The recent launch of ChatGPT has underscored artificial intelligence’s newfound scale, speed and mass accessibility, as well as its long-term potential to complete complex tasks currently performed by humans. Suddenly, AI is in the headlines everywhere, drawing the attention of commentators, business executives and even policymakers. 

For the insurance industry, AI and machine learning (ML) technologies are by no means new. Across the sector, organizations have begun launching AI and machine learning programs over the past several years and are using the technologies to drive efficiencies in core parts of their business, including underwriting, risk management and claims. More recently, these efforts have intensified, driven in no small part by concerns about an uncertain economy and companies looking to “right size” their workforces.  

According to a survey of IT leaders we conducted in 2023, the insurance industry has been accelerating the use of AI over the past five years. But recently the rate of adoption has recently seen a remarkable surge, with a 30% increase in new AI/ML projects from 2021 to 2022. 62% of insurers say implementation of AI/ML has resulted in an overall reduction of their overall headcount, while other companies have focused on retraining employees whose jobs have been affected by AI. Most remarkably, 81% of insurers cited AI as their leading strategic IT priority, outpacing the use of cybersecurity at 63% and cloud at 58%.  

Over half of insurers say they are using AI/ML for product lifecycle management, as well as to drive innovation and data analysis. Intelligent search, document processing and customer engagement are other fast-growing areas. A sizable majority (65%) said they are leveraging AI and ML to improve speed and efficiency, while half say it is helping them predict business performance, and 46% are turning to these technologies to better manage risk.  

Challenges and Pushback

AI adoption across the industry hasn’t been without its challenges. First and foremost, insurers continue to face internal resistance to implementing projects. Over half (56%) of insurance IT decision makers said they have encountered pushback or scrutiny regarding the use of AI/ML in their organization. This could be the result of differing perspectives between business leaders and IT departments.  

Building trust in the results of AI/ML projects is another common issue. When asked if they feel the data that AI generates is reliable, more respondents (42%) said they only “slightly trust” the data than those (38%) who “completely trust” it. Moreover, less than half say there is sufficient governance in place to safeguard against any misuse of the technologies. 

Finally, more than two-thirds of respondents (67%) identified a shortage of skilled talent as the greatest challenge to greater AI/ML adoption. Other roadblocks include a lack of new use cases, algorithm/model failure and a lack of infrastructure necessary to support AI/ML. Despite these hurdles, 90% of insurers say they have grown their AI and machine learning workforces over the past 12 months.  

See also: OCR Plus AI Opens New Vistas

Room to Grow 

Even though over half of insurers say they’ve already realized substantial benefits from AI/ML, the survey also makes clear that there is substantial room to grow. The list of benefits to date is impressive:

• 81% cite risk reduction and an increased understanding of customers
• 79% have seen increased sales
• 77% have used AI to create more personalized marketing
• 75% say AI has increased productivity
• 73% have seen new revenue streams and operational cost reductions
• 69% cite improved customer satisfaction
• 67% have benefited from faster time to profitability and reduced the cost of product development
• 65% say AI has made them more innovative 

These numbers are remarkable, given the technologies’ relative infancy, and would indicate that the insurance industry has just begun to scratch the surface of what AI can do. As companies assess their existing projects and become more comfortable using artificial intelligence across more parts of the organization, AI will become an increasingly critical strategic differentiator and springboard for business success.

KEY TAKEAWAYS:

--Unless AI systems are trained on accurate, unbiased data by unbiased trainers, they can make faulty decisions on underwriting and claims that create legal and ethical risks.

--If great care is taken to train and review AI systems thoroughly, they can provide a host of benefits in customer service, claims and underwriting.

----------

Artificial Intelligence (AI) is transforming the insurance industry by enabling insurers to process claims more efficiently and accurately while improving customer experience. However, the adoption of AI in claims management and underwriting, while useful, poses significant risks and challenges that insurers must address to ensure successful implementation and avoid costly mistakes. 

Big names in the tech world, such as Elon Musk and Steve Wozniak, recently penned an open letter urging labs to pause the training of AI systems for at least six months or until developers and leaders agree on better rules-of-engagement -- ensuring AI is developed to improve human life and is not weaponized or used for harm. Zurich and Lemonade are some examples of Insurers that are investigating and subsequently creating controversy around ways to leverage this technology.

Here, I will explore the risks and challenges associated with AI in claims management and underwriting, as well as strategies to mitigate them.

Risks of AI in Claims Management

AI systems are only as valuable as the data provided by their creator. Any information beyond what an AI system is fed is then inferred to create what experts refer to as hallucinations. Hallucinations can be very convincing, even though they aren’t based on good data. What constitutes “good” data from “bad” is a mix of factors such as the accuracy of the data, the perspectives, opinions and biases of the human inputting the data and who is regulating or training these individuals.

These are just a few of the risks associated with using AI. But one of the most significant is that AI can be uncontrollable. With more and more competition to create the most knowledgeable AI systems, creators are getting to the point where they can’t understand or predict their behavior or accurately explain how a decision was made. This can be problematic if humans rely on AI’s assessments to determine specific actions, such as denying or paying an insurance claim. 

Following are some other ways these risks apply to claims management:

Bias

From healthcare to recruiting, all businesses are subject to bias. The insurance industry is no exception. The U.S. Department of Commerce reports that one of the most significant risks of AI in claims management is the potential for biases, discrimination, and inaccuracy. If AI systems are trained on biased or inaccurate data, the decisions made by the AI system will be the same. This can result in unfair treatment of certain groups of individuals or incorrect claims processing, leading to reputational damage and legal liabilities.

