In the ever-evolving landscape of the automotive industry, Tesla has carved out a unique niche not only as an electric vehicle (EV) manufacturer but also as a disruptive force in areas such as autonomous driving and energy solutions. With their sights set on revolutionizing the insurance industry, Tesla Insurance aims to leverage their technological prowess to offer innovative and personalized coverage. However, despite the impressive technology at their disposal, the road ahead for Tesla Insurance is fraught with challenges.

1. Data Privacy Concerns

One of the cornerstones of Tesla's insurance strategy is the use of vast amounts of data collected from their vehicles. While this data can provide valuable insights for personalized risk assessment, it raises significant concerns about privacy. Customers may hesitate to share detailed driving habits and personal information, especially in an era where data breaches and privacy violations are at the forefront of public consciousness.

See also: Automakers Build New Insurance Future

2. Regulatory Hurdles

The insurance industry is heavily regulated, and each region has its own set of rules and requirements. Tesla Insurance must navigate complex regulatory landscapes, obtaining approvals and complying with diverse legal frameworks across different jurisdictions. Overcoming these regulatory hurdles demands a significant investment of time and resources, potentially slowing Tesla's ambitious plans.

3. Established Competition

Despite Tesla's success in disrupting various industries, the insurance sector is already populated by well-established players with decades, if not centuries, of experience. Convincing customers to switch from their current insurers to a relatively new entrant like Tesla Insurance may prove challenging. Building trust in an industry where reputation is paramount requires not just technological innovation but a nuanced understanding of customer relationships and industry dynamics.

4. Actuarial Challenges

The success of any insurance venture relies heavily on accurate risk assessment. While Tesla's vehicles are equipped with cutting-edge sensors and cameras, developing actuarial models that accurately predict and price risks associated with EVs and autonomous driving technology is no easy feat. Insufficient actuarial precision could result in financial losses for Tesla Insurance and dissatisfaction among policyholders.

5. Economic Sensitivity

The insurance industry is highly sensitive to economic fluctuations. During economic downturns, consumer spending on non-essential services, such as insurance, tends to decrease. As Tesla Insurance seeks to establish itself, economic headwinds could pose a significant challenge, affecting their ability to attract and retain customers.

See also: Maybe OEMs Aren't Such a Threat to Auto Insurers

6. Repair and Replacement Costs

Tesla's vehicles are renowned for their advanced technology, but this sophistication comes at a cost. Repair and replacement costs for Tesla vehicles can be substantially higher than those for traditional vehicles. This could translate to higher claim payouts for Tesla Insurance, potentially affecting the company's profitability and pricing competitiveness.

7. Scalability Issues

Tesla Insurance's success will hinge on its ability to scale rapidly. As the customer base grows, so will the demand for efficient claims processing, customer support and risk management. Ensuring scalability without compromising service quality is a significant challenge, and failure to address it could result in customer dissatisfaction and reputational damage.

Conclusion

Tesla Insurance possesses a technological arsenal that sets it apart, but the road to success in the insurance industry is laden with obstacles. Overcoming data privacy concerns, navigating complex regulatory landscapes, competing with established players, addressing actuarial challenges, weathering economic uncertainties, managing repair costs and achieving scalable growth are formidable tasks.

Tesla's journey into the insurance realm is undoubtedly ambitious, but success will require a strategic blend of technological innovation, regulatory acumen and a deep understanding of the intricacies of the insurance business. The journey may be challenging, but if Tesla can navigate these hurdles, the rewards could be transformative not only for the company but for the insurance industry as a whole.

With generative AI, I've advised you to Think Big, Start Small and Learn Fast -- my mantra of going on three decades for how to approach breakthrough innovations. But don't bother with the just-introduced Apple Vision Pro headset. There's no reason to even think about it at all for now. 

Despite the best efforts from hype-meister Apple, the virtual/augmented reality device won't have a significant impact for years, certainly not on business.

The device is a technology in search of a problem. Yes, the technology is wonderful -- the improvements in the display of images are spectacular, and the ability to call up apps in space is impressive, if a bit confusing. Apple will also make the Vision Pro seem a lot cooler than the ill-fated Google Glass heads-up display of a decade ago.

But there's simply no reason to strap a 1 1/2-pound device to your face (nearly the weight of a quart of milk) and put a three-quarter-pound battery in your back pocket so you can type with your two index fingers in mid-air while strangers or officemates gawk at you. Not when some combination of today's laptops, tablets and phones will do just fine.

One reviewer says the Vision Pro does have a killer app: It makes for a great kitchen timer.

But is that enough to get you to buy a slew of them at roughly $4,500 apiece, out the door, for your business? I suspect not.

