In many ways, the insurance industry helped hone the algorithms behind artificial intelligence technology.
From actuarial science, Bayesian statistics and generalized linear models, to cloud computing, telematics and advanced machine learning models, risk management models have always been designed to push the possibilities of prediction.
“AI is the decision support for that adjuster that is dealing with some very complex claims,” Heather Wilson, CEO of CLARA Analytics, told PYMNTS during a conversation for the “AI Effect” series. “… The models never sleep, and they are there to help the adjustor by surfacing information that the human in the loop may not see with the human eye.”
Wilson drew an analogy between AI in claims management and GPS navigation systems. Just as GPS systems provide real-time information and alternative routes to drivers, AI in claims management helps adjusters navigate complex cases. The AI platform analyzes data, detects signals and surfaces important information that may be overlooked by human adjusters.
This proactive approach saves time, improves efficiency and ensures better outcomes for both the claimant and the insurer.
Importantly, AI technology is applicable across all lines of insurance, including personal and commercial. By analyzing historical data and learning from past cases, AI models can help avoid loss costs, lower claim expenses and elevate best practices.
Read also: Insurance Claims Optimization Company CLARA Analytics Raises $24 Million
Employees working from home who get hurt on the job, or even professionals who may be infected with long-lasting symptoms of diseases like COVID-19, can introduce unfamiliar territory to the claims process.
The new, generative capabilities of AI mean that insurance companies don’t have to retrain existing teams to respond to novel and unique claim scenarios arising out of today’s hybrid employment landscape. And AI tools, for their part, are becoming much better equipped at detecting anomalies that could signal a potential concern.
“Insurance has always had an amazing statistical approach, particularly for reinsurance and for pricing, but bringing those capabilities over to the claims side is new,” Wilson explained. “… And because lost costs, which are medical costs, litigation costs — those costs are intensifying and escalating — it becomes about how does AI help the insurer? How do you help that adjuster to look at cases where the lost cost is really coming to play, and it’s surfaced for them to make a decision?”
AI algorithms can quickly analyze and categorize claims based on complexity, enabling faster processing of straightforward claims while directing more complex cases to human adjusters. AI systems can also automate routine and repetitive tasks, such as data entry and document processing, allowing adjusters to focus on more complex aspects of claims.
AI-powered computer vision systems can analyze images and videos to help in estimating costs more accurately, as well as be trained to recognize and classify damages.
All the same, Wilson acknowledged the concerns surrounding the potential for bias and even poor performance in AI systems while underscoring that it’s crucial to strike a balance between automation and human expertise to ensure optimal results.
However, she emphasized that because CLARA Analytics operates on the claims side of the insurance landscape, it can take a calculated and conservative approach with its AI models to ensure fairness and accuracy. By not using personally identifiable information and providing objective information, CLARA Analytics mitigates the risk of bias.
“We surface the potential tree branches of how this might go, not recommendations, and it is up to the human in the loop to decide what is best for that respective insurer and for that claimant … just like a GPS gives you routes,” she said, emphasizing that accuracy is being continuously monitored and benchmarked.
Wilson also explained that the insurance industry is highly regulated, and any AI solutions must operate within the regulatory guardrails to maintain compliance.
Looking ahead, Wilson said she sees agentic AI applications as the next great innovation in insurance claims management. These agentic AI systems would provide decision support and handle routine tasks, allowing adjusters to focus on more complex aspects of their work.
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.