This problem has been seen more recently regarding the use of FICO scores in the underwriting of insurance policies. Insurers are obligated to ensure that bias is removed from underwriting and claims decisions and that the information being used is accurate.

Legal/Ethical Risks

AI systems also require access to large amounts of data, including sensitive personal information, which can be vulnerable to cyber attacks and data breaches. This is a risk because it can result in significant reputational damage, legal liabilities and loss of customer trust.

There are legal and ethical risks with AI, as well. For example, an AI system denying a claim based on biased or inaccurate data could result in legal action and reputational damage. Additionally, the use of AI in claims raises ethical questions about the role of humans in decision-making and the responsibility of insurers to ensure fair treatment of their customers.

Overcoming the Challenges of AI in Claims Management

The first step in overcoming the challenges of AI in claims management and ensuring successful implementation is to review the data for accuracy and quality. It is crucial to confirm where the data came from, how it was vetted and how sensitive data is protected. This includes regular monitoring and auditing of AI systems to identify and correct errors and biases.

With the amount of content that is created online, it’s easy to pull data that stem from questionable sources. These quality checks can allow insurers to identify and correct inconsistencies in the data. Data can also change over time, so your data must be current and relevant. Routine checks will enable you to do this.

Next, you must understand how the AI system was trained and by whom. Industry experts are better-equipped to provide the knowledge AI systems are fed versus those with just tech experience.

Collaboration between humans and AI systems is another key to ensuring fair and accurate claims processing. Insurers should establish clear guidelines for when and how human intervention should occur in claims processing and confirm that humans are adequately trained to understand and collaborate with AI systems.

An independent review board inside an organization should also be implemented to test the technical efficacy of the AI system and confirm it’s working as expected. This board can also determine the moral and ethical implications and whether the system should exist in the first place.

By implementing these strategies, insurers can overcome the challenges of AI in claims management and leverage the benefits of AI to improve the efficiency and accuracy of claims processing, enhance customer experience and drive better decision-making. 

See also: 'AI' or Just 'I'? Most Adaptable Will Win!

Benefits of AI in Claims Management

While AI poses risks and challenges, it also presents significant benefits in customer service, claims and underwriting.  

Customer Service

One of the primary benefits of AI is that it can process large amounts of data quickly and accurately, reducing the time and resources required for customer service. Here is how AI can enhance the overall customer experience:

• Use chatbots or answer bots to answer customers' questions quickly, reducing wait time
• Submit policy changes and other simple endorsements on behalf of the customer
• Notify the agent or customer of any outstanding items, such as payroll audits
• Note any significant rate increases at renewal and automatically re-shop the policy
• Create and forward policy documents, such as ID cards, declarations pages or COIs

Improved customer service can help to protect margins on commissions as a result.

Claims Management

AI can process claims end-to-end and eliminate customer or agent frustration throughout the entire claims process by:

• Managing First Notice of Loss (FNOL) and First Report of Injury (FROI) correspondence
• Classifying, indexing, extracting and relaying claims data into agency/carrier systems
• Screening for potential fraud and validating eligibility
• Calculating and setting reserves and paying claims
• Identifying, servicing and following up on time-sensitive activity with adjusters, such as legal demands within demands packages
• Forwarding explanations of benefits (EOBs) and policy information
• Identifying accounts for recovery or subrogation based on the value at stake

Conclusion

With the challenges involved with AI and the potentially significant impact it can have on the world, there should be checks and balances to mitigate moral, ethical and legal concerns. As the AI open letter so succinctly states, rather than a race to the top, all future research by AI labs should focus on how to make the systems accurate, trustworthy and safe.

With the challenges involved with AI and the potentially significant impact it can have on the world, there should be checks and balances to mitigate moral, ethical, and legal concerns. By taking a break to establish these standards, we can be better prepared to use AI in the future and continue to revolutionize the insurance industry.

KEY TAKEAWAYS:

--In the midst of all the enthusiasm, it's right to imagine your full range of strategic innovation options in the context of your company’s particular markets, strategy and key business drivers.

--But don't fall in love with the one or two you think are the best. Instead, break them down into smaller pieces and run those opportunities through a rigorous screening and experimental process.

--Test with inexpensive, rapid prototypes (not pilots) and be extremely disciplined about which should be dropped, which merit continued exploration and which should move into the pilot phase.

----------

AI is having quite a moment. One high-profile AI app, ChatGPT, has become the fastest-growing consumer app in history. It garnered over a million users in its first five days and hit one 100 million active users in just two months. That’s faster than TikTok, which took nine months to reach the same mark, and Instagram, which took 2.5 years.

ChatGPT is one of many recently launched applications using machine learning techniques to identify patterns from very large data sets to generate content. As its name implies, ChatGPT uses a text-message-like user interface to “chat” with the user. It can generate and tune intelligible (though not always correct) responses to seemingly any prompt, and it can do so in the form of essays, jokes, poetry, recipes and even computer code in response to user direction. Other apps can generate content in video, art and audio form.

The race is on to leverage the underlying AI platforms, tools and techniques in a wide range of consumer and industrial applications. Microsoft has invested $10 billion in OpenAI, ChapGPT’s developer, in hopes of using ChatGPT to help free it from its laggard status in the very lucrative business of internet search. Other tech giants, like China’s Baidu, also see this as an opportunity, and are investing heavily in similar capabilities. Not to be left behind, Google has made clear its intention to respond in force with its own suite of deep AI capabilities. A range of other industrial applications are being explored and debated, including in media, healthcare, education and finance.