Now, Apple reportedly sold 200,000 Vision Pros before they even became available to the general public. I'm skeptical of that number. A lot of things get rumored when a hype machine is in full gear. But even if the rumor is way high, there are an awful lot of fanboys and fangirls out there who are trying to stir up enthusiasm with videos like this one of someone with a Vision Pro strapped to his face while his Tesla Cybertruck drives in fully autonomous mode. 

Here's the thing: While I'm sure that video looks like the future to some people, to me it looks like a crash waiting to happen. 

The device just isn't practical. The ability to open apps in space seems to be a big draw, because it makes the whole world your desktop. But one reviewer described having to walk around his house, looking for a document much as many of us look for lost keys. And opening a few dozen tabs on a laptop, while sometimes cumbersome, works well enough that there's little justification for switching to a whole new technology.

Typing on the Vision Pro's virtual keyboard is so awkward that every reviewer I've read said they switched to a physical keyboard, which didn't mesh smoothly with the virtual environment and which pretty much defeats the point of going virtual, anyway.

The Vision Pro is reportedly great for displaying movies -- but I don't know how relaxing it would be to sit there for a couple of hours with the equivalent of a quart of milk strapped to my face. 

Much is made of the ability to pinch two fingers together and mark an object on your Vision Pro's screen for annotation of some kind. I suppose that could be useful in some work environments... but I can't immediately think of one. Certainly, that capability doesn't fit well with the sort of knowledge work that insurance companies do.

The capability does work well for kitchen timers. As this reviewer notes, you can pinch your fingers together on a whole series of pots and pans, on the stove or in the oven, and set a timer for each. Every time you look back at the stove, you can see in an instant where each dish or sauce stands. 

In turns out that kitchen timers have been the killer app for other, supposedly breakthrough technologies. The review cites a 2023 study finding that the timer is by far the most used app on an Apple Watch and another study saying the timer is an extremely popular feature on smartwatches, in general. Another study found that the timer was the most requested use of Amazon's Echo, even ahead of "play a song."

Kitchen timers obviously aren't what Apple is going for here, but how quickly can the technology improve?

Apple has fabulous engineers and all the money in the world. The market it's after is almost boundless, so it has every incentive to keep making the product better. 

But physics is an unforgiving opponent. While electronic components keep getting smaller and lighter, displays do not, at least not quickly. Nor do batteries -- and improvements in battery density for devices like the Vision Pro are going toward making them last longer rather than reducing their weight.

So I'd say we're looking at many years where the Vision Pro is just too clunky to justify spending on what is, at best, a marginal improvement in the computing experience.

There will still be a cool factor to the Vision Pro. There always seems to be with Apple these days. But not enough to justify a real business investment.

Cheers,

Paul

In theory, mobility is shifting toward shared use, to the detriment of the private car. In the age of multimodal mobility, insurance will mechanically follow: We no longer insure a person, but their mobility in all its forms, with a single insurance contract.

But in reality...

Desperately seeking multimodal products

Ideally, I'd like to be able to take up an annual policy that covers my personal vehicle, as well as third-party liability, personal injury insurance, legal protection and assistance when I rent a bike or scooter, and personal injury insurance when I use ride-hailing services.

Does such a contract exist? I did a little analysis (without claiming to be exhaustive), and here's what I found:

• Insurance offers covering shared mobility: Ma Mobilité AXA, launched in 2016 by AXA in France, covers personal injury, legal protection and assistance, but this product is no longer listed. MyMobility by Allianz in Italy, launched in 2019, covers shared mobility for tourists visiting the country. But in both cases, these are temporary insurances, not targeted at year-round users.
• Products dedicated to certain types of alternative mobility: Aon supports Flee in Italy, which covers (via usage-based insurance, or UBI) the rental of a vehicle coupled with a car-sharing service with pre-designated drivers. Moonshot Insurance offers insurance dedicated to green micro-mobility in Europe (e-scooters, e-bikes, kick-scooters, etc.).
• Extended cover for renting (or sharing) vehicles is quite common with traditional insurers or car-sharing platforms.
• Finally, there is Swinz, in Belgium, which offers home insurance, coupled with personal liability, assistance, legal protection and green mobility comprehensive cover. We are getting closer to the multimodal concept, even if the core of the product is actually home insurance, with ancillary cover.

In short, the dream scheme doesn't seem to excite insurers... or maybe I'm the one with the twisted fantasies... or maybe the subject is more complex than it seems.

See also: A New Approach to Embedded Insurance

For insurers, is multimodal mobility insurance even a topic?

In reality, the concept of multimodal insurance faces serious insurance pitfalls, due to the very nature of the risks exposed.

