Planting the Seeds of AI

What do digitalization and artificial intelligence (AI) mean for insurers and their business models? Will the latest developments in digital technologies change the competitive landscape of the insurance industry? Insurers are “planting their seeds” by starting to use AI, exploring where it could add short-, medium- and long-term benefits.

The Geneva Association and Generali recently held a digital technologies conference in Milan that focused on generative AI and the future of insurance. In this article, we look ahead to 2035 and examine the potential impacts of AI and other digital technologies on insurance, drawing from our reflections on the topics that were presented at the conference.

10 Years Out: How Will AI and Digital Technologies Potentially Change Insurance?

Advancements in AI and other digital technologies potentially would give insurers more opportunities to engage with their customers, provide value-added services, orchestrate partnerships, tailor pricing to specific customer needs and administer their portfolios more efficiently. We observe three prevalent trends that we believe will change insurance:

1. The use of AI and generative AI will expand and potentially change the world, much like the revolutionary invention of the steam machine.¹ Insurers are currently defining use cases on how AI and generative AI can be applied in their day-to-day operations, focusing mostly on efficiency gains. In the future, we believe large language models might carry out knowledge engineering, and robots could assist us with daily tasks, from housekeeping to solving more complex, scientific problems.
2. If extreme weather events driven by climate change, such as wildfires and storms, continue to increase in frequency and severity, insuring natural catastrophe risks could become more challenging due to scale of loss. In 2023, economic losses due to natural catastrophes reached US$380 billion, of which US$262 billion (69%) was uninsured.² To help their customers deal with this increasing risk, insurers are establishing preventive measures such as early warning systems or installing monitoring devices in homes. These measures not only reduce losses overall but improve the customer experience and help protect customers’ assets and families. In addition to the natural catastrophe challenges, protection gaps are widely present in pensions, health care and cyberspace, all of which are large in size and potentially could result in greater societal challenges. We envision that AI and other digital technologies will empower health care and elder care services by reducing costs and possibly improving quality. AI also may enable automation and improve insurability of cyber risks. Therefore, AI and digital technologies could fundamentally address these protection gaps and their subsequent societal challenges.
3. How insurance products are supplied could change significantly. Focus could increasingly be placed on offering solutions that provide added value to customers rather than selling specific products. This implies that, for certain lines of business, insurance products may no longer be sold in a standalone manner but may be embedded in underlying service processes. Insurance may be one of many “added-value services” that customers buy.

Specifics: Possible Impacts of AI Applications

AI and generative AI will play a key role in improving customer engagement, productivity and operational efficiency. AI is already partially operationalized and used at scale in areas such as claims (focus on fraud detection and payouts) and underwriting (assessment of risk exposure and prediction of life insurance surrenders). The number of AI applications in insurance is expected to proliferate in the future.

Positive momentum is evident, but insurers are at the beginning of their AI journeys. We see three main considerations for the industry:

1. Vision and strategic direction. Given the potential disruptive nature of AI, it is important that insurers define a clear long-term perspective about the role they want to play and their “right to win”—that is, a clear strategic focus on what could drive the business value going forward. This approach then would drive investment allocation into technology, skills and operations. Vision and strategy act as the framework around which use cases could be developed, iterated and scaled to ultimately create value for insurance companies.
2. Data and technological infrastructure. With AI, it is generally known that the output can only be as good as the input. We foresee the need for insurers to build the right seeding ground for upcoming use cases to flourish. Therefore, it is critical to build a consistent data infrastructure with minimal complexity. Which data is needed could be defined in conjunction with the respective business functions. Insurance companies across the globe have been building highly complex technological infrastructures that run on their legacy systems. In this sense, insurers have not been as fast to modernize their legacies as the pace at which new technology is advancing. Technological infrastructures combine a wealth of data to develop targeted use cases at scale that lead to positive economic outcomes for customers and insurers.
3. Governance and operations. Knowledge of AI and generative AI does not need to be concentrated among a small number of people. We believe it is critical that capabilities are broadened across a wider knowledge base to reduce dependency. Rules and governance that establish a code of conduct—taking into consideration ethical and cultural considerations—could be challenging to create, especially for insurers that operate on a global scale. Depending on the regime, different standards for data protection and security may apply and, as such, insurers would then need to work closely with regulators and other relevant stakeholders to support and safeguard general rules (e.g., the EU Artificial Intelligence Act).³ The sophistication of AI models likely will continue to evolve rapidly, so we assert that rules and regulations need to be designed flexibly to mitigate the risk of lagging behind the pace of technological development.

Balancing Opportunities and Risks of AI and Digital Technologies

AI and generative AI are driving real change in the insurance industry, and it is fair to expect that development will continue at an unprecedented pace globally. These advancements will help insurance companies become truly customer-centric, using technology to drive efficiency and increase productivity. Large language models will get smarter over time, allowing for more sophisticated analyses and reasoning and further extending the range of potential use cases.

These new technologies also bring up new risks and worries, including deepfakes, job replacement and privacy concerns. So, insurers need to remain cognizant of their responsibility to their customers, employees and shareholders, as well as be mindful that they will need to adhere to any potential new regulations.

In the future, we believe successful insurance companies will reap the benefits of early investments in AI and digital technologies. They will set themselves up to flourish by building strong data foundations and robust internal governance; prioritizing use cases that have the highest added value; and ensuring use cases comply with ethical, cultural and potential regulatory standards.

Statements of fact and opinions expressed herein are those of the individual authors and are not necessarily those of the Society of Actuaries or the respective authors’ employers.

Copyright © 2024 by the Society of Actuaries, Chicago, Illinois.