Is it time for insurers to “go big” on this technology, too? The short answer is no. But, insurers shouldn’t just take a wait-and-see attitude, either. Instead, as I counseled in a previous article, insurers should follow Peter Drucker’s advice on innovation:

“Don’t subscribe to romantic theories of innovation that depend on ‘flashes of genius.’ Innovation begins with the analysis of opportunities. The search has to be organized and must be done on a regular, systematic basis.”

How might an organized, systematic approach to explore and leverage ChatGPT and other AI capabilities look? Here’s the outline of a three-stage approach that builds on many of the ideas that I’ve offered in previous IIS research articles.

1. Think Big

Thinking big means considering a full range of possibilities. Rather than accepting the widespread exuberance (or the emerging backlash) surrounding AI, make sure to understand your strategic innovation options in the context of your company’s particular markets, strategy and key business drivers. Consider both incremental and disruptive possibilities, including how developments might change customer preferences, enable new markets and upend business dynamics. Be willing to start from a clean sheet of paper. Don’t be too proud to explore downside and even doomsday scenarios, including how new developments might alienate your customers, dent your existing advantages and challenge your existing products.

What might be opportunities, for example, to automate repetitive tasks, help underwriters make better-informed decisions, improve the accuracy and speed of claims processing and enhance customer experiences? What are applicable lessons to take from Lemonade, which uses AI to automate the underwriting process and to handle claims processing. It also uses chatbots to communicate with customers, providing quick and convenient service. Allstate is using AI to prioritize claims for expedited handling. Metromile has long used AI algorithms to analyze driving data to help determine a customer's monthly insurance rate, offering more affordable insurance options to low-mileage drivers. Metromile’s AI capabilities were a key factor in its acquisition by Lemonade. Fraud detection is another area where AI is having a significant impact. AI algorithms can analyze large amounts of data, including claims data, policyholder data and social media, to identify unusual patterns and behavior that may indicate fraud.

See also: When You Have Too Many Good Innovation Ideas

AI might also be used to better address very large “change the world” kind of growth opportunities, such as in cyber-security and climate change. For example, AI algorithms can analyze large amounts of data to identify potential cyber risks, such as vulnerabilities in software and hardware, or unusual network traffic patterns. Insurers can use this information to help their clients better understand their cyber risk profiles and to develop more effective cybersecurity procedures, policies and products. AI algorithms can analyze vast amounts of data, such as weather patterns, satellite imagery and historical claims data, to identify patterns and trends that may indicate an increased risk of natural disasters. This information can be used to improve climate risk assessments and to develop more accurate pricing for insurance products.

It is important to explore not only the upsides but also the downsides and fault lines of this technology. AI tools are clearly powerful, but they also can inadvertently perpetuate bias due to flawed data or poorly designed algorithms. They can stimulate concerns about privacy, misuse of data and nefarious use by bad actors. And AI can provoke regulatory and customer backlash about automated rather than human-based decision making.

2. Start Small

Don’t fall in love with the ideas that come out of the “think big” stage. Resist the tendency of jumping on the bandwagon and going “all in” on one or two seemingly big ideas. Instead, break them down into smaller pieces and run those opportunities through a rigorous screening and experimental process, such as the one I proposed in my previous IIS article.

In the terminology of innovation portfolio management, the key to starting small is to pare all those potentially good ideas into a small basket of “option-creating investments” (OCI). OCIs are the earliest stage innovation initiatives. Companies typically pursue them to explore business opportunities that might yield large returns in the future. They resemble financial options—from which the name is derived—in that they are relatively low-cost, exhibit high risk and offer tremendously high potential returns but do not commit the organization to full-scale implementation.

It is important not to pursue too many OCIs. That’s because even though such efforts should be relatively low in cost, they will demand a lot of senior management attention. No truly disruptive innovation can succeed without sufficient  senior executive attention to protect the efforts from corporate antibodies as necessary, to terminate them when they don’t pan out and to allocate the considerable resources for scaling if appropriate.

3. Learn Fast

After thinking big and starting small, aim to “learn fast” by taking a scientific approach to innovation.

Many companies think that experimentation consists of setting up pilot projects to validate concepts before they’re rolled out. But this is a dangerous approach. Pilots aren’t the answer, at least not on their own. Once something gets anointed as a “pilot,” it’s no longer an option—it’s the destination. Pilots are typically touted as “the answer,” rather than the testing of a possible answer. Assumptions are inevitably made that the pilot will work and will be quickly rolled out at scale. Senior executives invest political capital, often promising eventual success to the board, investors and the press. Team members are judged based on the pilot’s success or failure. All this means that pilots are typically overbuilt. There are no graceful ways to kill them. And, even course corrections are hard to make because any significant changes are seen as a sign of defeat.

Instead of jumping into production-quality pilots, you should achieve your early learning with prototypes that are all chewing gum and baling wire. The difference between a prototype and a pilot is there’s no possibility or expectation that a prototype will turn into a final version of the product or service. Prototypes are just tests to explore key questions, such as whether the technology will work, whether the product concept will meet customer needs or whether customers will prefer it over the competitive alternatives. Prototypes should aim to provide real insight that informs further learning and development. Each stage should minimize costs and maximize flexibility.

See also: AI Offers Big Step Up in Underwriting

In most cases, you’ll wind up doing several rounds of prototypes as you search for vulnerabilities and reduce uncertainties about technology and product viability. You’re looking for evidence of both potential success and failure, but all you want to learn is whether you know enough to kill the project (let the option expire), to continue exploring (renew the option) or to move forward into the next phase of production (exercise the option).

Discipline is critical — if the results after a round of prototyping are still unclear, be very specific about what issues need to be explored. Then lather, rinse and repeat—until you’ve resolved all the key questions.