• The switch from one type of mobility to another depends on volatile patterns, such as the prevalence of public transport in a given city or part of a conurbation, existing alternative forms of mobility, the impact of weather and daily or seasonal peaks on the preference for means of transport, moral hazard in the use of means of locomotion, etc. Under these conditions, it is difficult for actuaries to model risk.
• The data linked to these different forms of mobility is distributed among different operators; they would have to be willing to share it, and insurers would have to be able to capture and analyze it. Not to mention the regulatory obstacles that the use of this personal data may pose.
• As alternative forms of mobility are most often billed on a pay-per-use basis, insurers need to align their pricing methods. CapGemini published a study in 2023 indicating that only 29% of the insurers they surveyed felt able to develop products adapted to new mobilities, and just 16% had the talents to do so.

Next come the challenges posed by the business model for insurers

• The development costs of unbridled multimodal insurance are disproportionate to the youth and size of the potential market. Hence the current vertical solutions, which only cover one segment of mobility.
• The stability of the mobility offer can be an issue, as numerous examples show that shared mobility operators can disappear because they are unable to find their business model, when it's not the public authorities who change the rules of the game without warning. Insurers need a stable ecosystem in which to assess risk.
• Obviously, the market for mobility insurance is largely covered by traditional products or by cover, embedded or not, in shared services. We are not in a situation where an obvious protection gap would lead to a flood of specific insurance offers.
• Speaking of business models, we obviously have to reckon with users' willingness to pay for insurance, which brings us to the next point.

Is there a demand for multimodal insurance?

At the outset, potential customers are faced with the challenge of understanding the risk and its need for coverage.

• I've done the test for myself: With ride-hailing services, where passenger injuries are covered by the driver's professional liability insurance; my shared bike subscription, where there is no personal injury or liability insurance included; and e-bike or e-scooter platforms, where insurance can be embedded, I need to put forth significant effort to understand what is and isn't covered.
• Empirically, I think users of mobility services fall into three categories: those who don't even understand that insurance is an issue, those who ask and try to answer (and struggle to make up their minds) and the residual minority of dangerous perverts who will model everything in Excel.
• And when this last category anticipates who to insure with, who to turn to in the event of a claim, depending on the different coverages and providers, I'm afraid the best intentions won't be enough.

And of course, there is the question of value for money.

• For the general public, buying insurance is often associated with an obligation (e.g. motor third-party liability, or MTPL) or a tangible risk (e.g. theft or fire). For an occasional service, it's more difficult to be palpable. "I don't buy insurance to take the bus, so why should I buy it for other occasional transport?"
• Secondly, you have to consider that the cost of this pay-as-you-go insurance is in addition to the insurance paid for by the year. If all insurances were charged per use, we could find an attractive pricing model for users, but that's not the case. Without incentive, it's hard to generate interest.

See also: Embedded Insurance and the Gig Economy

What's the trajectory for multimodal mobility insurance?

Clearly, this famous "mobility of the future," as the title suggests, is not yet here.

• McKinsey has evaluated the revenues of global mobility to 2030, and its message is clear, even if the figures must be taken with a grain of salt: Ride-hailing mobility will drive future mobility revenue (from $120 billion in 2019 to $450 billion in 2030), while micro-mobility would be at $50 billion in 2030). While these projections are based on the rise of robo-taxis and robo-shuttles, micro-mobility remains a niche market.
• With the digitization of cities, we're hearing more and more about multimodal mobility platforms (mobility-as-a-service). To date, however, this remains a little-implemented concept. Whatever assumptions are made about their development, public transport will remain the backbone of mobility. In this scenario, one question will guide future developments: Do public operators have the will to federate alternative forms of mobility and thus promote competition and its procession of despicable startups?
• In fact, public operators that deploy digital platforms dedicated to mobility are opening up to third-party services to a greater or lesser extent and are facing stiff competition from technological players such as Uber and Citymapper.

In this context, the opportunity for insurers is not necessarily where we think it is.

• CapGemini estimates that, by 2030, the individual car insurance market will only experience organic growth, whereas insurance for alternative forms of mobility (electric, connected, autonomous, shared) should increase eight-fold and reach 40% of the total volume.
• This growth in insurable volume will be driven by coverage for fleets (including leasing contracts), product liability (autonomous vehicles), professional liability (ride-hailing services) and embedded insurance (micro-mobility, hailed mobility).

This brings us back to my initial observations on the state of the insurance offer for multimodal mobility.

• B2C opportunities are likely to be fairly limited, essentially linked to micro-mobility.
• It is with mobility-as-a-service platforms -- i.e., around a single orchestrator carrying embedded insurance -- that true insurance for multimodal mobility has the best chance of developing.
• Before such platforms expand, an entry strategy would be to partner up with shared mobility operators, via embedded insurance, a subject already tackled by certain players, both traditional and insurtech.
• When the multimodal mobility market really takes off, insurers that are already in the market will benefit from their experience, particularly in the use of data. Perhaps this will enable them to respond to some of the insurance challenges mentioned above. But when? All bets are off!

