With artificial intelligence (AI) continuing to make an impact across the industry, by transforming the insurance underwriting process, and the value chain by boosting efficiency and improving accuracy, the technology is also creating potential for insurers to further improve productivity and minimize, if not actually eliminate, human touch points, according to credit ratings agency AM Best.
The agency explains that even as hiring levels decline and layoffs appear to be rising across the insurance industry, it is too soon to point towards AI as being the leading cause of the job losses, at least at this nascent stage.
According to Best, the recent layoffs across the industry are more likely fall into the cyclical, rather than the structural, category.
Sridhar Manyem, senior director, industry research and analytics, AM Best, commented: “Personal lines writers, including auto and homeowners’ insurers, are the most affected by the current layoffs. Carriers’ loss ratios and underwriting margins are being pressured by loss cost inflation, reinsurance capacity and pricing, and rising climate risk.”
However, as AI capabilities continue to broaden and insurers continue to become more comfortable with the technology, especially within using it for business processes, either through their own efforts or by outsourcing, automation will eventually impact industry employment levels and a much wider set of occupations, warns Best.
Adding: “The potential for lower premiums and greater flexibility in coverage limits may help reduce costs and generate savings for both insurers and customers.”
Interestingly, a new study that was recently launched by FintechOS showed that 73% of UK insurance executives anticipate that Generative AI will eventually take their jobs. Approximately 25% of those surveyed described themselves as feeling “frightened”, while 23% admitted to feeling “curious” about the technology.
Moreover, other parts of the insurance value chain, such as customer service, policy generation, claims handling, and areas that use document and image processing, could also be enhanced by the advent of large language models too.
Edin Imsirovic, director, AM Best, said: “As AI capabilities broaden and insurance companies become more comfortable using AI for business processes, either through their own efforts or by outsourcing, automation will impact industry employment levels and a much wider set of occupations.”
Best also highlights the potential for Generative AI to impact several applications. This includes writing code, creating marketing content, analyzing legal documents, and providing customer service.
But, given the fact that this technology is still within its early stages, having human expertise available in the near term is highly crucial.
Looking ahead, it is important to understand the different ways that AI technology can help insurers.
A key example is how AI can help insurers to gain a deeper insight into their customers through data-driven sets that are impossible through traditional underwriting.
By simply analyzing large volumes of data such as customer demographics and preferences, the technology can help organizations identify trends in risk profiles, as well as develop tailored solutions for each customer.
In addition, Best noted that AI may also provide real-time risk assessments and facilitate decision-making. As a result, this will allow companies to respond quickly to market changes and also offer pricing that corresponds more with the underlying risk.
By using predictive analytics to better gauge risk and provide real-time data for quotes on demand, insurers are able to customize policies based on each client’s needs.
Best also warns that the impact of the current disruption by AI is uncertain, which mostly falls down to its objective to replicate human intelligence.
Both data privacy and security concerns are highly important for the insurance sector given the sensitivity of consumer data. The use of more advanced forms of AI such as generative AI does raise many additional concerns about the moral use and protection of data, which may serve as a barrier to adoption.