Once you’re confident that you’re ready to exercise your option and proceed in a specific direction, don’t be tempted to go straight to full production and rollout. That can get you into trouble. Instead, this is the time to do a pilot.

* * *

Thinking big, starting small and learning fast won’t guarantee success. AI encompasses a complex set of technologies and capabilities at the advanced edges of computing. Figuring out your company’s strategy in the context of the immense opportunities and challenges that AI amplifies will be challenging, as it will be for your competitors. In the end, business is a contact sport. Some companies win; some companies lose. Taking a disciplined, long-game approach to innovation will put you in best position to win.

Independent insurance agents constantly face the challenge of finding new prospects to grow their agency's business. While casting a wide net may be appropriate in some circumstances, strategic targeting of commercial sectors with consistent hit ratios can lead to more efficient prospecting and, ultimately, higher conversion rates.

Having been a commercial lines producer for 25 years, I realize identifying industries with consistent hit ratios and prospecting to them can be challenging, but it is crucial for building a robust book of business. During my time as an agent, I developed specializations in retail auto parts and private colleges and universities in Kentucky. This enabled me to concentrate my marketing efforts and outreach toward these sectors. Through this focused approach, I gained a deep understanding of these industries, establishing myself as an expert in the field.

In my role at AHA Insurance Network, I conduct a two-day course designed for agents who want to specialize in writing small commercial business. A part of this program is dedicated to educating them on prospecting for industries with consistent hit ratios. I have highlighted some of the same course topics here and offer insights on identifying potential clients and industries with the most reliable hit ratios, as well as common pitfalls to avoid.

Industries With High Hit Ratios

In a highly competitive landscape, some agents may think targeting niche industries, such as golf courses, will be the remedy. However, this may not provide a long-term solution, as the market can be limited by your geography. Regardless of your location, there is an abundance of contractors such as electricians, plumbers, HVAC technicians and excavators who are vital to any construction or repair project. Therefore, I recommend pursuing contractors, especially artisan contractors who are physically present at the job site. Unlike general contractors, who outsource the work, artisan contractors are favored by insurance carriers because they are more regulated.

Developing a Hit List

Developing a hit list for a specific market typically involves examining carriers and their performance in that market. You'll want to identify the specific class of business you're interested in targeting, such as plumbing contractors, and then gather data on the carriers that write those types of policies. This data might include information on each carrier's market share, loss ratio and hit ratio in that market.

Once you have this information, look for carriers with a hit ratio of 30% or higher in that market. These are the carriers most likely to be successful in writing policies for that class of business. Once you compile a list of these companies, along with any other relevant information, such as their contact information, underwriting guidelines and any unique features of their policies, you’re ready to begin your outreach.

Developing a hit list requires access to comprehensive data on carriers and their performance in specific markets. This information can be obtained from a variety of sources, including industry reports, market research firms and insurance industry associations. Some insurance agencies also have access to proprietary data that can help inform their hit list development.

AHA Insurance Network members can request a lead list for specific classes in certain counties or ZIP codes. Members can review the lists to identify carriers they already use for direct appointments as well as the 10 most promising classes. Alternatively, agents can explore online directories or professional associations to identify potential clients within specific industries.

Building Relationships

If you want to excel in a specific industry, it's crucial to network within that industry and join the relevant associations. This strategy worked well for me when I managed an agency serving the automotive retail industry, particularly the independently owned auto parts stores affiliated with Carquest. I attended conferences and meetings to establish myself as the go-to insurance expert for these accounts by connecting with the district manager of this company and getting involved with the Kentucky Indiana Automotive Wholesalers Association.

This approach requires regular, active engagement in associations at the local, regional and national levels. Find a carrier that will collaborate with you and become well-versed in the industry’s language and nuances. This approach enables agents to position themselves as experts or at least come close to being one. 

See also: The Future of the Independent Agent

Staying Current on Industry Trends

It is crucial to regularly read trade publications and use online resources to stay current on industry changes, such as the recent change in Indiana regarding underground gas storage tanks. In Indiana, gas station owners are now required to have pollution insurance, whereas it is not required in Kentucky. This knowledge is essential when assessing an account to make sure it has the necessary coverage. An agent can remain ahead of the game and competitive by staying informed through trade publications, carrier newsletters and meetings with underwriters and marketing reps.

Mistakes to Avoid

It’s great to have written almost all the golf courses within a geographic area. But then what? Independent agents need to expand their reach beyond a certain geography or sector to continue growing the business. This might involve expanding the geographic area or including another class of business. It's important to ensure there are enough prospects to sustain an agency over the long term and not just in the immediate future.

Final Thoughts

For generalists, which is probably the case for most independent insurance agents, the key is to never stop learning. Find a mentor and develop relationships with underwriters on both the E&S and standard carrier side. Don't be afraid to ask questions of a mentor, marketing reps and carrier underwriters. They are amazing resources for information and guidance. Everything is constantly changing in commercial lines insurance, so asking questions and learning as much as possible is essential.

Independent insurance agents face many challenges when it comes to developing a robust book of business. With the right strategies and a focus on industries with consistent hit ratios, success in the small commercial space is attainable. Agents can position themselves as experts and grow their businesses in a sustainable way by developing specializations, building hit lists, networking within industries and staying current on trends.

Remember, success is not just about casting a wide net but about targeting the right prospects and cultivating meaningful relationships.

Amid the buzz surrounding ChatGPT, it’s helpful to consider another application of AI that seems to have hit a wall: self-driving vehicles. 

Google’s Waymo unveiled the first driverless car in late 2015, and so many bullish predictions followed. For example, Lyft President John Zimmer predicted half the company’s rides would happen in fully autonomous cars by 2020. 