In my opinion, the prospects for multimodal mobility insurance show a complementary dynamic to the one I've already mentioned, which means that the distribution of car insurance will -- in time -- increasingly escape traditional distributors.

The adage, “garbage in, garbage out,” is as true today as it was 20 years ago.  As Thomas Redman writes in Harvard Business Review, “Poor data quality is enemy number one to the widespread, profitable use of machine learning.”

Just because you can teach computers to help normalize data does NOT mean they can do the cleanup from years of bad habits. Whether your team is supposed to be logging information in two systems but only does so in one, whether you want to connect two disparate systems or whether you need to simply make up for a lack of data, there is work to be done before AI can take over.  

Think of where you’ve seen AI in action, like the craze for images created by AI. Sometimes they turn out well, yet sometimes they are comically bad. And nothing is funny when you’re trusting AI to help grow your business. Missing out on key pieces of information could lead to disaster. 

When we trust AI to do the work: This is an AI-generated image of an insurance agent issuing a policy to a local business owner. 

How does this translate to insurance? 

Imagine you have one system that records all customer interactions from a marketing standpoint – interactions on social media, website engagement and email tracking. You have a different system that tells you how many times they logged into your self-service customer portal (and what they did there). And of course, the customer could also be interacting directly with their carrier without your knowledge, perhaps buying additional insurance or researching whether their current insurance is good enough for them. 

Your main system may also be disconnected from the system that logs the number of policy change requests, questions about their policy, requests for certificate of insurance or auto ID cards and even when claims are made.

Your marketing engine may paint certain customers or prospects as highly engaged customers likely to buy, when, in fact, the customers who are truly likely to buy are barely being touched from an education or upsell/cross-sell opportunity standpoint.

See also: The Risks of AI and Machine Learning

Why does this happen? 

For AI to be successful, you need a data set that is normalized and “taught” with certain objectives in place. And you must have ALL required data in the data set to test your hypothesis.

You must also recognize that “AI is inherently probabilistic,” as Forrester suggests in a recent article on Bridging the Trust Gap Between AI and Impact.  

Your marketing team may assume that the most active customers/prospects on social media are the most likely to buy more insurance from your agency. However, many agencies have found that those that are the most “taken care of” during the policy change/question interaction are the ones most likely to buy additional insurance. Why? Insurance is still very personal, and in these cases the agent has created a bond of trust with that customer. Without a complete data set, you may get a distorted picture.

“But we are extracting our data and dumping it all into a data lake.”

Another tenet of AI best practices is iteration – systems need to be taught and learn through continuous feedback loops. Andrew Johnson shares more details if you’d like to dig deeper into how these work. If your data is disparate, even if you are extracting it, normalizing it and dumping it into a database, the feedback loop remains manual. At best, you are guided by the observations of your firm’s management team, not complex analysis of actual user behavior. 

With quality data, AI can begin making well-informed suggestions. Going back to our example, a system might recommend upsell opportunities based on complete information, better understanding customer sentiment. This hypothesis can be tested and refined through multiple iterations of conditions based on real-life relationships with the customer.

So, can AI solve our underlying data problem? The unfortunate truth is that you can’t even begin to use predictive analytics without good data.

But there’s so much you can do… once the data is right. If garbage in is garbage out, just imagine what the output could be when you start with good data.

For an agency ready to take on the future of technology and allow AI to work for its business, you must make sure the initial data set is clean, complete and housed in the same environment where your users interact with your business. Then, you can really start making progress.

As we step into 2024, the landscape of the cyber insurance industry is poised for significant evolution. The global market is expected to continue its upward trajectory, driven by increased pressure from threat actors and insurtech innovations geared toward providing both tailored coverage options and services to curtail emerging cyber threats like ransomware assaults and data breaches. 

Catastrophic cyber events are likely to become a major topic of conversation for both insurance providers and their customers. The recent decision made by the Federal Insurance Office regarding the potential expansion of the federal program for cyber catastrophic events will also influence and shape the industry. Insurers are preparing for coverage changes, too, potentially addressing specific areas such as coverage for cloud outages, significant software vulnerabilities and other widespread cyber events.

Moreover, modifications to the "acts of war" exclusions may significantly alter the value of cyber insurance policies. Catastrophe bonds are expected to expand and play a crucial role in risk transfer, with the participation of multiple new entrants.

