generative AI in claims management, and we can expect long-term uses to include automation of the entire claims processing cycle. While industry-leading TPAs are already revolutionizing efficiency in medical document review, future iterations may further automate claims processing tasks, learn from past claims to generate new information, enhance fraud detection and improve customer service through AI-powered chatbots. 

Have you ever found yourself hastily typing during a client call, trying to complete the system’s mandatory data fields and bypassing fields to save time? You might find yourself thinking, “I’ll go back and fill that out later.” But when this happens repeatedly and across the entire agency, it can lead to a data cleanliness issue.  

A study by IBM reports businesses can lose up to 20% in revenue for poor data quality. Simply put, better data quality can influence every part of an agency’s business, from client retention to overall growth. 

A data management plan is the systematic and strategic handling of large amounts of data that agencies generate, collect and process in their operations.

The benefits of a successful data management plan can include:   

• Clear visibility into cross-selling opportunities.
• Renewal improvement with auto-delivered information.
• Insights to improve client retention.
• Improvement in attractiveness as a potential acquisition.

Investing in a data management plan ensures that agencies can harness the full potential of their data, turning it into a powerful asset for informed decision-making and strategic growth. 

With gamification, agencies open the door to a fun and collaborative learning environment that benefits everyone.    

Understand the reasoning behind the game 

It’s understandable to push data management aside for another day. Some agents may ask why they should take that extra minute at the end of a client call to triple check if data fields are filled in correctly. It’s a fair question.  

This is where gamification kicks in. Gamification incorporates game-design elements and principles into non-game contexts to enhance user engagement, motivation and participation. This concept can be applied to a data management strategy to help keep data clean and foster a culture of accountability.  

Get everyone ready to play by first showing the necessity of data management. Ditch the Power Point presentation and instead amp up opportunities for Q&A, use fun GIFs to make a point or post interactive polls to gauge feedback and keep everyone engaged. 

Clear and transparent information will break down barriers. Once the team understands the importance of data cleanliness, dig in to understand exactly how data mistakes happen in the first place.

See also: 'Data as a Product' Strategy  

Uncover data challenges and turn solutions into wins

Data management challenges often arise from simple human circumstances, such as time constraints and misunderstandings, rather than intentional negligence. The term "field hijacking" refers to incorrect data entered into a mandatory field just to bypass requirements. An example of this includes entering “9999” as a legitimate SIC Code—meaning non-classifiable—rather than taking the time to find the correct SIC classification. Other examples are purely mistakes, such as keying in the wrong ZIP code or falling victim to the commonly known "fat thumb problem," accidentally hitting the space bar and creating a blank character in a field.  

Understanding the challenges up front will help staff know what to look for. They will also be better positioned to motivate coworkers and win rewards throughout the game. 

Organize teams and stay competitive     

The good news is that bad habits can be broken. 

To create the shift in approach and stay focused on gamification, incorporate game updates in company-wide meetings and showcase clear outcomes as data management improves. Within the updates, focus on three key pillars of strategy to keep messaging consistent and then repeat these pillars over and over again in messaging throughout the gamification process. The pillars can be simple, such as “achieving data accuracy,” “trustable reporting” or “integration-ready data.” Granular objectives for each pillar can be further annotated to help drive action.

After the three pillars have been established, develop a framework tailored to your data management needs by defining goals, rules, rewards and feedback mechanisms. Pick a fun name to brand the initiative, like "Datapalooza,” then create teams or appoint individual stewards. For example, create a team around each of your account executives. Each team will be tasked with initiatives catered to your data management framework, like removing blank spaces and filling in required fields. 

Track progress and share rewards 

Throughout the gamified data management process, use a data analytics tool to track progress and reward staff accordingly. A platform designed specifically for data management will make the process a breeze, with one easy-to-view dashboard. Through this software, you can pull weekly reports to check on team status, view areas for improvement and create visually appealing presentations to share during company-wide meetings.   

Continue to provide incentives to staff by offering rewards for team winners. Free lunches or Amazon gift cards inject healthy competition into the process, making data cleanup a quick weekly ritual rather than a daunting task. Engagement and accountability will naturally rise when infusing gamification into the data cleaning process. Keep the gamification elements fresh and exciting and regularly update challenges and rewards based on user feedback and performance data. 

See also: How External Data Is Revolutionizing Underwriting

A continuous process to keep momentum 

One-and-done training sessions usually fall flat. Implementing a continuous process with motivational incentives will give employees a better understanding of the importance of clean data. Keep up the conversation through lunch and learns and promote progress frequently through multiple internal channels. 

Don’t kick data management down the road. Valuable insights are guaranteed to drive business growth.