What’s the reality? A recent expert panel discussion in the Wall Street Journal pegged the commercial availability of Level 5 self-driving vehicles (i.e., zero human input required, zero restrictions) at 2035. 

Progress has been made. Today, over 60% of cars sold in the U.S. have Level 2 capabilities, equipped with dynamic cruise and lane assist (Tesla’s Autopilot is Level 2).

A question worth asking is: What is the Level 2 reality of AI in insurance today?  

The first use case is the automation of rote work that people--especially young people--don’t want to do, at seemingly any price. American Family, for example, raised their minimum wage to $23 an hour for mid- and back-office positions, in moderately priced Wisconsin. And they’re automating in response. 

The second use case is telematics and AI on the product side. Progressive’s pioneering Snapshot usage-based auto product is the leading example here. Another is Hartford Steam Boiler’s use of sensor data and AI to help predict machine breakdown and prevent the costly freezing and bursting of pipes.     

A third use case emerging is operational intelligence.   

What’s really going on in that sea of cubicles? What’s really working in work from home? A landmark study from IDC asserts inefficiency costs the average business 20% to 30% of revenue annually. Unclear boundaries, poor system integration, complacency and redundancy are the culprits, and they exist because leaders lack insights into them. If you can’t measure it, you can’t manage it.  

With platforms such as ours at Skan, combining computer vision and AI, measurable insights have arrived. 

Viewed one way, digitally observing humans and the tools they use as they’re using them is about Big Brother surveillance. Viewed another way, it’s about player stats toward continuous player and team improvement. As analytics have affected every major sport, operational intelligence is changing insurance by creating new ways to evaluate, build and manage human teams.    

See also: The Key to 'Augmented Intelligence'

Operational intelligence also enables new modes of work. A younger (or any) person may not be interested in processing claims for $23 an hour, but they seem interested in a gamified version of claims processing where they get paid for results, outputs, competing against their peers in the next cubicle or 10 states or an ocean away. Maybe the solution to the rising hourly wage problem is doing away with the hourly wage? 

New ways to manage, new ways to work. As Progressive uses AI to transform driver intelligence into new premium and profit, insurers are using AI to transform operational and worker intelligence into new margin.  

Come 2035, we may indeed have Level 5 self-driving cars, as well as Level 5 claims processers and sales agents operating at new levels of efficiency and effectiveness. Until then, let’s use the tools at hand to empower humans to find new ways to manage and work in service to other humans.

Let’s focus on replacing time-based cultures with cultures based on performance underpinned by intelligence. Let’s create modes of work that are more engaging and ultimately rewarding for the employee and the enterprise.

It’s challenging for insurers to differentiate themselves through products and services, but data and analytics can enable them to break free from commoditization with better and faster decisions. This potential is reflected in increasing levels of investment in data and analytics across the industry; these investments now exceed 0.7% of direct written premium (DWP), on average. Recent Aite-Novarica Group studies indicate that over half of all insurers are replacing or conducting major enhancements to their data environments and associated capabilities.

According to an Aite-Novarica Group survey of insurer CIOs, the average property/casualty insurer is spending 4.4% of DWP on IT. Of that, 17% is spent on data, resulting in average spending of 0.68% of DWP. (For more information, see the Aite-Novarica Group report Property/Casualty Insurer IT Budgets and Projects 2023.) The term “data” typically includes data warehouses, business intelligence (BI), predictive analytics, third-party data and practices such as master data management.

Where Insurers Are Investing

Insurer business units are prioritizing further expansion of BI and analytics over all other capabilities in 2023, reflecting the fact that carrier capabilities tend to lag in this area. Many insurers face data access challenges, resulting in significant effort expended for sourcing and merging data to produce reports. Data quality issues also affect insurers’ ability to receive the full benefit from analytics efforts. 

When asked to self-assess on various system capabilities, insurer CIOs indicated that the biggest gaps are in data and digital (which has a significant reliance on data). For larger property/casualty insurers, customer relationship management (CRM) is emerging as an area of concern; more than half of insurers rated their capabilities for CRM as “poor” or worse, up from over 40% last year. CRM systems are highly reliant on access to quality data. For midsize property/casualty insurers, customer portals and predictive analytics remain pain points.

To achieve their data objectives, insurers will need to improve their data maturity. Creating and maintaining a trusted source of data and analytics within an insurer requires competency and coordination across multiple disciplines. Insurers with mature data capabilities have access to unique insights and apply them in all aspects of their business to further competitive advantage. 

The True Meaning of “Data Maturity”

Aite-Novarica Group has developed the Insurance Data and Analytics Maturity Model (iDAMM) to facilitate an insurer’s journey to high levels of data maturity. The iDAMM provides a framework for insurers to evaluate their data-related capabilities and establish a plan to reach data maturity by assessing their data organization and capabilities across seven dimensions and 21 subdimensions, including leadership and organization, data governance and architecture and technology management.

The model uses three stages of maturity: Traditional, Evolving and Transforming. Insurers are likely to have different levels of maturity across different model elements. Moving into a more mature stage is a function of organizational and technological capability, not duration. Like transformations in other parts of the insurer, data and analytics innovation requires enabling technology, organizational change and executive sponsorship.

Sustaining data mastery requires an insurer to establish and maintain a data culture. Data cultures are those in which data is fully democratized, data literacy is high and tactical and strategic decision making are largely data-driven (but still informed by intuition). Further, all users value data highly and act as corporate data stewards in maintaining high data quality and protection.