See also: Cyber Insurance at Inflection Point

In parallel, several trends continue to shape the landscape of cyber insurance, which holds the crown as the newest specialty P&C product on the market. While the product initially entered the market out of necessity, particularly due to businesses grappling with cyber events within their business insurance, its evolution is still underway. The most important trends that will shape our world this year are:

• Growth: a growing demand spurred by the escalating sophistication of cyber-criminal activities and increased exposure to geopolitical conflicts.
• Stabilization of market rates: selective increases and decreases for specific market segments.
• Expansion: the global cyber insurance market continues to expand, despite experiencing a softening trend in the first half of 2023.
• Increase in data collection and use of AI: leading to potentially longer questionnaires for agents and policyholders. Sophisticated cyber providers will be able to collect this data without questions. AI’s rapid expansion will affect both exposures and how insurance will react.
• Reinsurance rate dynamics and how the property market and cyber market affect each other on large portfolios of risk.
• Further trends to monitor include:
  • Improved pricing structures.
  • Heightened cybersecurity requirements.
  • The growing impact of cyber insurance on a company’s initiatives to strengthen its cyber resilience.

The most prominent risks in the cyber insurance landscape include geopolitical cyber risks, ransomware threats, supply chain vulnerabilities, data breaches leading to liability issues and emerging technological trends. Conversely, the opportunities within this sector encompass the growing need for cyber insurance across a broad spectrum of businesses, emphasizing cyber risk management, the development of cyber risk modeling and the formulation of strategies for cyber risk mitigation and planning.

See also: Risks, Trends, Challenges for Cyber Insurance

The landscape of the cyber insurance industry is undergoing a dynamic shift as it navigates the complexities of emerging cyber threats and the evolving global market. The coming years will undoubtedly demand a dynamic approach to address the multifaceted challenges and opportunities within the cyber insurance industry. Despite these risks, promising opportunities are being driven by increased demand, highlighting the crucial role of cyber risk management and measures to address these evolving threats. 

More than 15 million people have already lost Medicaid coverage as states unwind pandemic-era protections, and that number could ultimately climb upwards of 30 million people. Almost three quarters (71%) of people whose coverage was terminated were disenrolled for “procedural reasons,” which can include paperwork errors. Many of these people may still be eligible for Medicaid. 

Losing Medicaid coverage can be life-threatening for low-income families and people with disabilities. Not only that, but cycling in and out of Medicaid is expensive for state governments, costing up to $773 per person. Continuous coverage for eligible populations benefits everyone by lowering costs and ensuring people have access to the healthcare they need.

Aggressive timelines for redetermining eligibility leave little room for error, especially when data on Medicaid recipients is not always accurate or organized. Many people don’t even realize that they have lost coverage, or that they need to redetermine their eligibility.

With these challenges in mind, here are three steps I believe could minimize loss of coverage during Medicaid unwinding. 

See also: How Data & AI Can Shape Group Benefits

Step 1: Use more channels for communication

Staying enrolled in Medicaid can be complicated. Low-income families tend to move frequently, which means states might not have their latest mailing address on file. More than 60% of Medicaid beneficiaries are not even aware that redetermination is happening. When letters do make it to families notifying them they must recertify, they may still struggle to understand and gather the right documents to reapply.

With redetermination stretching limited call center resources even further, people are left with little support in the process. For individuals without internet access, unlimited cell phone minutes, transportation to in-person centers or spare hours to dedicate to being on hold, this presents a unique challenge. Diversifying the communications channel mix is critical. 

Instead of relying on mailed redetermination notices, states should meet beneficiaries where they are. For example, they could reach out via text message or phone call, because someone’s phone number is less likely to change than their mailing address. 

States could also advertise consumer assistance on channels that are known to reach Medicaid beneficiaries. According to Nielsen, low-income consumers spend about nine hours a month on Facebook, which means targeted ads on social media could go a long way toward reaching them. Additionally, more than half (51%) of daytime television viewers have household incomes under $30,000, which means the audience likely captures a lot of Medicaid beneficiaries. 

Modernizing outreach efforts and shifting beyond mail-only will help minimize loss of coverage by ensuring individuals are aware of redetermination, and what steps they need to take. 

Step 2: Integrate data across systems

With the modern technology available today, there is no excuse for the volume of Medicaid paperwork problems and procedural disenrollments. Instead of relying on beneficiaries' response to a mailed letter, or live conversations, states could identify people who are still eligible by using data from other government programs. For example, people who are eligible for the Supplemental Nutrition Assistance Program (SNAP) are likely eligible for Medicaid, too.

Technology could also more efficiently manage call centers by collecting consumers' information for callbacks automatically and eliminating the need to wait on hold. Around one third of states have been warned by federal Medicaid officials that their call center wait times are too long, which can cause people to give up. Utah, for example, had an average call center wait time of 35 minutes in May, and one in four callers hung up, according to the Centers for Medicare and Medicaid Services. 

Better government data infrastructure and integration could offset lengthy wait times by either eliminating the need for calls altogether or routing people more effectively. However, not all states have the technical capabilities or infrastructure to do this efficiently. 

See also: We Must Prescribe Drugs More Accurately

Step 3: Public-private partnerships can help close the communications gap

State governments may not all have the resources or the capabilities to reach consumers where they are or to identify whether beneficiaries remain eligible; however, private businesses have different resources. 

Many consumers who lose Medicaid may be eligible to enroll in ACA or Medicare Advantage plans. The government is offering a special enrollment period for anyone losing Medicaid between March 31, 2023, and July 31, 2024, providing 16 months to enroll in a new ACA marketplace plan. Private businesses can reach these individuals efficiently to help them find a new plan and enroll. Insurance agencies and non-profit organizations can also provide much-needed support to Medicaid beneficiaries navigating the renewal process. 

Private businesses have an opportunity to step up and provide education and resources to those affected. States should embrace public-private partnerships to maximize outreach around redeterminations and help more people maintain the coverage they need.

2023 was challenging in insurance. The entire industry, specifically personal lines, has been on a roller coaster driven by a historically volatile macroeconomic environment. 

In my role at Clearcover, a next-generation car insurance company, my team and I speak daily with a variety of agencies and producers–those who have been working on the front lines of the industry and can attest firsthand to the challenges the industry experienced last year. Suffice it to say, we recognize that 2023 might not have been a career highlight for many. 

The good news: While challenges in the market will not correct themselves overnight, there’s still reason to maintain some optimism as we head into this year. But first, to ensure 2024 produces a better outcome than its predecessor, we need to sift through the lessons 2023 doled out to learn how best to move forward. 

See also: Top 5 Challenges Facing Agents in 2023

The market challenges of 2023 

We’re using the word “unprecedented” a lot right now to describe the state of the industry, and it seems to be an appropriate description. Veterans of the industry–on both agency and carrier sides–tell me they’ve never experienced a market like this. 

That's not to say that the industry didn't present opportunities for those who were willing to innovate and adapt. While connecting with agents, our team noticed that those who found themselves busier than ever in 2023 were the same agents who had implemented operational efficiencies in their agencies. They built more sophisticated automated workflows into their customer relationship management systems (CRMs) and dedicated specific staff to make renewal calls if they weren’t doing so already. They also diversified as much as possible, dedicating more marketing dollars and personnel to commercial lines insurance than they had in the past. 

Carriers, working feverishly to find profitability, began to lean on artificial intelligence (AI) as one of the major directions of innovation. In 2023, AI became much more than a buzzword as carriers implemented sophisticated predictive models to place risks. (I cannot help but be comforted knowing that AI helps us procure and analyze data that might help us predict the “2023s” of the future so we can get ahead of hard markets.) 

Because Clearcover felt some of the market impacts sooner than the average carrier, we made some early adjustments to our operational costs by using AI to save where we could. 

This is only the beginning. The market effects of artificial intelligence should only increase in 2024. As AI and machine learning tools improve, significantly refined models will evolve, optimizing predictions for target customer groups. 

See also: The Key for Agents: Lifelong Learning

What agents can do to position themselves for success 

We don’t have a crystal ball, but we are taking a few bets that 2024 will see the return of broadened customer appetites and fewer rate changes than last year. With that in mind, here are a few things I think agents should lean into in 2024: 

1. Embracing artificial intelligence: It’s easy to be apprehensive about AI, especially if you haven’t yet experienced all the ways it can help even small agencies succeed. For example, imagine a world where you don’t have to call every customer at renewal because you have an automated tool that can accurately predict who is most likely to respond to a renewal call—or who is most likely to shop at renewal based on previous behavior. This could save your agency a lot of time on the phone, which could translate to more time prospecting. Embrace and learn as much as you can about the AI landscape. It will no doubt be in our future. Educating yourself on the new technology will be key. 

2. Encouraging customers to reflect on their insurance needs: This is a great time to review and revise their coverage to fit their needs, as customers may have been (understandably) too price-sensitive to make adjustments in 2023. There could be new discounts that apply so it’s always a good idea to go over existing coverage with your clients. 

3. Refreshing your carrier relations: Understanding the evolving goals and appetite of the carriers you work with this year will help your agency be able to present as many options as possible to your customers. It will also keep your staff feeling knowledgeable, empowered and positive about the industry. If your top carriers offer a re-training or even a sit-down (virtual or in-person) with a carrier representative, this could be a great way to ensure you’re up-to-date on their appetite, offerings and technology—all of which may have been updated while new business slowed down. 

Fraud is a significant problem for the insurance industry, and companies are starting to take advantage of natural language processing (NLP) to expose it. 

Here’s what to expect: 

How Insurers Can Implement NLP for Fraud Detection

Experts say the NLP market will reach about $29 billion in 2024 and nearly double that by 2029. NLP has a place in insurance because it can often detect fraud when used in processing claims. 

NLPs gather mostly text-based client information, such as the claim description, police reports, medical records and phone transcripts. The software extracts the relevant data points and uses them to fill out claims. Without a human’s help, the NLP software can compare the data with clients’ past claims, criminal records and other crucial factors that could play into fraud detection. 