See also: Achieving a 'Logical Data Fabric'

Building a Data Culture

Building a data culture is challenging. Insurers that have achieved data mastery will have a culture grounded in data and analytics that can survive leadership changes and shifts in business focus. This means that everyone cares about data quality, as everyone understands innately the value of information and insights. Strategic and tactical decisions are heavily influenced by signals in the data, though intuition developed through industry experience also plays a role.

Insurance industry data masters derive significantly more value from their data than other insurers, putting them at a major competitive advantage. Insurers that wish to achieve and sustain data mastery should assess where they are in their data and analytics journey, define a target state and plan investments and initiatives to bridge the gaps between them. The iDAMM is a useful tool to assess an organization’s current level of maturity and align that with where the organization seeks to grow.

For more information on achieving data maturity, see Aite-Novarica Group’s new report Establishing and Sustaining Data Mastery: Introducing the Insurance Data & Analytics Maturity Model (iDAMM).

KEY TAKEAWAYS:

--A "human in the loop" model lets insurers fight rising costs through rapid adoption of advancements in AI.

--The trifecta of AI, automation and analytics allows for breakthroughs in requisitions, claims, personalization, predictive analytics and embedded insurance.

----------

The insurance industry dates back to ancient Babylon and China, where merchants would pool their resources to protect themselves against losses from piracy and theft. The concept landed on the U.S. shores in 1752, when Benjamin Franklin co-founded the first insurance company in Philadelphia. Since then, the industry has grown to become a massive global enterprise. According to Insurance (Providers, Brokers and Re-Insurers) Market Global Report 2022, the global insurance market is estimated to grow from $5.38 trillion in 2021 to $8.39 trillion in 2026.

Key Challenges – Inflation and Natural Disasters

The industry needs sharp efficiency improvements to combat inflation, which skyrocketed from 1.23% in 2020 in the U.S. to 8% in 2022 during the COVID-19 pandemic, consequently increasing the costs of paying claims. Furthermore, due to climate change, population growth and migration to vulnerable areas such as coasts and river floodplains, there were 60 $1 billion natural disasters in the U.S. in the last three years, which is already nearly half of the 128 in the entire previous decade (2010-2019), and those numbers are inflation-adjusted to 2022 dollars.

Technology has the potential to offset these rising costs. The insurance industry needs to adopt a "human in the loop" model for rapid adoption of technological advancements in artificial intelligence.

There are numerous debates on the advancement in AI, thanks to ChatGPT and several similar generative AI tools. While there are concerns about the potential risks and unintended consequences that could arise, generative AI models could revolutionize the way humans interact with machines. These technologies have the potential to reinvent the insurance industry -- increase requisition and claims efficiency by up to 80%, reduce premiums through more detailed customer information and insights and even improve the economy.

While AI, intelligent automation and analytics can revamp the insurance industry, insurers need to ensure that the data they collect is protected and used ethically. 

AAA Trifecta

AI can be leveraged to analyze large datasets and identify patterns and trends that would be difficult or impossible for humans to detect. Intelligent automation can streamline processes, such as claims, and can handle routine tasks, such as sending out policy renewal notices. Analytics can identify patterns of fraud and detect suspicious claims.

Listed below are five ways in which the trifecta of AI, automation and analytics can transform insurance:

1. Breakthrough in requisitions

The insurance industry still struggles with ingesting, aggregating, contextualizing and processing incoming applications. This is partially due to many sources of unstructured application data such as paper documents, emails and voicemails that must be processed and incorporated into applications to properly evaluate risk and provide the best quote based on the most complete possible applicant profile. AI can automate most of the data acquisition, contextualization and decision-making with up to 90% accuracy, resulting in business benefits of up to 80% greater efficiency. AI allows for processing five times the volume of applications, resulting in more revenue.

2. Advancements in claims

Automated processes and AI will largely replace the adjuster who travels to view and photograph damage and file a report, followed by days or weeks of review by an in-house expert. The new AI and image-processing technologies based on customer-generated images and geospatial data can reduce claims processing from days or weeks to hours or even minutes. 

3. Predictive analytics

Once policies are in place, AI and predictive analytics technologies can further refine quotes, risk assessment and customer service by predicting which customers, regions and ZIP codes will likely incur losses and when and what types of losses will occur.

4. Personalization by leveraging AI

The more an insurer knows about a customer and their requirements, the better the insurer can service the customer and provide the best possible coverage and rate. AI and predictive analytics based on current and historical data combined with marketing automation can help insurers tailor their offerings, resulting in better customer service, customer intimacy and revenue increases through new, appropriate offers and a higher percentage of renewals.

See also: Key Challenges on AI, Machine Learning

5. Innovation in embedded insurance

It’s common to embed offers for extended warranties with product sales or travel insurance with travel purchases, but there are many more possibilities for embedded insurance using AI, intelligent automation, analytics and geospatial data, which will help improve customer intimacy and revenue opportunities while not being overly invasive.

According to a recent report by PwC, 54% of insurance CEOs say implementation of AI solutions has increased productivity in their businesses and believe AI will transform their industry in the next five years.

In today's fast-paced world, businesses need to make better and more informed decisions to stay competitive and grow. This is where AI can be incredibly valuable with a "human in loop" model that can provide timely and efficient insights in real time and support insurance business leaders to make better decisions.

All the frothy talk about how ChatGPT changes everything and about how we all need to be part of a land grab in the metaverse suggests that it's worth taking a look back at how the changes-everything/land-grab adherents fared during the early days of the internet in the late 1990s. 

It wasn't pretty.