When comparing a client’s claim with similar filings with the same company, an NLP can quickly see how much correlation there is between claims and identify suspicious activities. 

The Coalition Against Insurance Fraud estimates it costs Americans more than $300 billion annually — or just under $1,000 for every person in the country. 

Other Advantages

NLPs analyze large datasets far faster than humans can, reducing paperwork for claims professionals and letting them increase client satisfaction daily by making them wait less. 

You can see NLPs helping insurance underwriters save time on routine work, too, so they can evaluate risk more precisely. 

NLPs can help address the expectation that the industry will lose about 400,000 workers by 2026. 

NLPs provide particular value when there is a backlog of claims — such as after a natural disaster. AI gives companies 24/7 support by talking with clients even when the customer service department isn’t available. In fact, contact center labor costs are expected to drop by $80 billion by 2026 due to conversational AI. Insurers can use NLPs as their support desk and reduce the need to outsource this feature. 

Legal and Ethical Issues of NLP for Insurers

Insurers must be aware of the various legal and ethical issues surrounding NLP and their ramifications for the company’s future.

Bias

Bias is one of the most pressing issues because NLP can discriminate against particular demographics if precautions aren’t in place. NLP programs can be inaccurate if the data fed into them aren't precise. NLP developers use historical data to train it, and this information could derive from biased outcomes, so NLP outcomes can exacerbate current biases. 

A 2021 Language and Linguistics Compass study finds bias occurs most often in data, models, research design, input representations and the annotation process. Some bias is typical, but excess leads to adverse outcomes for insurance companies. 

An example of NLP bias you may see is with language input. People speak English differently, and the NLP software might not be accustomed to specific dialects and accents, leading to inaccurate information. The NLP’s inadequacy could lead to an insurer wrongly denying a claim. 

While a 2023 study finds ChatGPT invariant to race and ethnicity and insurance type, there are statistically significant correlations in word frequency across race and ethnicity and a difference in subjectivity across the types of insurance. 

See also: How Technology Is Changing Fraud Detection

Client Privacy

NLP software’s reliance on personal data opens questions on clients' privacy. Any insurer will need someone’s physical address, email address, telephone number and other essential details when a claim is filed, and it’s up to the company to keep these records confidential. Unauthorized use of a client’s information, such as selling it to third parties, could breach privacy and invite lawsuits, depending on your jurisdiction.

Insurance companies must be aware of recent privacy laws that protect consumer data. Virginia, California, Colorado, Utah and Connecticut are five states with comprehensive privacy laws that insurers should be aware of. For instance, the Utah Consumer Privacy Act, enacted on Dec. 31, says consumers have the right to know what data a company collects, how it uses the information and if they decide to sell it. 

Sometimes, a privacy breach could be unintentional if a cyberattack occurs. These online attacks cause millions of dollars in damage and destroy reputations if an insurer doesn’t adequately protect its clients. NLPs draw personal attention and financial details from claims, so exposure could lead to devastating consequences. 

The 2023 MOVEit cyberattack has affected 94.2 million people worldwide and 2,730 insurance companies — emphasizing the importance of cybersecurity in the industry when using NLPs. Failure to protect clients will result in damaged reputations and possible fines from authorities if insurers don’t satisfy regulators. 

Evolving Regulations

There is little federal guidance on NLPs and what insurers can and can’t use them for. Most of these determinations are in the hands of state governments — and many have not taken action yet. However, evolving regulations will determine how much insurance companies can lean into this technology and the penalties for misuse. Following news and developments on NLP laws is crucial for insurance professionals.

Some states have taken action against AI, including various laws protecting consumers from profiling. Insurers may use AI profiling to determine whether a client is eligible for insurance coverage, but they may only do that if the customer consents to giving their information. Virginia, Connecticut, California and Colorado are four states that have implemented this policy. 

The most significant regulation yet comes from the European Union (EU), so insurers with an international presence should be aware of using NLP in their operations. Last December, EU regulators agreed to rules governing AI as developers continue to grow this advanced technology. Europe’s top governing body will ban AI systems using social scoring, biometric identification and categorization, and other unacceptable risk factors for NLPs. 

Every carrier wants to deliver an exceptional claims experience. It’s a pivotal moment, often the only one, where carriers engage directly with their customers. To make this experience count, carriers must settle the claim quickly and accurately. But at the same time, ensure only legitimate claims are processed and paid. This step, while crucial, can create unnecessary friction, slowing down the process. Nevertheless, carriers cannot afford to overlook this, as fraudulent claims cost the insurance industry billions in losses.