While there were a few monster successes, notably Amazon, there were far more face plants like those by Webvan and Pets.com. The companies that reliably prospered, such as Sun Microsystems and Cisco, were what those of us in the financial media referred to at the time as "arms merchants" -- they supplied the combatants in the internet wars rather than forming "armies" themselves.

I'd suggest that being a metaphorical "arms supplier" is the best way for companies, including insurers, to play on ChatGPT and the metaverse.

In many ways, I'm repeating myself here, but, as a veteran reporter told me when I was a young pup at the Wall Street Journal, "If you have a good story, you ought to write it every once in a while."

Let me provide a bit of context, then I'll get into what I think the "arms merchant" concept means for ChatGPT and the metaverse. I'll start in August 2017, when I wrote about the early internet days:

"There was a class of 'arms merchants,' including Sun Microsystems and Cisco, that made gobs of money by outfitting the pioneers—in fact, more than almost all the pioneers.

"The analogy was: If you try to name a miner who won big in the Gold Rush of 1849, good luck. But you know many of the outfitters, including Levi Strauss, who supplied blue jeans to miners, and Leland Stanford, who made his fortune mostly on the railroad that connected the miners to the rest of the country, before founding his eponymous university."

Turning to the lessons for the insurtech movement, then in its infancy, I added:

"Lemonade, Trov, Slice and some other truly new business models stretch our thinking and throw shade on those merely looking for 'faster, better, cheaper'.... The Lemonades of the world will be the most important and will transform insurance. In time. If they work. (Some will, but, if history is any guide, many others will fall by the wayside.) 

"In the meantime, there is an awful lot to be gained through incremental improvement. Insurance is such a paper-heavy, process-based, inefficient industry that the potential efficiencies from digital improvement exceed those in perhaps any other industry."

Many of the big ideas that flopped in the early 2000s weren't bad ideas. They were just ahead of their time. Customer behavior hadn't yet adjusted to the new possibilities, and, in particular, the technology infrastructure couldn't yet provide the sorts of robust, vivid, fast interactions and delivery capabilities that e-business models required. While Pets.com filed for bankruptcy in 1999, for instance, Chewy carries a market valuation north of $14 billion, operating the same sort of e-commerce pet supply business. 

Venture capitalist Marc Andreessen was once asked how he determined which nascent technologies would work and replied that they all will work -- the multibillion-dollar question was WHEN they would work. That uncertainty is why so many big ideas fail. In Silicon Valley terms, the founders and backers confuse a clear view with a short distance. That uncertainty is also why it's much safer to be an arms supplier rather than lead the charge on something like ChatGPT.

It's not that I worry about insurers running off and trying to establish whole new business models based on ChatGPT or the metaverse. But I do see suggestions about ChatGPT, in particular, that strike me as being much too ambitious for now.

A newsletter from a venture capital firm -- Andreessen's, in fact -- suggested in the past week that ChatGPT could answer questions for customer service at financial services firms because it could pull together information from all the firm's internal documents. Others have suggested ChatGPT could soon be used to make underwriting decisions automatically or handle claims with almost no human interaction. 

But if you look into "large language models" like ChatGPT and see, among other problems, their tendency to make things up -- to "hallucinate," as the scientists euphemistically put it -- you see that they won't be ready for prime time any time soon. Yes, there's a clear view to utility in all sorts of applications, but not a short distance.

It's better to take the view that ChatGPT and its kin can supply information for now, but not to yet fight the war. That perspective will not only help insurers with their own implementations but should inform their interactions with clients, who need to understand the risks associated with their own buildouts and to see what sorts of coverage they should purchase.

As I wrote back in February:

"The issue with large language models like those used for generative AIs like ChatGPT and Bard is that they don't know much about the real world. They've just been fed unimaginable amounts of text and learned to imitate it. You give one a prompt, and it figures out what word is most likely to go next and then next after that and after that... and on and on and on. The results are scarily impressive but have a tenuous relationship with reality, which is why Bard claimed that the James Webb telescope discovered exoplanets, why ChatGPT has claimed that the most-cited medical journal article of all time is a piece that doesn't actually exist, why ChatGPT told a friend that he was married to a number of women he'd never met, had children he'd never had and wrote books that didn't exist....

"It's important to see the results from these generative AIs as what they are: a very rough draft. Now, as someone who has spent decades doing his thinking with his fingers on a keyboard, I can tell you that even a very rough draft can be extremely valuable," but it still requires human insight to become the finished product.  

What about the metaverse? I don't think you need to worry about that any time soon. As I wrote in late 2021 (to quote myself one final time): 

"The vision of a metaverse laid out by Mark Zuckerberg last week is bonkers. Nutso on steroids. It won't be realized in my lifetime, yours or his, even if some of the wildest claims about longevity come true and we all live to be 150.

"The vision is essentially a fever dream for gamers who'd love to immerse themselves in their online worlds and not have to worry about the messy details of physical existence."

Some day, I'll tell you what I really think. :) Yikes. I stand by what I said about how insurers should avoid involvement with the metaverse, unless a client drags you there, but I must have been in some kind of cranky mood when I wrote that.

Cheers,

Paul

Two developments, in particular, in 2023 will inform the approaches that regulators will take in the years ahead on how (and how much) to regulate artificial intelligence’s use in the business of insurance

First, insurance regulators via the National Association of Insurance Commissioners (NAIC) are drafting a model bulletin on artificial intelligence use with the aim of guiding companies in establishing governance systems and regulatory expectations for such systems. As explained during the 2023 spring national meeting in Louisville, the NAIC’s Committee on Innovation, Cybersecurity and Technology is taking the lead on producing a commissioner-driven deliverable that likely will be exposed for public comment this year. Much like the NAIC’s AI Principles adopted in 2021, the bulletin likely will provide high-level, principles-driven guidance that will serve as a good guide for companies seeking to understand, at a minimum, what kinds of questions and information regulators will ask when seeking more information about artificial intelligence and machine learning products.

Second, Colorado continues to move forward with its rulemaking under the Colorado Privacy Act, with the first round of draft rules for life insurers exposed for public comment in February. Although focused on life insurers, the Colorado Department of Insurance has stated in public meetings that property and casualty insurers should expect the version of the rule applicable to them will be similar. Colorado’s rules are more prescriptive than anything coming out of the NAIC thus far in detailing the information insurers will need to have available for using AI as well as how to report such information to the department.

The differences in these approaches is a preview for regulatory differences insurers will face in the near future across jurisdictions. Some will adopt the NAIC model bulletin, while others will modify it. Still others may follow Colorado’s lead in seeking legislation or adopting rules specific to AI usage. At a minimum, carriers using AI as part of their insurance offerings in multiple jurisdictions, irrespective of line, likely will be faced with a somewhat disjointed regulatory regime in the near term, even as regulators work to find consensus wherever possible.

So what should savvy insurers do now? At a minimum, any insurer that is using or considering the use of AI should be giving thought to implementing a well-documented governance system for its AI and machine learning tools. In other words — how does the enterprise show its work? Whether a jurisdiction elects a more front-loaded approach to regulation (like Colorado, with significant reporting requirements) or back-loaded (guidance followed up with market conduct reviews, if necessary), much of the regulatory risk surrounding AI boils down to two questions: Do guardrails exist around a company’s AI and machine learning tools, and can the company defend those guardrails as appropriate and adequate through testing and documentation?

See also: Regulatory Interest in Big Data

In addition, companies with robust governance integrated into their AI and machine learning portfolios are in a much stronger position to shape regulatory requirements as they come into sharper focus. As regulators and policymakers focus more on how and to what extent companies should be prepared to explain AI guardrails, such carriers will not only be more prepared when regulation comes, they will also be in a much stronger position to speak up and be taken seriously when regulatory proposals become unnecessarily burdensome. As counterintuitive as it may seem to some, patient engagement with insurance regulators will make for a more navigable long-term regulatory framework.

Near-term regulatory uncertainty notwithstanding, establishing robust governance and testing regimes for AI and machine learning are smart investments for insurers in anticipating whatever regulatory requirements emerge. It will be much easier to tweak such systems as needed, once established, than scramble to implement wholesale systems in response to new regulatory requirements. Keep in mind: Governance is distinct from AI and machine learning tools themselves. If AI and machine learning are the economic engines of the future for insurance carriers, effective governance is the oil that will keep the engine running smoothly — and compliantly.

As first published in Digital Insurance.

Q1 '23 has shown a clear trend in the insurtech deals: It is not any more about raising money, it is about M&A. The Financial Technology Partners / FT Partners' monthly report has certified it:

• The number of investments in Q1 '23 has been 47, with less than $800 million invested (the most recent lower quarter was Q2 '18, with less than $600 million).

• The number of M&A deals has been 34, totaling $4.7 billion (already higher than the entire 2022).

Even looking at the listed insurtech carriers, we had some exciting takeaways (while waiting for the Q1 earnings calls). These takeaways are highlighted by the recent journeys of Lemonade and Oscar.

I've already compared the two players in this newsletter last July. Comments on their financials can be pretty similar eight months after: Both companies are still burning a ton of cash (Lemonade had $289 million in net losses in 2022, Oscar $610 million). However, their valuations diverged in these last few months: Lemonade has even lost "unicorn status," while Oscar has just gained it again.

This dynamic seems to be due to "insurance ignorance":

• Everybody remembers the original Lemonade celebration of their ignorance of the sector. ("Instead of reaching out to insurance experts to gain a better understanding of the industry, the co-founders resisted what Schreiber describes as a 'temptation' to access insider knowledge and opted for an alternate route: 'We milked our ignorance for all its worth.'")
  • 
• More recently, they offered the exotic rationalization that churned policyholders haven't left; they've become alumni!

The jump of Oscar's market cap instead has been due to the appointment of the new CEO at the end of March: Mark Bertolini, former chairman and CEO of Aetna. It is not cool anymore to be ignorant, better to have some knowledge and expertise. Robust insurance foundations are a necessary element for successful insurtech initiatives.

See also: Is My Organization Actually Innovative?

Talking about robust insurance knowledge -- and its virtuous combination with innovation -- I want to share some insights from a thought-provoking presentation John Ingersoll, head of strategy at CSAA, gave recently about their ambitious innovation journey.

"Find Time for Innovation" Josh Ingersoll, April '23

This 100-year-old insurance group -- focused on personal auto and homeowners insurance -- created an internal insurtech startup in 2019: Mobilitas Insurance. This startup is a commercial insurance carrier focused on new mobility. It has already increased the group's top line significantly -- representing 11% of the total premiums -- and aims to be as relevant as the traditional business by the next decade.

"Find Time for Innovation" Josh Ingersoll, April '23

The CSAA innovation journey is a great example that highlights how insurance insurers can be innovative. As I wrote the first time in 2016: "All the players in the insurance arena will be insurtech! Meaning, organizations where technology will prevail as the key enabler for the achievement of strategic goals."

Over the past few years, in any business line and in any international market, you can find insurers that have successfully applied and scaled insurtech solutions in their business. These success stories should inspire more insurance carriers to design and execute their innovation journey. Insurtech can make the insurance sector stronger and, therefore, more capable of achieving its strategic goal: to protect the way people live and enterprises operate!