Consider this:

• According to the Coalition Against Insurance Fraud (CAIF), the U.S. insurance industry loses $309 billion annually to insurance fraud
• About 10% of all insurance claims in the U.S. are fraudulent, which impacts loss ratio
• In Property and Casualty insurance alone, fraudulent claims account for 5-10% of all claims with losses amounting to ~$34 billion every year

Carriers often wrestle with identifying fraudulent claims in real time, challenging their efficiency and, ultimately, contributing to higher expenses. Giving insurance professionals real-time insights into the claims they’re handling and how fast they need to be investigated can help reduce this friction.

AI-driven automation and analytics can help fix that — these newer technologies make fraud detection a seamless part of the claims process, mitigating risk and enabling exceptional customer experiences. Unlike manual checks, AI-driven automation can process large claims volumes, allowing carriers to validate an applicant’s details at the quote stage, cross-referencing multiple data sets such as claims history, names and addresses, credit information, social media data, watch list data, and public databases. Plus, carriers can draw on many more sources to verify claims, gathering and filtering data that would otherwise be hard to access in real time.

AI-generated alerts provide claims professionals with key fraud indicators and actionable next steps for further investigation, empowering them to make more informed, data-driven decisions.

Three Types of Insurance Claims Fraud

All P&C carriers grapple with fraudulent claims. But tackling this fraud costs carriers time, money, and resources.

Before we dive into how newer technologies can assist insurers in detecting fraud, let’s take a look at the three types of insurance fraud:

Exaggerated claims: The insured inflates the claim value claiming far more than the actual value

False claims: The insured makes up a story of an incident that never occurred, such as an accident or theft. Examples include false or inflated property damage, burglary or theft reports, arson, intentional damage claims, staged accidents, auto shop scams, falsifying the date or details of an accident, and so on.

Duplicate claims: The insured person raises more than one claim from the same insurance provider for the same event/incident  

Challenges with Traditional Fraud Detection

The traditional approach to fraud detection relies on developing heuristics around fraud indicators, in one of two ways: 
1. Pre-defined rules determine if the case needs further investigation
2. Other times, a checklist is prepared with scores for various fraud indicators. These scores and the claim value will indicate if the case needs to be investigated further.

The problem with these methods is they rely heavily on manual intervention and have the following limitations:  

• Constrained to a limited set of known parameters based on heuristics, even though other variables/attributes can influence the decisions

• Inability to understand context-specific relationships between parameters (geography, customer segment, insurance sales process) that might not provide the complete picture

• Recalibration of the model requires frequent reevaluation to reflect changing behavior and ensure the model adapts to investigation feedback.
 

How do AI-Driven Automation and Data Analytics Help Detect Fraud?

AI-driven automation analyzes the data collected from various sources to identify patterns, anomalies, and trends that may indicate fraud. The automated fraud detection systems typically analyze the following data types to flag fraud:

• Transaction data: Tracking unusual patterns in spending, transfers, and withdrawals that may indicate fraud
• Document data: Scanning loan applications, tax forms, and other documents for signs of tampering, inconsistencies, or anomalies
• Customer data: Comparing new activity against a customer's historical transaction patterns and demographic information

Key benefits of fraud detection automation include:

• Speed: Automated systems can analyze transactions and documents at scale, flagging suspicious activity in real time
• Accuracy: AI and machine learning algorithms continuously learn, becoming more accurate at detecting patterns indicating fraud
• Consistency: Automated systems consistently apply the same logic and thresholds, eliminating human bias and variability
• Scale: Carriers can monitor a much larger volume of transactions and documents for potential fraud
• Revenue: Preventing, detecting, and flagging fraud early translates into lower premiums for the customer and a reduced loss ratio for the carrier, driving profitability

While automation cannot replace human judgment entirely, it can improve fraud detection by flagging suspicious activity for further review. This allows carriers to focus their resources elsewhere and operate more efficiently. With AI enabling more accurate fraud detection and fewer false positives, carriers can substantially reduce their financial losses while mitigating risks. Moreover, by automating repetitive processes like fraud detection, carriers can scale their operations without increasing headcount or costs.

Yet, the true beneficiaries of these advancements are the customers themselves. They need no longer bear the brunt of higher premiums to offset the losses from insurance fraud. And that’s a win-win for everyone, signaling a future where the insurance landscape is more equitable, efficient, and customer-centric.

Contact us if you’d like to discuss how you can automate with AI. Learn more by downloading the e-Book – Beyond the Bot: How AI Is Ushering in the Next Wave of Automation 

Murray Izenwasser, Senior Vice President, Digital Strategy

At OZ, Murray plays a pivotal role in understanding our clients’ businesses and then determining the best strategies and customer experiences to drive their business forward using real-world digital, marketing, and technology tools. Prior to OZ, Murray held senior positions at some of the world’s largest digital agencies, including Razorfish and Sapient, and co-founded and ran a successful digital engagement and technology agency for 7 years.

Sponsored by ITL Partner: OZ Digital Consulting